UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A psychological classification system based on child molesters' motivations Paris, Faye Ann 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1999-389537.pdf [ 6.99MB ]
JSON: 831-1.0089243.json
JSON-LD: 831-1.0089243-ld.json
RDF/XML (Pretty): 831-1.0089243-rdf.xml
RDF/JSON: 831-1.0089243-rdf.json
Turtle: 831-1.0089243-turtle.txt
N-Triples: 831-1.0089243-rdf-ntriples.txt
Original Record: 831-1.0089243-source.json
Full Text

Full Text

A PSYCHOLOGICAL CLASSIFICATION S Y S T E M B A S E D ON CHILD MOLESTERS' MOTIVATIONS By F A Y E A N N PARIS B.A. , Simon Fraser University, 1986 M.A. , University of British Columbia, 1992 A PROPOSAL SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A April 1999 © Faye Ann Paris, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of British' G l^umbia Vancouver, Canada Department of DE-6 (2/88) Abstract Despite social and legal consequences, child molesters engage in sexual activities with children. Speculation as to why molesters choose children as their sexual partners was examined from seven perspectives: psychodynamic, attachment, three mental functions, psychopathic, physiological, learning, and third force theories. The potential causal variables suggested by these theories were coded using information from the files of 200 male adolescent and adult child molesters. Variables were analyzed using Ward's minimum variance clustering technique to develop a psychological classification system based on child molesters' motivations. Five-cluster solutions were discovered when using either the adolescent or adult child molesters' samples. Clusters from the adolescent sample were labeled: Stable (47%), Pedophilic (13%), Disturbed Conduct (15%), Psychopathic Tendencies (10%), and Dysfunctional Families (15%). The clusters from the adult sample were remarkably similar, yet there were differences. These clusters were labeled Mentally Unstable (25%), Mentally Stable (40%), Brain Damage/Head Injuries (14%), Psychopathic Tendencies (10%), and Dysfunctional Families (11%). Using a split-half test, four of the five clusters for both samples appear to be stable. Only the clusters based on psychiatric diagnoses of pedophilia were unstable when these diagnoses were eliminated from the analysis. Although validation of the five-cluster solution using contrasting variables for the adolescent sample was weak, the five-cluster solution for the adult sample was convincingly validated. iii Table of Contents Abstract ii Table of Contents iii List of Tables vi List of Figures vii Preface viii Acknowledgements. xi Chapter I: Child Molester Classification Systems 1 Chapter II. Etiological Theories of Child Molestation 10 2.1 Psychoaynamic Theories 11 2.1.1 Classical psychoanalysis 11 2.1.2 A recent psychoanalytic perspective 15 2.1.3 Jungian theories 16 2.2 Attachment Theory 19 2.2.1 Continuity of attachment styles 21 2.2.2 Attachment theory and sex offenders 22 2.3 Three Mental Functions: Cognitive, Affective, and Volitional systems 26 2.4 Psychopathy 31 2.4.1 Incidences and degree of psychopathic tendencies 33 2.4.2 Risk assessment and psychopathy 35 2.4.3 Etiological theories of psychopathy 37 2.5 Physiological Theories 39 2.5.1 Mild brain dysfunction 40 iv 2.5.2 Impulsivity 42 2.5.3 Fetal alcohol syndrome and fetal alcohol effects 43 2.6 Learning Theories 45 2.6.1 Pornography/erotica 46 2.6.2 Child sexual abuse 50 2.7 The "ThirdForce" Theories 54 Chapter III: A Multi-Theoretical Approach 61 Chapter IV: Information Relating to Child Molesters' Offense(s) 68 4.1 The Victim(s) 68 4.1.1 Age, gender, and relationship to offender 68 4.2 Current Sexual Offenses(s) 69 4.2.1 Prior to the offense 69 4.2.2 Substance abuse 70 4.2.3 Seduction techniques 70 4.2.4 Type of sexual activity 71 4.2.5 Silencing techniques 72 Chapter V: Method 74 5.7 Subjects 74 5.1.1 Adolescent Sex Offender Files 75 5.1.2 Adult Sex Offender Files 75 5.2 Measurements 76 5.2.1 Overview of Data Collection 76 5.2.2 Content of Adolescent Sex Offender Files 77 V 5.2.3 Content of Adult Sex Offender Files 79 5.3 Procedure 81 Chapter VI: Results 85 6. 1 Child Molester Characteristics at Time of Offense 85 6.1.1 Background Information 86 6.2 Cluster Analysis 91 6.2.1 Adolescent Five-Cluster Solution 100 6.2.2 Adult Five-Cluster Solution 106 6.3 Contrasting Variables 112 Chapter VII: Discussion 117 7.1 A Summary of the Findings. 117 7.2 Validation of the Five-Cluster Solutions 121 7.3 Implications of the Five-Cluster Solutions. 124 7.3.1 Clinical Implications 124 7.3.2 Investigative Implications 127 7.4 Limitations and Future Directions of Research 131 REFERENCES 134 APPENDDC A1-C2 151 APPENDIX D 177 APPENDIX E 178 vi List of Tables Table la: Preferential Child Molester 3 Table lb: Situational Child Molester 3 Table 2: Explanations of Pedophilia 63 Table 3: List of Potential Causal Variables by Theory 66 Table 4: Descriptive Analysis for Adolescent and Adult Child Molesters 87 Table 5: Descriptive Analysis of Adult Child Molesters 89 Table 6: Reliabilities for the Descriptive Analysis Variables 90 Table 7a: Potential Causal Variables for the Five-Cluster Solution-Adolescent and Adult Molesters 96 Table 7b: Potential Causal Variables for the Five-Cluster Solution: Adolescent Molesters 97 Table 7c: Potential Causal Variables for the Five-Cluster Solution: Adult Molesters 98 Table 8: Reliabilities for Potential Causal Variables for the Adolescent and Adult Molesters' Data 99 Table 9: Reliabilities for the Contrasting Variables 115 Table 10: Contrasting the 5-Cluster Solutions for the Adolescent and Adult Child Molesters 116 vii List of Figures Figure la: MTC:CM3 1 5 Figure lb: MTC:CM3 II 6 Figure 2: Tree of Adolescent Child Molester Cases 94 Figure 3: Tree of Adult Child Molester Cases 95 Figure 4a: Cluster 1 for Adolescent Data: Stable 103 Figure 4b: Cluster 2 for Adolescent Data: Pedophilic 103 Figure 4c: Cluster 3 for Adolescent Data: Disturbed Conduct 104 Figure 4d: Cluster 4 for Adolescent Data: Psychopathic Tendencies 104 Figure 4e: Cluster 5 for Adolescent Data: Dysfunctional Families 105 Figure 5a: Cluster 1 for Adult Data: Mentally Unstable 109 Figure 5b: Cluster 2 for Adult Data: Mentally Stable 109 Figure 5c: Cluster 3 for Adult Data: Brain Damage/Head Injuries 110 Figure 5d: Cluster 4 for Adult Data: Psychopathic Tendencies 110 Figure 5e: Cluster 5 for Adult Data: Dysfunctional Families. I l l Preface Classification systems are a way of organizing the universe by reducing a vast array of information into definable categories or groups (Skinner, 1977). The groups are arranged according to a set of principles or predetermined rules (Gottfredson, 1987). Each group, as well as each entity in the group, is identified with a diagnostic term or label. In this case, the entities being labeled are people in mental health and criminological settings. Concern for the person being classified evoked criticism for utilizing classification systems in these settings (Blashfield & Draguns, 1976; Gottfredson, 1987). It has been argued that a label can never completely describe all the important attributes crucial to understanding a unique individual. Nor can a label accurately predict an individual's actions, since being labeled can even aggravate symptoms. Predictions based only on a label, rather than the individual's actions, run the risk of being inaccurate. Therefore, a classification system needs to demonstrate benefits that compensate for the possible harmful effects just described (Blashfield & Draguns, 1976). The classification of individuals into groups has the potential to serve five practical purposes or gains (Blashfield & Draguns, 1976). First, referring to a similar group of individuals with a diagnostic term, rather than describing each individual, increases effective and efficient communication between professionals. Second, classification systems assist professionals in retrieving information. For example, in botany the statement is often made that 'A plant's name is a key to its literature' (Sneath & Sokol, 1973, p. 67). Third, classification systems are a procedure for distinguishing ix between meaningful similarities and differences between individuals. Precisely, how this is accomplished varies according to the classification system. Fourth, through the use of classification systems predictions regarding etiology, prognosis, and treatment can be formulated. These predictions are frequently derived from past and present symptoms or behaviors. Finally, a classification system can serve as the basis for a theory or a theory may serve as the basis of a classification system. These advantages, as well as the disadvantages previously discussed, are the potential outcomes of utilizing classification systems. Developing a psychological classification system for child molesters is an exploratory exercise. The initial step involves determining whether child molesters comprise a homogeneous or heterogeneous group of offenders. Earlier studies have described child molesters and their offences as varying on numerous dimensions (Gebhard, Gagnon, Pomeroy, & Christenson, 1965; Bradford, Bloomberg, & Bourget, 1988). Nevertheless child molesters are often regarded as a homogeneous group distinguished only by setting (i.e., prison, community, or psychiatric hospitals) or other forms of diversity (i.e., occupations or ethnic groups) (Marshall, Fernanadez, Hudson, & Ward, 1998). Unfortunately, treating a psychologically heterogeneous group like a psychologically homogeneous group interferes with research focusing on etiology, investigations, and the evaluation of treatment programs For example, "relations among variables that might emerge if cohesive subgroups were examined could be masked or cancelled out in analyses of larger samples with great intra-group differences" (Knight, Rosenberg, & Schneider, 1985). This statistical error would impede progress towards a better understanding of the psychological nature of child molesters. Determining whether or not child molesters are a heterogeneous group is only the initial step in this exploratory process; therefore, the investigative and clinical implications of these findings should be applied cautiously until research progresses in this area. Acknowledgements First and foremost I would like to thank my supervisor, John Yuille, for his immense contribution to my graduate training that extended beyond the academic arena. I would also like to thank my committee members, Robert Hare and Charlotte Johnston, for their advice and encouragement throughout this project. Michael Chandler, Peter Graf, and Margaret Wright's suggestions significantly improved this dissertation. Without the support of fellow students at the university this project would never have been completed. Thank-you to T. Diane Sirkia for her perseverance in coding the adolescent and adult files, and to Mitchell Minh Cho and Silver Anderson who entered an enormous amount of data on their own time. Conducting research in a community agency setting is, no doubt, an inconvenience for the staff. I would like to thank Roy O'Shaughnessy and Dennis O'Toole for overcoming these difficulties at Youth Court Services, and Teresa Belluz at the Vancouver Clinic. The medical record staffs' assistance at these institutions was appreciated. Special thanks are expressed to Derek Eaves, Diane Lamb, Derek Wilson, George Tien, and other staff at Forensic Psychiatric Services. I can not express enough gratitude to my extended and immediate family for their unfailing patience during my absences. I would like to thank my parents, aunt, and siblings for their unconditional love throughout the years. Not only do I appreciate Larry's personal and financial sacrifices, but the walks along the beach when I needed to regain a sense of balance. Sheila and Joanne were an inspiration through the hard times, and a pleasure to be with during the good times. Brett and Lee thank you for listening. This thesis is dedicated to Dailyn, Steven, Shayla, and all the children in our community. 1 Chapter 1 CHILD MOLESTER CLASSIFICATION SYSTEMS A classification system focusing on types of child molesters dates back to 1886. Kraffl-Ebing described persons who violated children under fourteen years of age, as "psychopathological," "non-psychopathological," and "paedophilia erotica." Psychopathological conditions were attributed to an "acquired mental weakness" (Kraffl-Ebing, 1886/1924, p. 554) from "dementia senilis" (Kirn, "Allg. Zeitschr. F. Psyciatrie," 39, p. 217), "chronic alcoholism, paralysis, mental debility due to epilepsy, injuries to the head and apoplexy, lues cerebri" (pp. 554-555). Non-psychopathological males were believed to be motivated by novelty or a fear of adult women, while non-psychopathological females simply took advantage of the fact that children were left in their care. The categorical label "paedophilia erotica" referred to patients that suffered from a "morbid disposition" (p. 555), not psychological or physical conditions. Krafft-Ebing was the first person to use the label paedophile, or simply pedophile, to refer to a patient with a sexual orientation towards children (Hucker, Langevin, Wortzman, Bain, Handy, & Chambers, 1986). Cohen, Seghorn, and Calmas (1969) classified pedophiles from a treatment center into three categories: fixated, regressed, and aggressive types. The fixated-pedophile was described as incapable of forming mature relationships with adults. Instead he focused on children for sexual and non-sexual gratification. Unlike the fixated-pedophile, the 2 regressed-pedophile had normal adult relationships. At the time of the offense, however, he felt inadequate and lacked the ability to cope with stresses in his life. Alcoholic episodes were not uncommon. The pedophile-aggressive types' sole aim was sexual sadism directed at young children. Sexual arousal increases with aggressive acts towards the child, but orgasm is seldom reached. The first two categories of pedophiles that Cohen et al. described were elaborated upon by Groth, Hobson, and Gary (1982). According to Groth et al., the fixated child molester is aware of his sexual orientation (usually towards male children) during adolescence. The offenses are pre-planned and committed compulsively with numerous victims. The offender comes down to his victims' level and identifies with them, thus the offender is frequently described as immature. In sharp contrast, the regressed child molester's sexual orientation is towards female adults, as evidenced by his involvement in marriage(s) or common law relationships. These intimate relationships frequently lack meaning and support, therefore he turns to a child when under stress. The sexual acts are often impulsive, episodic, and related to alcohol consumption. According to Groth et al. (1982), both types of molesters, fixated and regressed, are trying to meet their psychological needs for recognition, affiliation, acceptance, validation, even mastery and control. Groth (1979) also acknowledged that some child molesters' assaults are expressions of anger, power, or sadism. Lanning (1992) developed a typology for child molesters based on two of Dietz's categories described in the Encyclopedia of Crime and Justice: preferential and situational molesters. The preferential child molesters, like the fixated offenders, are sexually oriented towards children (See Table la). There are three sub-categories of o ET. cu a 3 o n O 3 IO 3 o n> cn a -\ vi a 3* P < 5L < Vi or 3 B" 0 0 o c 3 o' p > p 5 ' 3 n a 3 r-»--n CD 3» 5 ' OQ x 3 / 0! o. 03 3 CD 0 3* 1 3 = -a 3" rO re ft -t ST tt a o « 13 O cn O O 03 a. o 3 03 VI O O 5". So' Cu a a S < CD OQ 3* CD < 03 5 ' a. 3 c CD r P C CD O o CD r c P 3 "5' c_ oT 5' 3 3 OQ 03 o < 3-x cT C/3 N a> 03 cu < P 3 03 OQ CD Rp o 1 3 c 3 3 O rs "1 . o o o o 12 5 ' OQ c/i 7T 5d fro -i ro Vi ft a C Vi n t-i 6 •-+> TJ fD O CD* t/5 m x c £L en x -a a> 3. 3 a 3 r-t o 3 on o o P . ' c/> S M a e 3/ C a o 3 P . o 3* CD o cr CD 3 cr 5* r P 3 OQ 0 0 o c P £?. o p < > P 5 ' 3 £L O CD 3 3> 3 OQ RP w x 12. ST 3.' a> Cu O s* s P 3 Cu fO 3 Cu CD -1 P ft Kg P 3 Cu H5g CD 3 Cu fD T3 -n CD CD' •n m 3 O CD C/3 n o 3 3 o s ft sr a » . rs rs O C/3 - CD ro ' a o £L rs r* If* cn T 3 1 . CD ^ 8 3 " OQ »n P O •a cr cr 3J << Cu _ -1 O CD -1 3 CD >-l o O P H I: CO* p re CD o cr CD Vi <-*• CD 4 preferential offenders: seductive, introverted, and sadistic. Seductive child molesters court their victims offering them affection, presents, outings, etc. Introverted child molesters, lacking interpersonal skills, resort to sexual activities involving exposure, obscene telephone calls, and child prostitution. Sadistic child molesters simply lure or force their victims into sexual activities. Painful and degrading acts forced upon their victim increases their sexual response. While the preferential child molesters typically have the largest number of victims, the situational child molesters comprise the largest group of offenders (See Table lb). There are four sub-categories of situational child molesters whose primary sexual orientation is adults, rather than children: regressed, morally indiscriminate, sexually indiscriminate, and inadequate. Regressed offenders have turned to children as a substitute for an adult partner. They lack self-esteem, along with the necessary social skills to approach an adult. Morally indiscriminate offenders use children in the same manner as they use other adults, that is, to satisfy their own desires. It is simply an attitude that pervades their lifestyle. The sexually indiscriminate offender has this same attitude, but only in relationship to sexual matters. The inadequate offender is suffering from a psychiatric disorder, and feels like a social misfit. He chooses children, elderly, or handicapped victims, because they are vulnerable. Lanning (1992) observed that dangerous offenders are usually sub-categorized as sadistical, morally indiscriminate, or inadequate. A limitation of this system is that offenders are classified according to how, rather than why they committed their crimes (Lanning, 1992). Another classification system with a strong focus on the concept of fixation is referred to as the Massachusetts Treatment Center: Child Molester Typology, version 3 or MTC:CM3 (Knight, Carter, & Prentky, 1989; Knight & Prentky, 1990). However, in > 2 o a-i-i ~ § 3 3 5« > > l _ 3. 3 c/J o CD PS 3 O cr sr PJ =r a. < > & a. a> o I S ? I « r-t-§ § 3 § B- 3 o o' 1 C/J. >r c« 3 . 3 |C9 O 5 3 ^ 3 o_ o" |t3 o B: 3 o to w 72 cr a I t/3 adi No cn *-»• 3 i o° 7 contrast to the previously discussed systems that merely described the various types of child molesters, the MTC:CM3 is an operationalized system with demonstrated reliability and validity (Knight, Carter, & Prentky, 1989). The system assesses the child molester's degree of fixation (e.g., thoughts and fantasies of children) on Axis I (high versus low fixation) (See Figure la). On Axis II the amount of contact the molester had with children is assessed (high versus low amount of contact) (See Figure lb). Each category of Axis I is sub-divided into two more categories based on the offender's level of social competency (high versus low social competency). For Axis II, child molesters assessed as high on amount of contact with children are sub-divided according to the meaning of their contact with the children (interpersonal versus narcissistic). The child molesters assessed as having a low amount of contact with children are divided into groups reflecting the degree of physical injury their victims experienced (high versus low). Then, child molesters who were assessed as causing either high or low physical injuries were classified as non-sadistical or sadistical. According to the authors (Knight & Prentky, 1990, p. 34), "Crossing the four types of Axis I with the six types of Axis II yields 24 possible two axes combinations, which seems an excessive number of types." Another limitation of this system is the exclusion of nuisance (e.g., exhibitionism), indiscriminate (i.e., sex offenders whose victims were adults, as well as children), and incest (i.e., biological or nuclear family) offenses (Knight, Carter, & Prentky, 1989). In response to the lack of attention to incest offenders when constructing classification systems for child molesters, Finkelhor and Williams (1994) recently developed a typology from information obtained through interviews with offenders. Five types of incestuous offenders were labeled "sexually preoccupied," "adolescent 8 regressives," "instrumental sexual gratifiers," "emotionally dependent," and "angry retaliator." Using a cluster analysis technique, two of the typologies became indistinguishable: the emotionally dependent and instrumental sexual gratifiers. In addition, data on 33% of the fathers did not fit into any of the five typologies. Thus, the evidence only partially supported their classification system. The authors suggested abandoning this typological approach "to simply articulate a set of theoretical dimensions or attributes on which incestuous fathers may differ, and to study the causes and consequences of such differences" (p. 25). The purpose of the present project is to develop a psychological classification scheme for adolescent and adult male child molesters taking into consideration Finkelhor and Williams (1994) suggestion that theoretical dimensions or attributes may distinguish between sub-groups of child molesters.1 Initially, a literature review was undertaken to examine why molesters sexually abuse children. The explanations were separated into seven theoretical frameworks (i.e., using a bottom-up approach), although some researchers acknowledged using more than one theoretical framework (e.g., Marshall, 1989; Marshall & Christie, 1981). The term "theory" is being applied to these frameworks in a pragmatic, rather than a formal or informal sense of the term. Each theory refers to "a general principle or a collection of interrelated general principles that is put forward as an explanation of a set of known facts and empirical findings" (Reber, 1987, p. 789). Four of these theories (i.e., psychodynamic, physiological, learning, and third force theories) have been used for conceptualizing modern psychology (Robinson, 1979) and psychopathology (Coles, 1982), while the remaining three theories (i.e., attachment, three mental functions, 9 and psychopathy) contain principles directly applicable to child molesters. A brief summary and description of the implications of the general principles of each theoretical framework for data collection follows. Information pertaining to the characteristics of the offenses child molesters committed was also collected to understand how the psychological differences between child molesters were manifested. \ 1 An overwhelmingly large number of men, rather than women, are incarcerated for sexual assault (Correctional Services of Canada, 1991), therefore, this project focuses solely on male offenders. 10 Chapter 2 ETIOLOGICAL THEORIES OF CHILD MOLESTATION All classification systems require a theoretical framework to be used as a guide in the selection of variables (Knight, et al., 1985). The aim is to select variables that will discriminate between the clusters to increase the probability of developing a useful classification scheme. In this chapter seven theoretical perspectives focusing on why male adults engage in sexual activities with children were examined to choose potential causal variables to be entered into a cluster analysis technique. "Even statistical cluster techniques, which might seem at first glance to epitomize an inductivist approach to taxonomy construction, presuppose specific theoretical models and are unsuccessful when naive empiricism guides the selection of their input variables (Blashfield, 1980)" (cited in Knight, et al., 1985, p. 224). The following seven theoretical perspectives were reviewed for the sole purpose of selecting variables. This review was not intended to be a critique of the various theoretical perspectives, nor were the solutions obtained from the cluster analysis technique used to support a particular theoretical position. 11 2.1 Psychodynamic Theories 2.1.1 Classical Psychoanalysis On April 21, 1896, Sigmund Freud gave a lecture entitled The Aetiology of Hysteria (Masson, 1984a; Mason, 1984b). Freud "put forward the thesis that at the bottom of every case of hysteria there are one or more occurrences of premature sexual experience, occurrences which belong to the earliest years of childhood but which can be reproduced through the work of psycho-analysis in spite of the intervening decades" (Freud, 1896/1964, p. 203). This is accomplished through psycho-analytic techniques that associate hysterical symptoms with sexual experiences that occurred during puberty. Then, pubescent experiences are associated with earlier childhood sexual experiences, extending as far back as infancy. These intertwined sexual memories or sexual memory-traces lead to the roots of hysterical symptoms. Freud discussed 18 (6 male and 12 female) patients who had uncovered what they considered to be premature sexual experiences through psycho-analysis. The case histories of the patients were divided into three groups, according to the type of offender. The first group had been sexually stimulated by a stranger. The sexual act had been a single or isolated incident without physical injury. The second group had been sexually stimulated by their caregivers, such as a nurse maid, governess, tutor, or close relative. The abuser maintained a sexual relationship with the child over an extended period of time. The majority of cases fell into this group. The third group consisted of pairs of children who were of the opposite sex: one was the abuser, the other abused (e.g., brother 12 and sister). In some of the cases where the male had been the abuser, it was discovered that he had been sexually abused earlier by an adult female entrusted with his care. During the lecture, Freud did not discuss the possibility of the hysterical patient's father being the abuser. However, this possibility was raised by Freud in a letter to Fliess on December 6, 1896 (Masson, 1985). "It seems to me more and more that the essential point of hysteria is that it results from perversion on the part of the seducer, and more and more that heredity is seduction by the father" (p. 212).2 As strong as Freud's stance sounds, he changed his position. Initially, this is evidenced by the fact Freud failed to report the father as the seducer in two cases which were described in the Studies of Hysteria (Lerman, 1986). According to Masson, (1984b), Freud finally rejected his Seduction Theory3 because "the father, not excluding my own, had to be accused of being perverse" (p. 264) (perversions lead to seduction). There are other explanations for Freud's shift in positions (e.g., incest appeared to be an inadequate explanation for the high incidents of hysteria) (Lerman, 1986), and interpretations of Freud's work in this area remain controversial (Simon, 1992). Freud discussed the psychological and physiological sources of perversions in Three Essays on the Theory of Sexuality (Freud, 1915/1962). Perversions were attributed to the innate psychic energy of the libido (sexual energy) existing in everyone, even children. The perversions experienced by children are of a mild form, unless they are sexually stimulated prematurely. Sexual stimulation causes children to act upon these 2 On May 31st 1897 in a letter to Fliess, Freud recalled a dream revealing "overaffectionate feelings for Mathilde" his young daughter (Masson, 1985, p. 249). Freud stated, "The dream of course shows the fulfillment of my wish to catch a Pater as the originator of neurosis and thus [the dream] puts an end to my ever-recurring doubts" (p. 249). 13 perversions excessively. This is unavoidable as children lack the internal barriers (i.e., shame, disgust, and morality) to resist sexual activities. However, the strength of the perversion is dependent upon disposition and life experiences. In addition to innate psychic energy, Freud believed there are innate physiological remnants of the opposite sex in every male and female. "These [remnants] either persist without function as rudimentary organs or become modified and take on other functions" (Freud 1915/1962, p. 28). Bisexual remnants served as an explanation by Freud for why men were attracted to boys in Ancient Greece. The boys had physiological and psychological characteristics of males (e.g., male genitals, approach to playfulness) and females (e.g., lacks body hair, shy). Even though Freud viewed bisexuality as normal, pedophilia was considered a perversion. Freud reasoned (1915/1962), "People whose sexual objects belong to the normally inappropriate sex—that is inverts—strike the observer as a collection of individuals who may be quite sound in other respects. On the other hand, cases in which sexually immature persons (children) are chosen as sexual objects are instantly judged as sporatic aberations" (p. 36). The pedophile's approach to sex is viewed as an aberration, because the pedophile lingers over preparatory activities (e.g., touch and sight) or foreplay (Freud, 1915/1962). Even though prolonged touching is pleasurable, remaining at this phase without completing the sexual act is considered a perversion. The same logic applies to sustained visual impressions of the love object. Visualization is the most frequently utilized system for sexual excitation, especially when the body is concealed. Regardless of how pleasurable the act, it is considered a perversion when fixated at this point. It has been 3In this theory, according to Masson (1984a), "what Freud meant by sexual seduction was a real sexual 14 documented that pedophiles often engage exclusively in "mutual stroking, fondling, and masturbation," as well as "disrobing and inspection" (Sgroi, 1978, p. 131). According to Freud the perverse sexual activities engaged in by pedophiles are a consequence of being "cowardly," "an urgent instinct," or as a "substitution" for an adult sex partner (Freud, 1915/1962, p. 36). One form of perversion, incest, was believed to be a consequence of an unresolved Oedipus complex (Freud, 1915/1962). The Oedipus complex refers to a young boy's attraction to his mother through incestuous fantasies. The boy fears his father will discover his incestuous fantasies and punish him by removing his sexual organs (i.e., castration anxiety). His father is also viewed with hostility as he is a competitor for the mother's attention. The boy's resolution of the complex involves identifying with his father, consequently fantasizing about sexual relations with his mother is channeled through his father (Stagner, 1988). Unresolved, the Oedipus complex leads to neurosis, reliance on the parents' authority, and incest. According to Freud, incest is particularly damaging to the child, as well as, society (1918/1946). Severe restrictions are placed on the members of various societies to discourage acts of incest, as described in Freud's Totem and Taboo (1918/1946). For example, Australian tribes are broken-down into clans that are named after a totem (e.g., an animal, plant, or force of nature). "Almost everywhere totem prevails there also exists the law that the members of the same totem are not allowed to enter into sexual relations with each other" (p. 7). If the law is broken and sexual relationships do occur, traditionally the penalty is death. act forced on a young child who in no way desired or encouraged it" (p. 34). 15 2.1.2 A Recent Psychoanalytic Pespective Hammer's psychoanalytic understanding of sex offenders is a more recent representation of this perspective (Rada, 1978). The central theme of Hammer's article (1957) is that sex offenders are experiencing castration anxieties. To cope, pedophiles turn to a less threatening sex object. This idea was explored by Hammer (1968) through projective tests, interviews, and therapy by comparing 286 male sex offenders (i.e., incest offenders, heterosexual/homosexual pedophiles, exhibitionists and rapists) to incarcerated offenders (convicted of a non-sexual offense) and private patients. The psychological patterns observed in the sex offenders were consistent with Freud's essays on sexuality--with one exception. The "sex offenders were characterized by a pervasively concrete, as opposed to an abstract, mental orientation" (p. 10). This finding was used by the author to explain why over half of the pedophiles were in relatively unskilled (e.g., bus drivers), rather than professional (e.g. elementary school teacher) positions. In contrast, Freud (1915/1962) stated, "the sexual abuse of children is found with uncanny frequency among school teachers and child attendants, simply because they have the best opportunity for it" (p. 37). These professionals appear normal in other aspects of their lives, but falter in controlling their sexual instincts. On the other hand, an individual with a psychological abnormality will undoubtedly engage in an abnormal sex life. Regardless of this and other minor discrepancies between Freud's and subsequent psychoanalysts' perspectives, psychoanalytic therapy "was virtually the sole treatment method for child molestation" for years (Lanyon, 1986, p. 179). 16 2.1.3 Jungian Theories Jungian analysts in collaboration with Freudian psychoanalysts wrote a book on pedophilia titled The Forbidden Love: The Normal and Abnormal Love of Children (Kraemer, 1976). Like Freud, two of the Jungian analysts (i.e., Kraemer and Gordon) focused on pedophiles' early life experiences. Kraemer viewed pedophilia in terms of fusion. A state of fusion transpires when a mother idealizes the child as part of herself. In turn, the child idealizes the mother, along with his own self-image. When the child grows-up, "the unconscious identification with his mother's idealized picture often goes with his own tendency to enter into a state of fusion with yet another insufficiently separate object" (i.e., a child) (p. 3). Fusion with a child involves parallel participation in activities, rather than a relationship. This is evidenced sexually during mutual masturbation, when each participant drifts-off just as soon as it is their turn to be stimulated. The child closes his eyes while sucking his thumb; the adult is lost in his own world of images. "He is fascinated by fantasy images which are superimposed on children who may themselves have very little similarity to the idealizing person or idealized figure" (p. 3). Idealization of the child lends itself to the creation of elaborate poetry and other literary pursuits. Ultimately, these creations are a reflection of the pedophile's own idealized self-image. Gordon (Kraemer, 1976) focused on seductive experiences that occurred during the pedophile's childhood. One or both of the child's parents acted seductively either unconsciously or consciously. Unconsciously, even normal parents are attracted to their child's innocence and purity. This attraction is advantageous for the child, since it encourages the parent to be nurturing and protective. This "normal paedophilic drive" (p. 38) is beneficial to the parents, as it preserves their own "inner child" (i.e., the child within 17 themselves). The existence of an inner child is considered essential as it is the essence of art, music, literature, and religious experiences; without an inner child we merely exist. The paedophilic drive only becomes deviant when the parent is no longer aware of the child's needs. Instead the parent is driven by their own childhood anxieties, involving feelings of ineffectiveness, fearfulness, and dependency. The parent cannot relate the "inner child" in himself to the "inner child" in his sex partner. Instead, the parent can only relate consciously "to an actual child or youth" needing "the concrete reality of such a partner" (p. 45). In contrast to the two previous Jungian analysts who focused on childhood, Lambert viewed pedophilia as a projection of the self as reflected upon a child. Within the self is an archtype, or unconscious image, of a divine child combined with a heroic archtype. The negative aspect of the heroic archtype, defiance, is overcome by the positive aspects of the archtype of a child, represented by an unblemished future. To overcome the imbalance, the pedophile uses a child to mirror his own potential without ever knowing the real child. This theme of youth and divine innocence is prevalent across Jungian analysts. Williams stated, "It is as if the individual is renewed by contact with a young prototype of the self as yet untainted by 'the dirty devices of the world', as the poet Traherne has it" (Kraemer, 1976, p. 145). Summary. Freud offered various explanations for pedophiles engaging in perverse activities with children, including bisexual tendencies, feelings of cowardice, an urgent sexual instinct, as well as using children as a substitute for adult partners. In the case of incest, the offender may be inhibited developmentally, which can be traced to an unresolved 18 Oedipus complex. Using Freud's Oedipus theory, Hammer suggested that pedophiles experience castration anxieties, therefore, they turn to a less threatening sexual object—a child. Jungian analysts focused on the offender's attraction to innocence and youth. Kraemer was concerned with a mother's attempt to fuse with her child, while Gordon's concern was either parent being seductive with their child. Lambert believed the abusive parent projects aspects of themselves on the child in order for it to be reflected back to them. Parents who cannot relate to their own "inner child," attempt to relate concretely to a child's body. Hammer also believed that child molesters exhibit a concrete mental orientation. These explanations for pedophilia by psychodynamic theorists continue to be influential. Implications of the Psychodynamic Theories.4 The explanations suggested by psychodynamic theorists' for adults engaging in sexual activities with children were transformed into potential causal variables to be coded (See Appendix B l , adolescent sample; Appendix B2, adult sample). Freud emphasized the importance of understanding any premature sexual experiences in terms of the relationship between the abuser and the abused (Appendix B l and B2, Questions la, lb and lc). These premature sexual experiences may result in the abused excessively acting out perversions (Appendix B2, Question 2b), eventually even becoming the abuser. In addition to perversions, any functional physiological remnants of bisexuality may contribute to the development of pedophilia. Bisexual tendencies can be evaluated through the offenders' scores for "gender preferences" on the penile plethysmograph (this 19 test was administered by a psychiatric nurse to adolescent, but not adult offenders) (Appendix B l , Questions 18a-g). Finally, the issue of whether or not the offender is capable of abstract reasoning was evaluated using psychological assessments (Appendix B l , Question 16; Appendix B2, Question 15) and projective tasks (Appendix B l , Question 32). 2.2 Attachment Theory Ever since Freud made his famous, and in my view disastrous, volte-face in 1897, when he decided that the childhood seductions he had believed to be aetiologically important were nothing more than the products of his patients' imaginations, it has been extremely unfashionable to attribute psychopathology to real-life experiences (Bowlby, 1988, p. 78). Bowlby's statement describing the extent to which psychopathology has attributed to real-life experiences is debatable, although it is evident that attachment theorists focused on one form of real-life experiences—maternal deprivation. Very brief incidents of maternal deprivation demonstrate the type of affective tie (i.e., secure or insecure) between the child and mother. For example, one type of insecurely attached child when separated from his/her mother, responds to her as if she were a stranger when reunited. The child does not keep her in eye range while exploring, nor turn to her in need. The child is considered temporarily detached from the mother. After a few hours or even days later, the child ceases ignoring their mother. Instead s/he clings to her as an expression of fear and anger at the thought of being abandoned. "What this means is that a system controlling such crucial behaviour as attachment can in certain circumstances be rendered 4 The reader may want to refer back to the details contained in the implication sections after the 20 either temporarily or permanently incapable of being activated, and with it the whole range of feeling and desire that normally accompanies it is rendered incapable of being aroused" (Bowlby, 1988, p. 34). A possible consequence of a temporarily (i.e., insecure attachment) or permanently (i.e., psychopathy) deactivated system is psychopathology. To understand child molestation, one form of psychopathogy, the relationship between attachment and sexual systems needs to be considered (Bowlby, 1969). Bowlby believed there is a "link" between these two systems, but suggested three reasons for conceptualizing them separately based on ethological studies. First, these systems can change independently of each other. The attachment system is activated very early and intensely, while sexuality involves a system that appears slowly and incompletely. Second, attachment and sexualized patterns of behaviors are frequently directed towards different objects. Third, each system has its own period of sensitivity. Bowlby's reasons for conceptualizing the systems separately were not intended to negate the fact that interactions occur between systems, nor diminish the similarities between each system. The attachment and sexual systems do appear to share similar components (or behaviors). For example, an artificial cross-over between system components can be triggered: Young male birds of several species and also young male guinea-pigs, when treated with testosterone to accelerate sexual development, show sexual behaviour towards any object on which their attachment behavior is already imprinted. Control animals similarly injected but not yet imprinted show no sexual behaviour when presented with similar objects (Bowlby, 1969, p. 233). methodological procedures are explained. 21 Obviously, in humans accelerated sexual development cannot be related to imprinting. However, threats of abandonment appear to trigger the same mechanism for both systems, as evidenced by behaviors. For example, clinging behaviors of insecurely attached infants and children towards their caregivers appear to carry-over into their adult sexual relationships in the form of jealousy (Bowlby, 1969). 2.2.1 Continuity of Attachment Styles Empirical evidence does suggest continuity in forms of attachment across the life-span, according to a literature review undertaken by Rothbard and Shaver (1994). There appears to be a strong relationship between attachment measures as infants and various measures at 6 years of age. In addition, infants' attachment styles towards their opposite sex caregivers may ultimately affect their intimate relationships as adults, or their expectations about romantic heterosexual relationships (Collins & Read, 1990). These adult attachment styles have been categorized using a model developed by Bartholomew (Bartholomew, 1990; Bartholomew & Horowitz, 1991). The four categories of the model were derived from two dimensions: a model of self (dependence) and a model of others (avoidance). The first of the four categories describes secure adults with positive images of themselves and others. The second category describes adults who are preoccupied, that is, they are preoccupied with their relationships (Bartholomew & Horowitz, 1991). The fearful adults, described within category three, have a negative image of self and others. Finally, dismissing adults comprise the fourth category of adult attachment styles, as they try to ignore their needs for intimacy (Bartholomew & Horowitz, 1991). The author is unaware of any attempts to categorize types of sex 22 offenders according to Bartholomew's two-dimensional model, however, attachment theory has been used as a means of explaining the etiology of sexual offenses. 2.2.2 Attachment Theory and Sex Offenders Marshall stated that the critical deficiency of the sex offender is a "failure to develop the capacity for intimacy which is so important in establishing emotional and sexual relations with other adults (Brehm, 1985)" (Marshall, 1989, p. 491). He described three components of intimacy (citing Perlman & Fehr, 1987). First, intimacy involves feelings of closeness between two people. Second, people in an intimate relationship disclose information about themselves. Third, there are feelings of warmth, or genuine affection toward each other. The ability to experience intimacy can be charted through the use of a continuum ranging from feelings of alienation to experiences of connectiveness. When intimacy is present there is contentment and a sense that living is meaningful; in the absence of intimacy there are feelings of emptiness, loneliness, and alienation. Marshall (1989) speculated that the majority of sex offenders are consciously or unconsciously seeking intimacy. Sex offenders' personal histories frequently reveal that they were unable to form secure attachments to their parents as children. Both parents often acted erratically. If as children they were maltreated, there appears to be at an even greater risk of developing insecure attachments (Carlson, Cicchetti, Barnett, & Braunwald, 1989). Marshall suggested that if even one of the parents had been emotionally available, these devastating effects might have been neutralized. As well, other caregivers (e.g., teachers) could have neutralized the effects of poor parenting (Marshall, 1993). A disruption in the attachment process can also occur in adolescence 23 when feelings of being connected to their parental figures are transferred to peers. If adolescents are blocked from their association with peers, they do not develop a sense of independence. This leads to impairments in their adult intimate relationships. A lack of intimacy in adult relationships may result in loneliness (Marshall, 1993). Loneliness refers to "the discrepancy between desired and achieved levels of intimacy (Peplau & Perlman, 1982; Perlman & Peplau, 1984)" (Marshall, 1993, p. 114). Sex offenders who described themselves as "loners" often strove for a desired level of intimacy through superficial sexual encounters. Pleasure was obtained from the sexual release, but there was a sense of failure in not achieving intimacy (i.e., a deeply personal interaction with another individual). The offenders may have achieved some satisfaction, but a victim could not completely satisfy their emotional and sexual needs. Therefore, they continued their attempts to achieve intimacy, which led inevitably to frustration, even sexual aggression. "Aggressive individuals are often described as psychopaths (Hare, 1970), and such individuals, by definition, have extreme difficulties in establishing intimacy in their adult relations" (Marshall, 1989, p. 496). Marshall's theory of intimacy and loneliness in sex offenders is supported by Steidman, Marshall, Hudson, and Robertson's (1994) research. The first of two studies included seven groups of non-incarcerated males (i.e., exhibitionists, non-familial and familial child molesters, rapists, wife batterers and two control groups). The males from the experimental groups admitted, or were charged/convicted of an offence. They were recruited prior to being treated or incarcerated. The control groups were recruited through advertisements. All of the males were individually administered seven scales measuring intimacy, loneliness, hostility, trait anger, sexual functioning, history of family 24 violence, and social desirability. Results indicated that sex offenders scored lower on intimacy than the wife batterers and controls. This was particularly evident for non-familial child molesters and rapists. Sex offenders also indicated they were lonelier than the controls and wife batterers. Nevertheless, there was no difference in loneliness among various sex offender groups. Contrary to the authors' expectations, all of the sex offender groups indicated greater feelings of hostility towards women than the controls or wife batterers. The second study examined incarcerated, rather than non-incarcerated offenders. The incarcerated offenders included child molesters (without subgroups), rapists, violent non-sex offenders, and nonviolent non-sex offenders. Three scales were administered to measure intimacy, loneliness, and trait anger. Results indicated an overall group effect for the intimacy scale. The nonviolent non-sex offenders appeared to be more intimate than the remaining groups. On the loneliness scale the violent and nonviolent non-sex offenders appeared less lonely than the other two sex offenders groups, thus supporting the previous findings, and the primary focal point of Marshall's theory. Marshall and colleague's speculations on the etiology of sex offending also acknowledge other theoretical approaches. Marshall (1989) stated, "we have drawn attention to socio-cultural influences (Marshall, 1984a), biological underpinnings (Marshall, 1984b), developmental experiences (Marshall & Barbaree, 1984a), and conditioning processes (Laws & Marshall, 1988)" (p. 491). Likewise, earlier works by Marshall and Christie were influenced by psychoanalytic and socio-cultural theories. Psychoanalytic theory influenced Marshall and Christie's (1981) work on the etiology of child molestation. They suggested that males as children were attracted to their peers. As 25 males aged, they were continuously attracted to age appropriate partners. "The wonder of it is not so much that pedophiles fail to make this transition, but that by far the majority of males succeed [in making this transition]" (p. 156). Socio-cultural influences were considered by Marshall and Christie (1981) when examining the etiology of sexual aggression. Four influential factors were described. First, the personal histories of sexually aggressive child molesters often revealed parents who used punitive forms of discipline. Second, these child molesters were frequently exposed to aggressive, drunken, male models who would beat their mothers and siblings. Third, as adults they used alcohol during their offenses. Four, these child molesters may have associated high levels of arousal in aggressive acts with the high levels of arousal involved in sexual activities. These psychoanalytic and socio-cultural theories, as well as Marshall and colleagues attachment theory, took into account various approaches when considering the etiology of sexual activities and violence against children. Summary. Through ethological observations, Bowlby determined that attachment and sexuality are separate systems that interact. Each system can be temporarily, even permanently deactivated. Deactivated systems interfere with feelings and desires, thus contributing to psychopathology in adulthood. Applying attachment theory to sex offenders, Marshall suggested that offenders are actually pursuing intimacy through sexual encounters. In support of Marshall's theory, Steidman et al. (1994) found child molesters reported greater deficits in intimacy, than wife-beaters and controls. This was particularly true for non-familial child molesters. Aggressive or violent child molesters may have experienced punitive discipline, as well as 26 an aggressive, drunken, father figure. These offenders may have associated high levels of arousal in aggression with high levels of arousal in sexuality. Marshall and colleagues have continued to take an eclectic approach, while exploring components of the cognitive, affective, and volitional systems. Implications of the Attachment Theories. Reports in the offenders' files describing the degree of attachment the offender exhibited during childhood and adolescence were coded (i.e., unattached to various degrees of attachment) (Appendix Bl , Question 5; Appendix B2, Question 7), as well as forms of maltreatment that might have interfered with the attachment process (Appendix B l , Question 2, 3, and 4; Appendix B2, Questions 4, 5, and 6). For example, childhood experiences associated with aggressive and violent child molesters were measured (Appendix B l , Question 6; Appendix B2, Question 8). Attachment styles were evaluated in terms of self-reports (of intimacy, loneliness, and hostility) (Appendix B l , Question 7; Appendix B2, Question 9), an attachment model (i.e., evaluating images of self and others) (Appendix B l , Question 8; Appendix B2, Question 11), and number of intimate relationships (Appendix B2, Question 10). 2.3 Three Mental Functions: Cognitive, Affective, and Volitional Systems Cognitive, affective, and volitional systems were historically referred to as the three "mental functions" (Reber, 1987, p. 15). A cognitive system is intended to organize thoughts, conceptions, and reasoning, while an affective system is used to organize emotions, feelings and moods. Perhaps volition (or the pursuit of goals) is the most 27 difficult of the three systems to adequately explain as volition implies "generally and loosely, conscious, voluntary selection of particular action or choice from many potential actions or choices" (p. 824). All three types of mental functions may be discussed in relationship to other theoretical frameworks (e.g., attachment theory), however research on child molesters has focused directly or indirectly on these three functions. Horley and Quinsey (1994) examined child molesters' cognitions pertaining to themselves and others. Fifty-seven incarcerated child molesters (i.e., extra-familial offenders), 50 non-sex offenders, and 30 males from the community rated themselves on 21 bipolar adjectives (e.g., deceitful-truthful) using a 7 point scale separating each pair. Overall the three groups gave more positive than negative responses. Nevertheless, there were statistically significant differences in the extent to which the groups rated themselves positively. Compared to other groups, child molesters' described themselves as less attractive, less clean, and soft (versus hard). These responses may reflect accurate assessments, excuses, or low self-esteem. Child molesters' viewed women as more frigid in comparison to the other groups. Specifically, they viewed their spouses as less erotic, seductive, and sexy. It is conceivable that child molesters with a negative view of women are drawn to sexually unresponsive partners. Or, this line of reasoning may simply serve as an excuse to justify their involvement with children. Finally, child molesters who killed their victims perceived males as more seductive and deceitful. The authors' admitted that 5 they did not have an adequate explanation for this finding. Horley, Quinsey and Jones' (1995) replication of this study also found that child molesters, compared to non-child molesters (i.e., committed a non-sexual personal offense) viewed themselves and women in negative sexual terms (e.g., sexually 28 unattractive). However, they perceived boys as "kinder, more trusting, beautiful, calmer, cleaner, and more mature" (p. 13) than non-child molesters. This is consistent with Jungian analysts' observations that pedophiles are attracted to innocence and youth. The latter finding (i.e., regarding child molesters who kill their victims) was not addressed. In addition to these apparent distortions involving oneself and others, child molesters frequently engage in another type of cognitive distortion—denial. Laflen and Sturm (1994) illustrated four stages of denial used by sex offenders in a clinical setting. The first stage involves denying ever having committed a sexual offense. This may be an attempt to protect their own self-image, as well as maintaining an acceptable image of themselves to present to others. When they do accept the possibility that they committed the offense, they may deny any memory of their actions due to alcohol or drug abuse at the time of the offense. The second stage involves minimizing the sexual offense to avoid treatment. An offender may minimize their actions by stating that they really loved the child, or they were only providing the child with a good sex education. These errors are corrected when the offender realizes the potential harm involved in having sex with a child. The third stage of denial involves relinquishing responsibility for the offense. For instance, the offender may not admit responsibility because he did not plan the offense—it just happened. Overcoming the last stage of denial requires the child molester not only to accept responsibility for his actions, but experience "guilt" for committing the offense. At this point, the authors' recognize the importance of affect to conquer states of denial. The etiology and maintenance of sexually deviant activities involve both negative affect and cognitive errors, according to Hudson, Ward, and Marshall (1992). They referred to Weiner's concept of "attribution dependent affect" (p. 438), that is, a child 29 molester's reason for why the sexual activities occurred predicted their affective reaction thereafter. For example, if the child molester believes the activities can be controlled by his own internal efforts (i.e., "this kid is really sexy but I could stop looking at her"), this belief predicts feelings of guilt (p. 439). In contrast, if the child molester believes he should be in control but cannot maintain control (i.e., "I have no willpower. I am a disgusting person"), feelings of shame prevail (p. 439). When the offender views the sexual activities as external to his efforts, situational feelings of hopelessness or anger dominate. However, "child molesters more often express guilt or shame about their offenses which suggests that they attribute responsibility to themselves (i.e., internal and either controllable or uncontrollable)" (p. 440). Ward, Hudson, and Marshall (1994) extended this line of reasoning, theorizing that potential child molesters experience a negative affective state from interpersonal conflict and loneliness (see Marshall, 1989; 1993). To deal with these unpleasant feelings the molester fantasies about sexual activities with children. However, child molesters still need to rationalize their behaviors to themselves, if not to others. To accomplish this task the fantasies are accompanied by "a stream of cognitions" (p. 431). Sexual activities are planned and carried-out—the fantasy becomes sexual abuse in reality. During this process there appears to be a significant increase in negative emotions (i.e., disgust, hostility, shyness, and guilt) with a decrease in a positive emotion (i.e., interest). This is particularly evident between the time period of lapsing (i.e., engaging in a sexual fantasy) and relapsing (i.e., committing a sexual offense). Further exploration of the cognitive, affective, and volitional processes contributing to and maintaining sexual transgressions is provided in another article by 30 Ward, Hudson, and Marshall (1995). The authors' described Baumeister's cognitive deconstruction theory as a way of escaping awareness of oneself or one's circumstances by seeking pleasure through addictive or compulsive activities. For example, criminals have been observed to emphasize the details of their crimes to avoid the full impact of these experiences. Or they may dwell on the sensations, movements, and superficial aspects of a current experience. During this cognitive deconstruction process thought processes are reduced from their former level of abstraction. At this level "attention is not unduly influenced by higher-level concerns for the welfare of children or even the offender's own broader interests" (p. 75). Generally speaking, reasoning, emotionality, and morality deteriorate. The deterioration can be restricted to one area (e.g., the offense); it does not necessarily involve other areas. To reverse the process, the therapeutic aim is to maintain higher abstract levels in the unaffected area, while assisting "offenders in developing appropriate standards. Some offenders have low self-standards, while others have impossibly high standards" (p. 80). Interestingly, this theory is consistent with Hammer's finding that sex offenders (incarcerated and private patients) displayed concrete levels of reasoning, in contrast to Freud's viewpoint that offenders are capable of complex levels of reasoning, as evidenced by their position in society. Summary. To understand child molesters' cognitive processes, studies examined molesters' reasoning about self and others. It appears that child molesters frequently describe themselves, their wives, and other women in negative sexual terms, compared to non-child molesters. It is unknown whether the child molesters' assessments are ever accurate (e.g., 31 their spouses are less erotic) or simply excuses, or attempts to overcome a sense of low self-esteem. Theories have been put forth emphasizing the contributions of the cognitive and affective systems while considering volition in the initiation and maintenance of deviant sexual activities. For instance, experiencing negative affect (e.g., loneliness) leads to fantasies and rationalizations, prior to planning or executing the sexual activities. Implications of the Three Mental Functions Approach. To evaluate the extent that adolescent child molesters' engage in thought distortions in regard to sexual issues, their scores on the Cognitive Scale were recorded (Appendix B l , Question 10). Additionally, adolescent child molesters' descriptions of themselves and adult females and males in general were recorded (Appendix B l , Questions 11, 12, and 13). A specific type of cognitive distortion, denial, was evaluated (Appendix B l , Question 14; Appendix B2, Question 13) in relationship to the adolescents' and adults' current offenses. Both cognitive and affective reactions to their current offenses were considered (Appendix B l , Question 15; Appendix B2, Question 14). 2.4 Psychopathy Psychopathy is a socially devastating disorder defined by a constellation of affective, interpersonal, and behavioral characteristics, including egocentricity; impulsivity; irresponsibility; shallow emotions; lack of empathy, guilt, or remorse; pathological lying; manipulativeness; and the persistent violation of social norms and expectations (Hare, 1996). In 1941 when Cleckley (1976) was preparing a manuscript on psychopathy (i.e., The Mask of Sanity), he acknowledged that the term sexual psychopath was frequently 32 used to describe a wide range of sexual deviations (e.g., homosexuality, sadistic and masochistic inclinations, as well as exhibitionism, voyeurism, fetishism, and pedophilia). Applying the same term to various types of sexual deviations implies that they share a fundamental abnormality—psychopathy. "It is true that some sexually aberrant patients show additional pathologic features. Sometimes, but by no means regularly or even frequently, these features include those typical of the psychopath" (Cleckley, 1976, p. 287). Conversely, psychopaths' approach to and involvement in sexual activities frequently appears abnormal, in the sense of being impulsive, immature, detached, frivolous, and meaningless. They engage in prostitution, bigamy, or affairs with other men's wives as simply a means of enjoyment or a way to gain practical favors. Despite the overlapping features between psychopathy and various types of sexual deviations, a distinct line needs to be drawn, otherwise unique features of these sexual deviations are lost. To illustrate, consider the following sexual deviations (i.e., sadism, and masochism) in relationship to psychopathy. The psychopaths' lack of concern about anything or anybody gives them distinctive characteristics, in a very comprehensive sense, of both a sadist and masochist: "they bring humiliation and emotional suffering upon those who love them, as well as failure and unpleasant circumstances upon themselves" (Cleckley, 1976, pp. 290-291). A psychopath who is sadistical, using a more restrictive definition of the term, is extremely dangerous. "People of this type are often responsible for perverse and murderous attacks on children frequently noticed in the newspapers" (Cleckley, 1976, p. 291). For example, the Province newspaper reported that a man assessed by the court psychiatrist as sadistic and psychopathic (Hare, 1993) was questioned for approximately 140 sexual offenses, 33 including offenses against children (Horwood, 1987). It was the court psychiatrist who assessed him as both sadistic and psychopathic (Hare, 1993). Barbaree, Seto, Serin, Amos, and Preston (1994) studied sadistic rapists with psychopathic tendencies. On the Psychopathy Checklist-Revised (PCL-R), sadistic rapists scored higher on Factor 1 (emotional and interpersonal), Factor 2 (social deviance) and the total score, compared to non-sadistic rapists. However, the differences were only statistically significant for Factor 2. This study did not distinguish between possible types of sadistic rapists. One type of sadistic rapists may inflict pain as a means of sexual arousal, since they have difficulties experiencing affective arousal (i.e., psychopaths). Another type of sadistic rapist may be driven by a desire to degrade their victims to avenge their own emotional experiences (i.e., a truly sadistic rapist). It appears that psychopaths display characteristics that are common to other sexual deviations on the surface, however, this relationship breaks down when comparing their underlying drive or motivation. Various types of sexual deviations, like sadism, masochism, voyeurism, beastility, and pedophilia share a specific drive towards a sexual object. Apparently psychopaths do not have a drive of sufficient strength directed towards any object to be considered a sexual deviation. Instead, psychopaths display "poor judgment shown in what seems rather to be a vague blundering along the roads of folly and frustration toward what is essentially life rejection" (Cleckley, 1976, p. 290). 2.4.1 Incidences and Degree of Psychopathic Tendencies Among Sex Offenders The incidence and degree of psychopathic tendencies among non-sex offenders and sex offenders has been under-investigated, considering the seriousness of this condition. 34 In 1988 Prentky and Knight (unpublished manuscript cited in Hare, 1990) used the Psychopathy Checklist-Revised (PCL-R) to assess 59 child molesters and 95 rapists. Even with a rather high cut-off score of 32 out of 40 to indicate psychopathy, a large percentage of child molesters (30.5%) and rapists (45.3%) were identified as psychopaths. Similar results were obtained by Brown (1994) using the PCL-R with a cut-off score of 30 out of 40 to assess rapists. Thirty-five percent of the rapists scored above the cut-off point. In 1994, Froth and Kroner (unpublished manuscript cited in Hare, 1996) using the PCL-R on federal inmates found 26.1% of the rapists, 18.3% of various types of sex offenders (e.g., child molesters), and 5.4% of incest offenders had a psychopathy score of 30 or above. In contrast to the previous findings, Serin, Malcolm, Khanna, and Barbaree (1994) identified a rather small percentage of federally incarcerated child molesters (7.5%) and rapists (12.2%) as psychopaths. These findings were unexpected considering the authors use of the PCL-R cut-off point was low, that is 29 out of 40. Serin et al. (1994) attributed these discrepancies to procedural differences. They used file information to assess psychopathy, rather than a semi-structured interview accompanied by file information. Procedural differences for scoring subjects on the PCL-R was a crucial consideration, when deciding how to gather data for this project. Unfortunately, when evaluating previous research it was unclear whether procedural differences actually accounted for lower rates of psychopathy. For example, Wong (1988) assessed male inmates for psychopathy using two different procedures: file information with and without the interview. There was high agreement between these two procedures, although only using file information tended to underestimate the number of inmates rated high in 35 psychopathy. Nevertheless, Serin (1993) did not find the exclusive use of file information, compared to file information and interviews, produced a more conservative approach— just less reliable. Both authors (Serin 1993; Wong, 1988) were in agreement that assessing psychopathic tendencies only from files requires comprehensive information: "clinical and background information," as well as "clinical judgments and caseworker opinions" (Serin, 1993, p. 370). Serin et al. (1994) also found a statistically significant correlation between psychopathy and deviant sexual arousal using phallometric assessments. The stimuli focused on age, gender, and sexual violence. Responses to inappropriate and appropriate stimuli were compared. It appears that the positive relationship between psychopathy and deviant sexual arousal was particularly strong for extra-familial child molesters. However, for intra-familial child molesters the (negative) correlation was non-significant. Similarly, Seto, Barbaree, and Serin (1994) found that deviant sexual arousal to same sex prepubescent and pubescent children was characteristic of high-fixated child molesters. Higher Factor 2 and total scores on the PCL-R, in addition to more non-sexual offenses, was characteristic of low-fixated child molesters. "These results suggest there are at least two types of sexual offenders against children: One resembling the prototypical pedophile in terms of sexual interests and another type resembling a more generally antisocial offender" (Seto et al., 1994, p. 51). 2.4.2 Risk Assessment and Psychopathy Additional studies on this topic need to be pursed, since psychopathy (i.e., scores on the Psychopathy Checklist), deviant arousal (i.e., phallometric assessment), and sexual 36 offense history "are useful in planning treatment and supervision programs and in assessing risk" of re-offending (Becker & Quinsey, 1993, p. 173). Planning treatment and supervision programs for psychopathic offenders is particularly difficult. Treatment may actually aggravate the psychopaths problems as evidenced by an increase in recidivism rates among institutionalized offenders (Rice & Harris, 1992). For assessing risk of recidivism, measures of psychopathy and deviant arousal may surpass demographic and psychiatric variables (Rice, Harris & Quinsey, 1990). A psychopathy measure alone was predictive of recidivism in adolescent sex offenders. One hundred and ninety-three adolescent male sex offenders' files were scored at a forensic institution on the PCL-R (O'Shaughnessy, Hare, Gretton, & McBride, 1994). Twelve months after release, 55% of the psychopaths and 15% of the non-psychopaths had been convicted of non-sexual offenses. However, convictions for sexual offenses were low and independent of psychopathic tendencies. At present, the study has been extended to include 223 adolescent sex offenders whose mean score on the PCL-R was 22.00 (SD=7.08). The adolescents labeled "mixed type offenders" (i.e., committed sexual offenses against children and adults) and "high" on psychopathy committed offenses earlier than subtypes who committed offenses solely against children and adults and scored high on the PCL-R (R.D. Hare, March 18, 1997, personal communication). Other adolescents labeled "deviant psychopath" (i.e., according to their high scores on the penile plethysmograph and PCL-R) committed offenses sooner than "non-deviant psychopaths," "deviant non-psychopaths," and non-deviant non-psychopaths" (R.D. Hare, March 18, 1997, personal communication). In adult sex offenders, rates of risk are also dependent upon type of sex offender. In a 6 year follow-up study at a maximum security psychiatric 37 institution "inappropriate sexual age preferences as measured phallometrically were related to convictions for new sexual offenses" (Rice, Quinsey, & Harris, 1991, p. 385). 2.4.3 Etiological Theories of Psychopathy Numerous theories have been put forth to explain why molesters abuse children, including those with psychopathic tendencies. In regards to psychopathy, Hare (1993) speculated that an interaction exists between biological and social forces. Specifically, there is evidence to suggest that "genetic factors contribute to the biological bases of brain function and to basic personality structure, which in turn influence the way the individual responds to, and interacts with, life experiences and the social environment" (p. 173). The social environment (e.g., parenting) may affect how psychopathic tendencies develop and are expressed through actions: "psychopaths from unstable backgrounds committed many more violent offenses than those from stable backgrounds" (p. 175). Nevertheless, socialization has only a minute impact on the development of empathy or a conscience. In contrast to Hare's position, attachment theory (See section 2.2) has been adopted as an explanation for the development of psychopathic tendencies. Attachment theory assumes that psychopaths when infants failed to form a binding affection to their caregivers. Criticisms of this theory are: 1) it has been used too extensively to explain disorders ranging from schizophrenia to crime, 2) most of the evidence supporting this theory is based on retrospective reports, and 3) evidence is lacking to explain how a lack of attachment causes all the symptoms the psychopath exhibits (Hare, 1993). "While some assert that psychopathy is the result of attachment difficulties in infancy," Hare stated, "I turn the argument around: In some children the very failure to bond is a 38 symptom of psychopathy. It is likely that these children lack the capacity to bond readily, and that their lack of attachment is largely the result, not the cause, of psychopathy" (p. 172). Summary. Cleckley was critical of the broad use of the term sexual psychopath applied to various forms of sexual deviations. Unlike other sexual deviants, psychopaths lack a drive or goal. The incidence of child molesters with psychopathic tendencies ranges from 7.5% to 30.5%, compared to rapists with psychopathic tendencies ranging from 12.2% to 45.3%. The incidence of psychopaths among specific types of child molesters has not been reported. However, for extra-familial but not intra-familial child molesters, there was a positive correlation between deviant arousal and psychopathy. Greater deviant arousal for prepubescent and pubescent children was evidenced by high-fixated child molesters (i.e., the "prototypical pedophile"). However, an anti-social lifestyle characterized the low-fixated child molester (i.e., an "antisocial offender"). For both the high- and low-fixated child molesters, the effects of nature (i.e., physiology) versus nurture (i.e., learning) on these disturbances are still being explored. Implication of Psychopathic Theories. Considering Cleckley's observation that psychopaths lack a deviant sexual drive, the DSM diagnostic categories were recorded for each child molester, in addition to other sexual problems (e.g., fetishes, low or high sex drive, etc.) (Appendix B l , Question 21; Appendix B2, Question 19). It was anticipated that deviant sexual arousal separates pedophiles (as a sexual orientation) from psychopaths who engage in pedophilic activities (Appendix B l , Question 18). Psychopaths can also be distinguished from other sexual 39 deviants on measures of psychopathy (i.e., the PCL-R) and empathy (e.g., UCLA Empathy Scale), according to Hare (1993) (Appendix Bl , Question 19 and 30; Appendix B2, Question 17). 2.5 Physiological Theories There is evidence to support physiological explanations for pedophilic tendencies in child molesters (Freund, Heasman, Racansky, & Glancy, 1984; Freund & Watson, 1992). Freund et al. (1984) compared the prevalence of homosexual pedophilia to androphilia. Homosexual pedophilia refers to an attraction for immature male partners (i.e., boys), while androphilia is a physical attraction to mature males (i.e., men). The authors cited previous research estimating the prevalence of androphilia to be 4 or 5 percent. In contrast, previous research estimated the prevalence of homosexuality among the pedophilic population as 32-46%. This is consistent with Freund et. al.'s (1984) finding that 36% of457 pedophilic patients' offenses were against males. The authors suggested that: development of partner sex preference and age preference are not independent of each other and that sex preference in pedophilia is etiologically different from the sex preference of males who prefer physically mature partners of either sex. Moreover, taking into consideration that pedophilia does not exist, or is extremely rare, in women, one could further hypothesize that there is a physiological basis for pedophilia (p. 198). Speculations as to how physiology relates to pedophilia are continuously being put forward (Flor-Henry, Lang, Koles, & Frenzel, 1991; Gaffney, & Berlin, 1984; Gaffney, Lurie, & Berlin, 1984). Recent speculations include impulsivity and fetal alcohol 40 syndrome or effects. Impulsivity has been related to sexual offenses in general (Prentky & Knight, 1986), and pedophilia specifically (Bradford, & Gratzer, 1995). However, the long-term effects of fetal alcohol syndrome (FAS) and fetal alcohol effects (FAE) in relationship to pedophilia have been overlooked in the literature on mild brain dysfunctions (Hucker et al., 1986; Regestein & Reich, 1978; Scott, Cole, McKay, Golden, & Liggett, 1984). 2.5.1 Mild Brain Dysfunction Hucker et al., (1986) compared pedophiles (i.e., homosexual, bisexual, and heterosexual offenders) to controls (i.e., offenders who were not convicted of a violent or sexual offense) on psychological, neuropsychological, IQ, and medical tests. Results indicated that pedophiles scored significantly lower than controls on performance and full IQ, but the same on verbal IQ. On the Reitan Neuropsychological and the Luria-Nebraska Neuropsycholgical test batteries, pedophiles showed significantly more impairments than the controls. The Computerized Tomography Scan (CT-scans) also revealed significantly more abnormalities for the pedophiles than controls. These findings could not be adequately explained by age or substance use. According to the authors, "the indication of brain pathology may be relevant to the etiology of the disorder" (Hucker et al., 1986, p. 447). Similar results were obtained on neuropsychological tests from pedophilic patients who had committed a sexual assault on a prepubescent child (PED), compared to those patients who had forcibly sexually assaulted an adult (FSA) (Scott, Cole, McKay, Golden, & Liggett, 1984). There was a control group consisting of hospitalized and 41 nonhospitalized volunteers. All subjects were administered the Luria-Nebraska Neuropsychological Battery (LNNB). Results indicated that individuals accused of committing a sexual assault scored lower on all of the LNNB scales, excluding two scales (i.e., Left Hemisphere and Tactile). Within the sexual assault group those who committed an offense against children scored lower on all scales than those who committed an offense against adults. When focusing on the LNNB scores (controlling for age and education) both assault groups had a high percentage of brain dysfunction (PED, 36%; FSA, 55%). An additional observation by the authors was that the police reports for these offenses imply that many of the subjects had "episodic discontrol syndrome," possibly indicating that the limbic system is not adequately inhibiting impulses. Symptoms of episodic discontrol syndrome include egocentricity, attention seeking, and impulsiveness. The authors concluded, "for a large proportion of the rapists and pedophiles, cerebral dysfunction may be a contributing or dominant factor" (p. 1118). Criticisms of research using neuropsychological tests were put forth by Hart, Forth, and Hare (1990) when considering the possibility of brain dysfunctions in psychopaths. These same criticisms may be applicable to the neuropsychological testing of child molesters. The authors were concerned that the level of performance on neuropsychological tests could be accountable by age, education, or substance use. Hucker et al. (1986) did account for age and substance use, while Scott et al. (1984) controlled for age and education. Another criticism was the lack of control over Type I errors, since large test batteries are used on small sample sizes. The previous studies on child molesters did rely on small to moderate sample sizes with extensive test batteries. 42 Improving the methodology when using neuropsychological tests, in addition to studies using CT-scans, IQ tests, and other physiological measurements are warranted. 2.5.2 Impulsivity The research on impulsivity and sex offenders is sparse, but of interest when pursing physiological explanations for child molestation. Prentky and Knight (1986) developed three measures of impulsivity for child molesters and rapists. The first measure evaluates the degree of impulsivity evidenced by their offenses. Offenses were rated as "planned in detail," "partially planned before victim encountered," "partially planned after victim encountered," or "no planning" (p. 149). The second measure focuses on the offender's general lifestyle as evaluated by the Massachusetts Treatment Center (MTC) classification system. Finally, transiency was rated according to the offender's demonstrated ability to maintain stable employment and residency. Results indicated that child molesters who were impulsive when committing the offense were less apt to have serious sexual offenses as an adolescent, but more abusive of alcohol. However, the most interesting findings for child molesters was on the third measure, transiency. "Both measures of transiency were more important for the child molesters than rapists, and seemed to define a subgroup characterized by a schizoid lack of cathexis, itinerant individuals, withdrawn from the outer world, with a seclusion and eccentric style. Although the absolute amount of acting out was less, the quality of the behavior seemed to be more disturbed" (pp. 160-161). The authors' labeled this subgroup the "expressive fixated child molesters" (p. 161). 43 As mentioned previously fixated child molesters are sexually orientated towards children, or suffer from pedophilic tendencies. Both pedophilic and impulse control disorders have been associated with obsessive-compulsiveness (Bradford & Gratzer, 1995), as all three of these disorders are characterized by repetitious behaviours (Abel, Mittelman, & Becker, 1985; Weiss, Hechtman, Perlman, Hopkins, & Werner, 1979; Wilson, 1981). Consider the case study of a 38 year old male who committed incest, as well as displaying pedophilic tendencies, depression, and trichotillomania (i.e., pulling out of one's own hair) (Bradford & Gratzer, 1995). When treated for 2 months with sertraline, all three disorders diminished or were eliminated completely. Nevertheless, the authors' suggested that "Nonexclusive incest-type pedophilia is a more heterogeneous category than true pedophilia and likely has a multideterminant etiology" (p. 5). 2.5.3 Fetal Alcohol Syndrome and Fetal Alcohol Effects Impulsivity is also a characteristic of Fetal Alcohol Syndrome (FAS) and Fetal Alcohol Effects (FAE) (Short, & Hess, 1995), nevertheless, the primary concern is that FAS/FAE is a leading cause of mental retardation (Streissguth, 1994). FAS/FAE occurs when alcohol is ingested during pregnancy, and may be accompanied by other contributing factors (e.g., use of drugs and poor nutrition). Sociallly-disadvantaged groups are particularly at risk. For example, in British Columbia the non-native rates for FAS are .25 per 1,000, while the native rates are 4.7 per 1,000 (Wong, unpublished study cited in Short & Hess, 1995). Initially the diagnosis of FAS/FAE was restricted to children, however, in the mid-70s and early 80s adolescents and adults began to be diagnosed (Streissguth, 1994). 44 "Longer noses and bigger chins often gave their faces a coarser look after puberty" (p. 75). They exhibited growth deficiencies for either height or weight. The long-term effects of FAS/FAE on adolescents and adulthood included low IQ and mathematics scores, as well as problems understanding abstractions, cause/effect, and generalizations. Memory and attentional problems were also exhibited. Superficial verbal skills concealed these deficits only temporarily, as their lack of communication and socialization skills soon revealed their problems (Streissguth, 1994). Adolescents and adults with FAS/FAE "often have multidisciplinary problems that involve the home, the schools, the health care system, vocational training, the criminal justice system, and the community" (Streissguth, 1994, p. 79). A number of FAS/FAE adults are institutionalized in mental health or criminological settings (Lemoine & Lemonine, 1992 cited in Streissguth, 1994). The rates of FAS/FAE adults specifically involved in sexual crimes against children is unknown, although FAS/FAE cases have recently come to the awareness of professionals working in this area (J.C. Yuille, personal communications, September 1996). Adolescents and adults with FAS/FAE may be distinguishable from other types of child molesters. Summary. There is evidence to suggest a physiological basis for pedophilia, since sex and age preferences did not appear to be independent (Freund, Heasman, Racansky, & Glancy, 1984; Freund & Watson, 1992). The possibility of mild brain damage in child molesters should continue to be explored through the use of psychological, neuropsychological, and medical tests. Previous findings do suggest that pedophiles exhibit more brain pathology than would normally be expected. This may be exhibited through poor impulse control or 4 5 characteristics associated with FAS/FAE. Even if some sort of physiological damage has occurred, this may only affect a sub-group of child molesters. In some of these cases, the damage may be mended through sociological interventions. Implications of the Physiological Theories. The possibility that some child molesters have a mild brain dysfunction was evaluated using traditional approaches, including IQ measurements (Appendix B l , Question 20; Appendix B2, Question 18) and information from psychological assessments (Appendix B l , Question 17; Appendix B2, Question 16). Less traditional approaches were used to evaluate impulsivity and FAS/FAE. For impulsivity, information was evaluated in terms of the offender's lifestyle (Appendix B l , Questions 22 and 28; Appendix B2, Question 21). For indications of FAS/FAE two types of file information were considered: 1) the offender's biological parents' drinking habits (Appendix B l , Question 23 and 24; Appendix B2, Question 22 and 23), and 2) any relevant psychiatric diagnoses of the offender during childhood (Appendix B l , Question 25). The ability to engage in abstract reasoning is an important variable to be considered in the diagnosis of FAS/FAE. This variable was previously mentioned under the Psychodynamic Theories section. Other sexual problems related to physiology were also recorded (Appendix B2, Question 2a and 2c) 2.6 Learning Theories The application of learning theories to the etiology of sexual deviancies implies that sexual activities are a consequence of various forms of conditioning (e.g., classical, 46 operant, and avoidance conditioning), as well as modeling (i.e., observing and imitating others) (Robinson, 1979). McGuire, Carlisle, and Young (1965) illustrated how sexual deviations can be conditioned through a case study of a child molester. A male approximately 40 years of age described his sexual orientation as relatively normal until his 20s, in spite of the fact that he had experienced impotency during attempts at intercourse. He attributed his impotency to a strict religious upbringing that discouraged casual sexual encounters. It was in his 20s when pursuing a military career that he engaged in sexual activities with prostitutes by having them masturbate him. If prostitutes were unobtainable, he turned to female and male children to relieve his sexual urges. When masturbating alone, these experiences became a part of his fantasies. After leaving the military, he approached young males where he was residing for sexual favors. These experiences eventually contributed to the fantasies he used, while having sexual relations with his wife. Their son was born. When the boy was between 4 to 6 years of age the father encouraged his son to masturbate him. The authors' believe this case study demonstrates "how force of circumstances and self-masturbation to a sexual memory can shape, by simple conditioning, an individual's sexual behaviour" (p. 188). 2.6.1 Pornography/Erotica A form of conditioning commonly associated with sexual deviance is pornography. The word "pornography" is derived from Greek writings depicting the "life, manners, etc. of prostitutes and their patrons; hence the expression or suggestion of obscene or unchaste subjects in literature or art" (Oxford, 1984, p. 2242). What is obscene or unchaste is usually defined according to the prevailing societal standards. However, "in the final 47 analysis, obscenity is much like beauty and exists primarily in the eye of the beholder" (Reber, 1987, p. 486). A distinction between "pornography" and "erotica" has been made by feminists (Nemes, 1992). Erotica refers to the depiction of equality in sexual acts; in contrast, pornography involves degrading or demeaning acts. Both terms refer to sexually explicit materials. One of the first well-controlled studies of institutionalized offenders' use of sexually explicit materials was conducted by Goldstein, Kant, Judd, Rice, and Green in 1971 (cited in Marshall, 1988). Interviews were administered to pedophiles, rapists, homosexuals, transsexuals, users of pornography, and two control groups (Goldstein et al., 1971). Pedophiles were broken down into sub-groups based on their preference for male or female sexual partners. Both groups of pedophiles reported "a strikingly low degree of exposure to representations of mature sexual activity (heterosexual intercourse) during adolescence" for various forms of stimuli (e.g., photos, movies, and books) (p. 6). (This was also the case for the homosexuals, transsexuals, and pornographic users compared to the controls.) For homosexual pedophiles, however, the group differences for exposure to sexually explicit materials were smallest for homosexual activities depicted in photos and books. "This suggests that their homosexual interest was already present during their adolescent years and influenced their choice of erotica" (p. 6). Likewise, both groups of pedophiles reported a limited use of erotic materials in the previous year. (This was also the case for the rapists and transsexuals compared to the controls.) All of the groups of sex offenders and the pornographic users were more apt to use erotic materials for the purposes of arousal or masturbation. Rates of masturbation appear to diminish with maturity in normal males. "When asked if the subject did in fact follow through with 48 such sexual activities either immediately or shortly thereafter, all the reports dropped sharply" (p. 12). However, heterosexual pedophiles and one of the two control groups scored the highest for following through with sexual activities. Goldstein (1973) examined the results from the interviews used in the previous study focusing on the offender's and his parent's attitudes toward sex. Results indicated that homosexual pedophiles' described their families during childhood as very conservative in their attitudes towards sexuality and nudity. Interestingly, 50% reported that their first sexual encounter was with a male prior to age 14. "The generally low exposure to erotica reported by the male pedophiles suggest that their sexual development was more likely to have been influenced by actual sexual contacts as a child than by erotica" (p. 217). Heterosexual pedophiles also reported a conservative attitude towards sexuality within their homes as children; that is, issues around sex were rarely discussed. Of all the groups, heterosexual pedophiles' reported the highest percentage (31%) of first sexual encounters being with a prostitute. "Possibly the choice of immature girls as sex objects represents an attempt to find sex partners who are innocent and free of the connotation of sin" (p. 217). This interpretation is consistent with the Jungian theme that children represent an image of youth and divine innocence. From Goldstein's (1973) findings it appears that both pedophilic groups were unaffected by erotica, however, early sexual practices distinguished the two pedophilic groups. Another noteworthy study in relationship to pornography was undertaken by Marshall (1988). Four changes were undertaken to improve previous methodologies: 1) subjects were not interviewed until rapport was established, 2) only "hard-core," (i.e., attainable only through specialized stores or illegally) not "soft-core" (i.e., attainable at 49 local stores) pornography was considered, 3) exposure to these materials during pubescence was considered, and 4) subjects were asked whether they had used these materials prior to committing a crime. The subjects consisted of 89 offenders from a clinic (i.e., 51 child molesters, 23 rapists, and 15 incest cases) and 24 non-offenders. During pubescence, the child molesters, rapists, and non-offenders had come in contact with hard core pornography more frequently than the incest offenders. By adulthood the child molesters and rapists were more apt to use these materials than incest offenders and non-offenders. Remarkably, "slightly more than one third of the child molesters and rapists claim to have at least occasionally been incited to commit an offense by exposure to one or the other type of the sexual materials specified in this study" (p. 284). Occasionally, this occurred unintentionally; however, 53% of these child molesters and 33% of these rapists used the materials deliberately when planning an offense. These results are in conflict with Nutter and Kearns' (1993) findings that 84% of the child molesters "did not believe that pornography led them to engage in child molesting" (p. 82). The reason for this discrepancy may be due to the latter studies use of questionnaires, rather than in-depth interviews. A review of the literature exploring the relationship between sex offenses, pornography, and attitudes was completed by Nemes (1992). The literature relating non-violent pornography and changes in antisocial attitudes towards violence appeared ambiguous, but the literature on violent pornography and changes in antisocial attitudes was more substantial. In reference to violent pornography the author stated, "The available data do seem to suggest a causal relationship between exposure and callous sex attitudes, as well as an increased likelihood to rape in those men with a propensity to rape" 50 (p. 475). The data from laboratory studies were particularly convincing. Unfortunately, the author could not determine whether it was the sex or violence that contributed to the problem. 2.6.2 Child Sexual Abuse The expectation would be that a sexually abused child has learned a set of deviant behaviors by observing the offender. The set of behaviors might be imitated immediately after observing the event, or months, even years later. "Imitation of a model's behaviour is most likely to occur if that person is perceived by the child to be attractive, prestigious, competent, and/or powerful; if the model is the dispenser of rewards, either monetary or emotional; and if the model is of the same gender" (Orr, 1991, p. 95). This process is frequently referred to as modeling, "the fundamental learning process involved in socialization" (Reber, 1987, p. 447). Modeling sexually abusive behaviors may explain the etiology of child molestation in some cases, but the percentage of cases appears to be quite low, according to Hindman (1988). This thought-provoking study had convicted child molesters write an extensive autobiography. They were informed that on completion of their biography they would be polygraphed on the accuracy of their work. If they failed the polygraph test, they would be returned to jail. Prior to the study (i.e., before 1982) 67% of the offenders claimed to have been abused as children, compared to 29% of the offenders who participated in the study. In addition, 21% of the offenders prior to the study reported committing an offense as a child, compared to 71% during the study (immunity was granted for previous sex crimes). "In fact, what this suggests is these men began their deviant cycles very early in 51 life and rather than being victims as children, these men were abusing as children" (p. 3). These striking results must be evaluated cautiously, in view of a weak time-lagged experimental design, and the unethical use of coercion by the experimenters. Freund, Watson, and Dickey (1990) also examined the accuracy of sex offenders' self-reports, although they took a somewhat different approach. Male subjects were heterosexual (77) and homosexual pedophiles (54), child molesters (51), adrophilic clients (51) and gynephilic offenders (36) and controls (75). These groups were formed on the basis of a phallometric test to reveal their age and gender preferences for a sexual partner. Pedophiles' preferred immature partners; child molesters' preferred mature partners even though they had committed an offense against a child. Self-report questionnaires, the Erotic Preference Examination Scheme (EPES) and the Pedo Admitter scale, revealed two major findings. First, heterosexual and homosexual pedophiles' reported having been sexually abused by a male offender in childhood, more than the gynephilic controls, gynephilic offenders, and the androphilic controls. Second, the pedophiles' reporting child sexual abuse had the highest scores on the Pedo Admitter scale. This interdependence between admissions of child sexual abuse and pedophilic tendencies taints the creditability of these reports. To further investigate these findings Freund and Kuban (1994) undertook a more extensive analysis of the data. Their major concern was that the pedophiles' reported more child sexual abuse than the gynephilic offenders and controls, but not the child molesters. (For this analysis, incest offenders as well as those who denied their crime, were eliminated from the groups.) The group consisting of pedophiles was separated into three sub-groups based on answers from the EPES regarding sexual fantasies of young children or early pubescent children. They were described as either "admitters," "partial admitters," and "nonadmitters." Results indicated that pedophiles not admitting to a deviant sexual orientation differed from these other two groups. "This might indicate that those who admitted to their pedophilic erotic preference did not fabricate as much and therefore were disadvantaged in the comparisons with the remaining offender groups (p. 562). Overall, there appeared to be a strong relationship between pedophilia and reports of child sexual abuse. A more direct approach to modeling deviant sexual activities was undertaken by Becker, Hunter, Stein, and Kaplan (1989). They examined the sexual responses of adolescent sex offenders who had themselves been abused as children. Using a phallometric procedure suggested "that a history of having been sexually abused is associated with higher erection responses to cues involving noncoercive as well as coercive sexual interactions with children, relative to nondeviant cues" (p. 359). To explain these results the authors' suggested that the previously abusive experiences activate deviant masturbatory fantasies, or serve as a basis for enacting their previous victimization. Becker et al.'s (1989) data do not prove that child sexual abuse causes pedophilia, in fact, "most abused persons never become offenders" (p. 562). Critics of learning theories emphasize that males who have been sexually abused as children have a choice to continue or end the cycle of abuse. This is illustrated by a case study from Conran, Neugebauer, and Wizesinski-Conran (1989). A child molester named Hal had been sexually abused repeatedly by both his father and uncle during childhood. Therapists working with Hal suggested that his offenses were a consequence of this abuse. Hal 53 rejected this explanation, he believed associating "with criminals and drug users contributed to his abuse of children" (p. 38). Hal chose to meet new friends and associates through a religious organization. According to the authors, Hal refused to view his deviant behaviors through a learning paradigm. Summary. The learning of deviant sexual activities through conditioning (e.g., pornography) and modeling (i.e., child sexual abuse) were considered in relationship to sexual offenses committed against children. Goldstein and colleagues (cited in Marshall 1988; Goldstein, 1973) found that pedophiles' reported low levels of exposure to sexually explicit materials as an adolescent or adult. In contrast, Marshall (1988) found child molesters, rapists, and controls reported being exposed in adolescence to hard core pornography more frequently than the incest offenders and controls. In adulthood, the child molesters and rapists were more apt to use pornography. Most important, over one third of the child molesters and rapists were incited to commit a sexual crime through the use of pornography. Additionally, there may be a relationship between violent pornography and the propensity to rape (Nemes, 1992). Prior to addressing the issue of whether the sexual abuse of a child contributes to deviant behaviour through modeling, researchers have questioned whether offenders' reports of abuse are genuine. Pedophiles who admit to a deviant sexual orientation appear to be less likely to fabricate experiences of child sexual abuse, than those who do not admit their sexual orientation. Adult pedophiles' arousal levels need to be evaluated according to their sexual experiences during childhood (e.g., abusive and non-abusive). Homosexual pedophiles frequently reported their first sexual experiences being with a 54 male under 14 years of age; while, heterosexual pedophiles often reported their first sexual encounter was with a prostitute (Goldstein, 1973). Implications of the Learning Theories. Although specific information relating to pornography was only briefly mentioned in the adolescent and adult files, the following information was recorded. First, any information reflecting conservative attitudes towards sexuality in his family of origin was considered (Appendix B2, Question 3), as these attitudes may shield him from exposure to pornographic materials. Second, offenders' reported exposure to either soft or hard core pornography during adolescence and adulthood was recorded (Appendix B l , Question 26; Appendix B2, Question 24). Information readily available in the files was reports of offenders' experiences of sexual abuse during childhood or adolescence (Appendix B l and B2, Questions la, lb and lc). This information was considered in relationship to the child molester's sexual preferences (Appendix B l , Question 18). 2.7 The "Third Force "Theories "Third force" theories, such as phenomenology, humanism, and existentialism, contrast sharply with deterministic psychology. These theories do not focus on explanations in terms of physiology or unconscious motivations like psychoanalysis, nor reductionistic or mechanistic explanations like learning theory (Robinson, 1979). "If there is a common tie joining the majority of third force theories, it is the European intellectual tradition begun in the 1930s and 1940s: a tradition whose own origins are to be found in 55 the writings of Hegel, Marx, and Brentano" (Robinson, 1979, p. 238). These philosophers reacted against Kant and rationalism, although even these ties are weak (Robinson, 1979). Phenomenological explanations focus less on physiology and more on the subjective perceptions and experiences stemming from physiological events. An apperception for explanations based on subjectivity is shared by humanists, even though their explanations emphasize personal agency, choice, and freedom. The freedom to choose, according to the existentialists, exists in an environment lacking reason and purpose (Reber, 1987). Third force theories, particularly humanistic explanations, have been applied to social issues ranging from labor-management conflicts to child sexual abuse (Giarretto, 1978). In 1971, a Child Sexual Abuse Treatment Program (CSATP) based on humanistic psychology, began taking referrals involving incest cases (Giarretto, 1978). Members of the program reached a joint decision to adhere to 10 major premises: 1. The family is viewed as an organic system. Family members assume behavior patterns to maintain system balance (family homeostasis). 2. A distorted family homostasis is evidenced by psychological symptoms in family members. 3. Incestuous behavior is one of the many symptoms possible in troubled families. 4. The marital relationship is a key factor in family organic balance and development. 5. Incestuous behavior is not likely to occur when parents enjoy mutually beneficial relations. 6. A high self-concept in each of the mates is a prerequisite for a healthy marital relationship. 7. High self-concepts in the parents help to engender high self-concepts in the children. 8. Individuals with high self-concepts are not apt to engage others in hostile-aggressive behavior. In particular, they do not undermine the self-concept of their mates or children through incestuous activities. 9. Individuals with low self-concepts are usually angry, disillusioned, and feel they have little to lose. They are primed for behavior that is destructive to others and to themselves. 56 10. When such individuals are punished in the depersonalized manner of institutions, the low self-concept/high destructive energy syndrome is reinforced. Even when punishment serves to frustrate one type of hostile conduct, the destructive energy is diverted to another outlet or turned inward (p. 67). An assumption underlying all of these premises is that the psyche and society are interlinked, thus the offender's responsibilities to various societal roles (i.e., son/parent, employee/employer, citizen/politician, etc.) are addressed. To assist the offender in fulfilling his roles, the CSATP provided treatment not only for the offender, but for the mother-daughter and father-daughter dyads, along with marital, family, and group counseling. The principles and methodologies developed at the CSATP reach beyond a traditional medical model that focuses on individual psyches or pathologies to the exclusion of the individual within their social context. Another humanistic perspective on the issue of child sexual abuse was expressed by Szasz (1980, 1991). Szasz (1991), a professor of psychiatry, viewed adults who sexually abuse children as criminals (although criminality requires an explanation)—not mentally ill. Professionals frequently overlook this fact, because "sexual acts involve the use of the human body, inviting the illusion that sexual problems are medical problems" (Szasz, 1991, p. 34). The effect of this illusion is that abusive adults are believed to be suffering from medical problems, ending in a lack of control over their sexual impulses. Sexual crimes are reduced to sexual disorders requiring court ordered treatment to be rehabilitated. When rehabilitation inevitably fails it is attributed to the patient being untreatable, or a lack of medical knowledge with respect to the disorder. Strongly opposed to this line of reasoning, Szasz (1980) stated, "there is not a shred of evidence that "sexual deviates" are less able to resist their erotic preference and rituals than 57 devoutly religious persons are able to resist their theological preferences and rituals" (p. 138). According to Szasz (1980/1991), sexual deviates are not out of control, they choose to commit a crime. They should not be labeled "mentally ill" (as suggested by psychiatric theories), nor should they be referred to as "sex offenders" (as suggested by psychological theories) (Szasz, 1991). "When a man robs a bank, we do not think of him as a financial offender: When he leaves prison, he is not subjected to "psychological counseling for financial offenders" in the expectation that it will rehabilitate him" (p. 37). Nevertheless, this is the therapeutic approach taken with adults who sexually abuse children. Children are left vulnerable and the government loses funding when court ordered treatments fail, yet professionals providing the treatment benefit financially. Personal investments make it difficult for professionals to entertain the notion that sex crimes are a legal or ethical issue, not a psychiatric disorder (Szasz, 1980). Viewing child sexual abuse as an ethical issue impinging upon our legal system does not diminish the need for an etiological explanation for child molestation. Brandt (1959) examined ethical values from both a statistical (i.e., factors that may affect a person's values) and theoretical (i.e., laws of development) standpoint. Statistically, it appears that numerous observations have been made regarding the source of ethical reasoning: "home influences," "other prestige figures," "information, consistency, and personal experience" "personal interests, needs, temperament" and "sympathy." According to Brand, it appears children learn the meaning of justice in terms of discipline and punishment from their guardians, or home influences. They notice that their guardians are judgmental of their behaviors. Other prestigious figures' (or even groups') influences outside the home are dependent on attributes of the child. For adults, 58 prestigious figures or groups do not change their fundamental belief systems, instead ethical judgments regarding complex issues are "modified to some extent, somehow" (p. 126). Other modifying factors are new information, consistency, and personal experiences. New information includes reports, literature, films, testimonies, religious materials, and observations. Personal experiences comprise not only an event, but an affective reaction to the event. Modifying factors are reasoned through to construct a consistent set of beliefs, but do personal interests, needs, and temperament interfere with this process? The statistical evidence, according to Brandt (1959), is lacking on the effect of one's personal interests and needs when structuring principles. Interests and needs do appear to affect social and political attitudes, while personalities may affect the kinds of values held (Brandt, 1959). Finally, sympathy can modify a belief system. A sympathetic response "covers such phenomena as the impulse to relieve a person or animal perceived to be in distress, being shaken emotionally by perception of a pain or distress situation, and experience of a thrill of joy at seeing another transported by joy on account of some good that has come to him" (p. 133). Evidence indicates that sympathetic responses appear early in life and are not dependent upon conditioning or modeling as learning theorists suggest (Brandt, 1959). After examining the development of values from a statistical perspective, Brandt considered three theoretical positions to explain the development of ethical values. As mentioned previously, it was the sympathetic response that interfered with his ability to accept learning theories. (Brandt held that there were unlearned tendencies within the human mind.) The second theory is Freud's psychoanalytic theory. The limitation of this theory, according to Brandt, is that it does not provide a means for modifying ethical 59 values, nor an explanation for universal principles. The third theory, gestalt psychology, is based on innate rules or laws governing the process of perception. Like perception, ethical values involve following rules or universal laws. "Basic values never change; perhaps what changes is just the meaning or psychological character of the situation being appraised" (p. 147). Summary. Third force theorists discuss freewill when considering the etiology of sexual deviancies. Giarretto emphasized an individual's choice to be involved in incestuous activities within the context of family and societal relationships, while Szasz discussed deviant sexual activities in terms of personal choice regardless of the social context. Szasz insisted that engaging in child sexual abuse is an ethical or legal issue, not a medical condition. The question remained, "What is the etiology of unethical reasoning?" Statistically, Brandt considered both internal and external factors. Theoretically, Brandt focused on moral laws parallel to gestalt psychology's laws of perception. Overall, professionals who are advocates of the third force theories believe each person chooses to value or devalue sexual relationships. Implication of the Third Force Theories. In response to the debate on the effects of mental illness and social relationships on child molestation, both the offenders' psychiatric diagnoses (Appendix B l , Question 21; Appendix B2, Question 19), contact with mental health professionals (Appendix B2, Question 20) and participation in social roles (Appendix B l , Question 9; Appendix B2, Question 12) were recorded. Furthermore, child molestation was viewed as an ethical issue (Appendix B l , Questions 14 and 31; Appendix B2, Question 13). Brandt's 60 suggestion that home influences (Appendix Bl , Questions 27), other prestige figures (Appendix B l , Question 5; AppendixB2, Question 7), information, consistency, and personal experience (Appendix Bl andB2, Questions la, lb, and lc), personal interests, needs, temperament (Appendix B l , Questions 28 and 29; Appendix B2, Question 25), and sympathy (Appendix B l , Question 30), was taken into consideration. 61 Chapter 3 A MULTI-THEORETICAL APPROACH After examining the seven theoretical frameworks relating to child sexual abuse it is apparent that a wide range of explanations are provided for offenders engaging in sexual activities with children. Psychoanalytic and analytic theorists stress the importance of premature sexual experiences, since possible consequences of these experiences include excessively acting out of perversions (Freud, 1915/1962), or eventually becoming an abuser (Kramer, 1976). Physicalistic theorists explore the likelihood of mild brain damage in child molesters as indicated by psychological neuropsychological, and medical tests (Hucker et. al., 1986). While these theorists present traditional points of view in the psychological field, other psychologists are in the vanguard of modern psychology putting forth novel explanations for child molestation. For example, Marshall's (1989) attempt at applying attachment theory to sex offenders, resulted in the suggestion that offenders are actually pursuing intimacy through sexual encounters. His work with colleagues elaborated on this perspective by suggesting that child molesters' negative affect and cognitive errors increase prior to fantasizing about, or committing a sexual offence (Ward, Hudson, & Marshall, 1994; Ward, Hudson, & Marshall, 1995). Such explanations can not be considered formal theoretical or even informal theoretical perspectives, but these explanations do provide avenues to be explored. 62 Some theoretical perspectives, or explanations, may be more successful than others in revealing certain types of molesters' psychological motivations for engaging in sexual activities with children. For example, when learning theorists examined institutionalized offenders' use of sexually explicit materials, they found that homosexual pedophiles had an interest in explicit sexual materials depicting homosexual acts during adolescence (Goldstein et al., 1971). Lanning (1992), an agent with the Federal Bureau of Investigation, took into consideration pedophiles' choice of pornography or erotica when investigating crimes. However, he did not specify if child pornography used by this type of offender depicts homosexual or heterosexual acts. Nor is it known whether homosexual pedophiles are more apt to commit offenses after using these sexually explicit materials. This issue is crucial. Evidence suggests that sexually explicit materials have been used by one-third of child molesters to incite a crime (Marshall, 1988). Clinicians require clarification on how to discourage child molesters from using these materials to combat the problem. Learning theorists, in particular, have a unique opportunity to contribute to the literature on child molesters (i.e., homosexual pedophiles) and pornography. Likewise, other theoretical perspectives may be useful in exploring the motivation of specific types of offender that have investigative and clinical implications. Finkelhor and Araji (1986) developed a four factor model using a number of theories to explain the diversity in pedophilic behaviors. Their model consists of four explanations (i.e., "emotional congruence," "sexual arousal," "blockage," and "disinhibition"), evaluated from two levels (i.e., psychological and socio-cultural) (See Table 2). It should be mentioned that this system was not designed to address the psychological complexities of pedophilia that requires detailed analysis of pedophiles' 63 Table 2. Explanations of Pedophilia Theory type Level of Explanation Individual Social/Cultural Emotional congruence Arrested development Low self-esteem Symbolic mastery of trauma Identification with aggressor Narcissistic identification Male socialization to dominance Sexual arousal Blockage Arousing childhood experience Traumatic childhood sexual experience Operant conditioning Early modeling by others Misattribution of arousal Biological factors Oedipal conflict Castration anxiety Fear of adult females Traumatic experience with adult sexuality Inadequate social skills Marital disturbance Child pornography Eroticization of children in advertising Repressive norms about masturba-ion, extramarital sex Disinhibition Impulse disorder Senility Alcohol problem Psycholsis Situational stress Failure of incest avoidance mechanism Cultural toleration Pornography Patriarchal perogatives 64 motivations, however, it is useful in explaining the process or outcome of having pedophilic tendencies, For this project, the seven psychological theories were reviewed to explain child molesters' motivations from various perspectives. Nevertheless, all seven perspectives shared common factors (Gold, 1996). First, all of the theoretical approaches viewed child molesters as disturbed individuals. That is, none of these theories considered a sexual orientation towards children as normal, or engaging in sexual activities with children to be acceptable. The disturbances experienced by molesters were usually viewed as psychiatric disorders, although one theorist (i.e., Szasz) simply considered the disturbance an ethical or legal issue. Second, all of the theoretical orientations were plagued by the problem of obtaining accurate information from the child molester (e.g., PPG Scores, Self-Reports, and Psychiatric Reports). Third, the methodologies used within each theory involved an empirical approach (i.e., observations and empirical data). Fourth, none of the seven theoretical approaches provided an adequate explanation for the occurrence of child molestation. Each theory is limited in its potential to explain and predict sexual offenses against children. Fifth, all theoretical approaches assumed child molestation is the product of multiple factors. Finally, all of the theories were useful in providing information regarding child molesters' motivations for committing sexual offences against children, in order to collect data using potential causal variables. Not all of the potential causal variables suggested by these theories, could be coded using information from adolescent and adult child molesters' mental health files. A preliminary examination of the files revealed that some of the files did not contain information pertaining to these topics, or the information was only available inconsistently. 65 As a result, it was not possible to test each theory. Instead the theories were only used in the selection of potential causal variables for the cluster analysis, if they could be consistently coded. The potential causal variables were used in the cluster analysis, then analyzed to see whether information pertaining to the child molesters' offenses (i.e., contrasting variables) would distinguish between the clusters. Reasons for collecting specific types of contrasting variables regarding the child molesters' offenses are discussed in the next section. 66 Table 3. List of Potential Causal Variables by Theory Theoretical Perspectives Potential Causal Variables Coding Sheet Question No. s (adolescent) (adult) Psychodynamic Attachment Three Mental Functions Sexually abused as a child Sexual activities Penile plethysmograph Concrete reasoning Projective test Close relationships Emotionally abused Physically abused Neglected Sexual Aggression Model Intimacy, loneliness, etc. Images of self/other Intimate relationships Cognitive Scale Description of self/others Responsibility for offense Reactions to offense la, lb, lc 18a-g 16* 32* 5* 2 3 4 6* 7* 8* 10 11,12,13* 14 15* la, lb, lc 2b 15 7 4 5 6 8 11* 10 13 14* Psychophathic Physiological Primary diagnoses Secondary diagnoses Other diagnoses (Penile plethysmograph PCL-R Scores UCLA Empathy Scale Sexual problems IQ Index Intelligence Brain damage/head injuries Transient Temperament Biological parents, alcohol Biological parents, drug Childhood diagnoses (Concrete reasoning 21 21 21 18a-g 19* 30 20* 17 22* 28* 23 24 25 16* 19 19 19 -) 17 2a, 2c 18 16 21 22 23 15) *Variables, unable to code Parenthesized, previously mentioned (table continues) 67 Table 3. (continued) Theoretical Perspectives Potential Causal Variables Coding Sheet Question No.s (adolescent) (adult) Learning Masturbation - 3 Use of pornography 26 24 (Sexually abused as a child la, lb, lc la, lb, lc) (Penile plethysmograph 18a-g -) Third Force (Primary diagnoses 21 19) (Secondary diagnoses 21 19) (Other diagnoses 21 19) Mental health involvement - 20 Socialization 9 12 (Responsibility for offense 14 13) Internal/External Control 31 * Fear of Negative Evaluation 27* (Close relationships 5* 7) (Sexually abused as a child la, lb, lc la, lb, lc) (Temperament 28* -) Personality measure 29* Needs/Interests - 25* (UCLA Empathy Scale 30 -) * Variables, unable to code Parenthesized, previously mentioned 68 Chapter 4 INFORMATION RELATING TO CHILD MOLESTERS' OFFENSE(S) A literature review relating to child molesters' current sexual offense(s) was undertaken to consider possible contrasting variables to distinguish between the cluster solutions. Characteristics of the victim were considered including, age, gender and their relationship to the offender. Other characteristics of the incidents that occurred prior to, during, and after the offense(s) are fantasies, substance abuse, seduction techniques, sexual activities, and silencing techniques. These variables were recorded in the last section of the coding sheets (See Appendix Cl and Cl). 4.1 The Victim(s) 4.1.1. Age. Gender, and Relationship to Offender A classic study by Gebhard, Gagnon, Pomeroy, and Christenson (1965) described the victim in terms of age, gender, and in relationship to the offender. Heterosexual pedophiles targeted slightly younger children than incestuous, heterosexual aggressive, or homosexual pedophiles. The relationship of the offender to the victim was described as either "stranger," "acquaintance," "friend," or "relative." (Incestuous offenders by definition are "relatives.") Forty one percent of the heterosexual and 49% of the homosexual pedophiles were considered "friends." When the last three categories were 69 collapsed, it became apparent that the majority of heterosexual (60.1%) and homosexual (67.6%) pedophiles were known to the victims. The one exception was heterosexual aggressive offenders who were frequently strangers to the child. Since these three characteristics of the victim(s) appear to be distinguishing features of some child molesters, this information was recorded (Appendix C l and C2, Victim Sections, Questions 1-3). 4.2 Current Sexual Offense(s) 4.2.1 Prior to the Offense Sexual fantasies have not only been associated with the development of a deviant sexual orientation, but to preliminary activities leading up to an offense (as mentioned previously). Looman (1995) interviewed 23 child molesters, 19 rapists, and 19 non-sexual offenders about their fantasies. "Child molesters reported that they were more likely to fantasize about a child than an adult if they were feeling depressed, argued with their wife or girlfriend, felt rejected by a woman or were angry" (p. 327). Rapists were somewhat more likely to fantasize about an adult when angry, and non-sexual offenders never reported engaging in fantasies while angry. Child molesters reported feelings of fear and guilt when fantasizing about children, but not when fantasizing about adults. There were no differences in emotional states when child molesters, rapists, and non-sexual offenders fantasized about adults. It was suggested that intense negative feelings accompanied by rationalizations occur prior to planning a sexual offense (Appendix C l and C2, Current Sexual Offense(s) Sections, Question 1). 70 4.2.2 Substance Abuse Rada (1976 cited in Coleman, 1982) assessed 203 child molesters: 49% percent had consumed alcohol during the offense. Of these offenders, 52% were categorized as alcoholic. Alcoholism appears to be a major problem for pedophiles (Langevin & Lang, 1988), although less of a problem for incestuous fathers (Williams & Finkelhor, 1990). Alcoholism can cause impairments in three areas: physical, mental, and social (Langevin & Lang, 1988). Physically, alcohol may stimulate sexual arousal in the short-term, particularly among young offenders. In the long-term alcohol may lead to sexual dysfunctions, such as an inability to maintain an erection. Mental impairments, like poor judgments, may interfere with the ability to maintain social relationships. Overall, drug abuse is less prevalent than alcohol abuse for child molesters (Langevin & Lang, 1988). Specific types of drugs can be devastating, for example, amphetamines or phencyclidine (PCP) has been associated with mental impairments (e.g., paranoia) that trigger violence. Assaults and homicides have been attributed to the effects of PCP, although violent activities are rare among child molesters (Sgroi, 1978). Any form of substance used during the offense was recorded (Appendix C l and C2, Current Sexual Offense(s) Sections, Question 2). 4.2.3 Seduction Technique Child molesters often initiate their offense by "courting" a child in the same manner as an adult male might pursue an adult female. One suspect, "began his courtship with dinners, ballgames, trips to an amusement arcade, and two expensive gifts, a pinball 71 machine and a stereo set" (Caplan, 1982, p. 50). To develop a sexual relationship with the child, the molester lowers the child's inhibitions. For instance, the molester may arrange an event where the child must undress (e.g., at a swimming pool) (Lanning, 1986) or is exposed to pornograpy (e.g., leaving an x-rated video in the VCR) (Erickson, Walbek, & Seely, 1988). In contrast to the courting technique, Sandfort (1984) suggested that sexual activities may occur abruptly, even during the first contact with the child. A typical scenario involves a male victim who is "13 or 14 years old, is at a park or playground, and is offered money or recreational rewards for his co-operation. The assault.. .involves fondling and fellatio" (Duncan & Hern, 1980, p. 13). Another typical scenario involves incest, "My father came into my room and I would think that I was dreaming... At first I'd wake up just as he'd be leaving the room, and later on I felt him fondling me, touching my breasts and my genitals" (Delin, 1978, p. 121). The technique for initiating sexual activities (Appendix C l , Current Sexual Offense(s) Section, Question 3; Appendix C2, Current Sexual Offense(s) Section, Question 3-4) was recorded. 4.2.4 Type of Sexual Activity, including Frequency and Duration "Most people imagine that sexual assault of a child by an adult will be a brutal and violent act involving physical trauma to the child by forcible penetration of the vagina, rectum, or mouth. Although these acts may occur, and occasionally do occur, most often the sexual assault will be nonviolent and without forcible penetration" (Sgroi, 1978, p. 131). Sgroi's comments appear to be substantiated by recent studies. Mair (1993) found that sexual acts imposed on children not involving rape are twice as prevalent as acts of 72 rape (i.e., penile penetration involving the vagina or anus). Prevalent sexual acts consisted of disrobing, fondling, digital penetration and oral-genital contact. For female victims under 10 years of age, fondling and vaginal contact were common; male victims in the same age range frequently experience anal contact or oral-genital contact (Erickson, Walbek, & Seely, 1988). The youngest children were often viewed as "masturbatory aids," while the older children were considered "partners" (p. 84). In addition to the child's age, the opportunity to commit sexual acts may influence the type of acts committed. For instance, incest cases "ordinarily involve multiple contacts over time, with gradual progression from fondling to masturbation of the perpetrator and efforts at penetration" (Erickson et al., 1988, p. 84). The type of sexual activity that occurred were coded for each child molester (Appendix C l , Current Sexual Offense(s) Section, Questions 4, 5, 6, and 7; Appendix C2, Current Sexual Offense(s) Section, Questions 5, 6, 7, and 8). 4.2.5 Silencing Techniques Examining cases with female victims, Lang and Frenzel (1988) compared the silencing techniques of EF and IF offenders. Both types of offenders (40%; and 85%, respectively) told their victims that the sexual activities were a "special secret" (p. 311) between them. This served to evoke feelings of distinction and privilege in the children "but in a bewildering way" (p. 312). Other types of statements that intended to silence the child included "I love you" (EF 6%; IF 68%) and the "Daddy's girl" or the "You're special" approach (EF 16%; IF 28%). Non-verbal techniques for maintaining the child's silence included doing a favor for the child (EF 4%; IF 23%) or portraying a fatherly 73 image (EF 16%; IF 17%). For incest cases only, fathers (43%) threatened the possibility of a family disintegration if the child disclosed. Further information is needed to determine whether type of offender has an influence on children's disclosures of sexual abuse (Appendix C l , Current Sexual Offense(s) Section, Questions 8, and 9; Appendix C2, Current Sexual Offense(s) Section, Questions 9, and 10). Information documenting the child molesters' current sexual offences (i.e., the contrasting variables) was found in police reports that were readily available in the adolescent and adult files. The police reports contained detailed and extensive descriptions of the offenses. However, not all of the information (e.g., fantasies, silencing techniques) regarding the offenses was mentioned in the reports. Additional sources (e.g., psychiatric or psychological assessment, pre-sentence reports, and victim impact statements) were used. 74 Chapter 5 METHOD 5. 1 Subjects Two hundred male adolescent (i.e., 12 to 17 years of age) and adult (i.e., 18 years and over) child molester files were coded through Forensic Psychiatric Services. They provide services for persons who are remanded, in custody, or under the Lieutenant Governor's direction or a court order. The adolescent offenders (n=100) had been assessed and treated at Youth Court Services in Burnaby, British Columbia. The adult offenders (n=100) were assessed, treated, or supervised at an outpatient clinic in Vancouver, British Columbia. Adolescent child molester files were considered for this project if they admitted, or were charged with or convicted of committing a sexual offense against a person under the age of 12 or 5 years younger than the offender. Likewise, the adult child molester files were examined to see if they admitted or were charged/convicted of a sexual offense against a person under the age of 18, and at least 5 years younger than the adult offender. The following is a list of possible sexual offenses, using the numbers listed in Martin's Annual Criminal Code-1997 book (Greenspan & Rosenberg, 1996): 151 Sexual interference 152 Invitation to sexual touching 155 Incest 75 159 Anal intercourse 160 Bestiality 170 Parent or guardian procuring sexual activity 171 Householder permitting sexual activity 271 Sexual assault 272 Sexual assault with a weapon, threats to a third party or causing bodily harm 273 Aggravated sexual assault In cases where the adolescent or adult offender was not charged/convicted with a sexual offense, they were usually charged with a non-sexual offense (e.g., assault) but admitted committing at least one of the sexual offenses listed above. 5.1.1 Adolescent Sex Offender Files A list of child molester files was generated from information collected for another study being conducted at Youth Court Services (YCS) (D. O'Toole, personal communication, July 8, 1997). The YCS list consisted of 210 active and inactive assessment and treatment files on adolescent sex offenders (not necessarily child molesters). These files were coded, if they pertained to child molesters, contained sufficient information (i.e., psychiatric, psychological, and social work assessment, plus the police report) and met the criteria described above. Five additional files were coded, that were not on the list to reach a sample size of 100. \ 5.1.2 Adult Sex Offender Files Active or recently closed adult child molester files were coded, if they contained sufficient information (i.e., to complete the Psychopathy Checklist) and met the basic criteria. The files were coded chronologically, going back in time until 100 files were 76 completed. The information from the files was primarily derived from the Pre-Sentence and Police Reports, as well as from psychological and psychiatric assessments. 5.2 Measurements 5.2.1 Overview of Data Collection The coding sheets consist of three sections titled Appendix A, B, and C. Appendix A contains variables that describe the child molester's background (e.g., birth-date, education, parents, legal history, and substance use or abuse). Some of the items were modified for the adult population. For example, educational achievements after high school were described. In addition, the coding sheets for the adult child molester files contained two extra items (i.e., marital status, and number of children). The potential causal variables extracted from the literature are included in Appendix B. These variables were constructed for this project, or obtained from questionnaires, diagnoses, and other tasks/measurements. Different tasks/measurements were contained in the adolescent and adult files, an in-depth description of Appendix B is provided below. Appendix C contains variables describing the child molesters' current sexual offense(s). One additional variable was coded for the adult child molester files, that is, the child molesters' attempts to lower the victim's inhibitions (i.e., by offering the victim cigarettes, alcohol, drugs, or pornographic magazines/videos). 77 5.2.2 Content of Adolescent Sex Offenders' Files Questionnaires. Another study currently being conducted at Youth Court Services administered questionnaires to child molesters who had been assessed or treated at the agency (D. O'Toole, personal communication, July 8, 1997). Scores from two of the questionnaires were used in this project: a modified version of the Cognitive Scale for adolescents (Hunter, Becker, Kaplan, & Goodwin, 1991) and UCLA Empathy Scale (Mehrabian & Epstein, 1972). The Cognitive Scale is a 43-item scale designed to evaluate adolescent child molesters' cognitive distortions pertaining to sexuality. The cognitive scale for adults is a reliable (i.e., test-retest reliability was .76) and valid measurement (Abel, Gore, Holland, Camp, Becker, & Rathner, 1989), however, scales designed for adolescents needed to be revised (Hunter, Becker, Kaplan, & Goodwin, 1991). The authors of the latter article suggested using a Likert-scale (with a 1 to 5 range), rather than a forced-choice scale (using true or false) to reduce impulsive responses that are socially desirable. The Youth Court services study did use a 5-point Likert-scale, that is, the same type of Likert-scales used with adult child molesters. A questionnaire designed by Mehrabian and Epstein (1972) measures emotional empathy. Split-half reliability for this measure was .84 (Mehrabian & Epstein, 1972). Previous studies demonstrated that individuals high in affective empathy were less aggressive and more altruistic, than individuals low in affective empathy (Mehrabian & Epstein, 1972). The questionnaire consists of 33 items (e.g., "Little children sometimes 78 cry for no apparent reason.") rated on a 9- point scale ranging from "very strongly disagree" to "very strongly agree." Diagnostic and Statistical Manual of Mental Disorders (DSM). In 1952 the first edition of the DSM manual was published (American Psychiatric Association, 1980). The unique feature of this manual is the descriptions of different categories of mental disorders (American Psychiatric Association, 1994). The mental disorders are described (e.g., pedophilia), not the individuals who are disordered (e.g., pedophiles). Patients may share the same disorder, but be distinctive in other ways. The mental disorders described in the manual "reflect a consensus of current formulations of evolving knowledge " in the field (American Psychiatric Association, 1994, p. xi). For this project, three psychiatric diagnoses were recorded for each offender: "primary," "secondary," and "other" diagnoses. Unfortunately, these diagnoses were not tested for reliability or validity, since the adolescents were assessed prior to this study for clinical and legal purposes within the community. Penile Plethysmograph. The Penile Plethysmograph measures sexual physiological functioning (Youth Court Services, 1997). The equipment used to measure physiological arousal was a Farrell SP-300 (2 channel) biofeedback machine. The gauge connected to the machine was either an Indium-gallium or Barlow. Slides and audio tapes were used as the stimuli. There were eight slides consisting of nude males and females at various ages. The audio tapes describing sexual activities were two minutes in length. In response to these stimuli, the average arousal level (i.e., percentage of full erection) for seven categories (i.e., male, female, non-coerced, coerced, post-pubescent, pre-pubescent and child subjects) were calculated by a medical personnel. The reliability, validity, and 79 ethics of using the penile plethysmograph remains a controversial issue (See Gordon, & Verdun-Jones, 1983). For example, initial testing does appear to reflect the offenders' previous deviant sexual behaviours (Wormith, 1986) and recidivism rates (Malcolm, Andrews, & Quinsey, 1993), but it is unknown whether repeated testing can overcome the problems of faking (See Wormith, 1986). "Thus we are forced deal only with voluntary and honest research participants and patients at present" (Langevin, 1983, p.9). Other Measures Constructed for the Project. Information was coded on whether or not the adolescent molester had been emotionally, physically, or sexually abused as a child. Reports of being neglected, or experiencing a transient lifestyle during childhood were also coded. There was an interest in whether the adolescent molesters' biological parents were involved with alcohol or drugs. Diagnoses of fetal alcohol syndrome, conduct disorder, attention deficit disorder, concrete reasoning, or brain damage were recorded. Attachment variables focused on the child molesters' relationships, not only with their parents, but siblings, peers, and other significant people in their lives. Questions relating to the Sexual Aggression Model, pornography, and their feelings of responsibility for their current offense were collected. 5.2.3 Content of Adult Sex Offenders' Files Psychiatric Diagnoses. The psychiatric diagnoses recorded for adult child molesters were taken from the assessments written by psychiatrists, or previous assessments that were referred to in other reports. 80 Intelligence Measures. Psychological assessments frequently contained information describing the intellectual level of functioning demonstrated by the adult child molester during testing procedures. Psychopathy Checklist-Revised (PCL-R). Hare (1990) developed the Revised Psychopathy Checklist (PCL-R) to measure deviant personality traits and behaviours as described clinically by Cleckley (1976). The PCL-R rating scale consists of 20 items scored as 0, 1, or 2. A total score ranges from 0 to 40. Eight items comprise a Factor 1 score, which measures personality traits. Nine of the items measure actual behaviours and are referred to as the Factor 2 score. The PCL-R has been shown to be internally consistent, with interrater and test-retest reliabilities ranging from .85 to .95 (Hare, 1990; Hart, Hare, & Harpur, 1992). It is frequently used to predict recidivism and violence in male offenders (Hare, 1990; Hare, 1993) Other Measures Constructed for the Project. Other measures constructed for this project were similar to those collected from the adolescent molester files, with the addition of two types of information. The first type focused on the adult offenders' previous and current sexual activities. The second type evaluated whether the offenders' personal needs or interests were satisfied prior to their current sexual offense. Any modifications and additions that were made reflected the information contained in the files. 8 1 5.3 Procedure The literature on the etiology of child molestation was reviewed from seven theoretical perspectives: psychoanalytic, attachment, three mental functions, psychopathic, physiological, learning, and third force theories. As mentioned previously, four of these theories (i.e., psychodynamic, physiological, learning, and third force theories) are commonly used approaches when conceptualizing modern psychology and psychopathology, while the remaining three theories (i.e., attachment, three mental functions, and psychopathy) are directly applicable to child molesters. A list of potential causal variables was constructed from these theories. The adolescent and adult files were coded on as many of these potentially discriminating variables as possible considering the amount and type of information that was available (See Appendix A l - C l and A2-C2). The adolescent child molesters' files were coded by a graduate student (i.e., the author) and an undergraduate student at Youth Court Services. The same students coded the adult child molesters' files at the Vancouver Clinic. At both sites, the coding procedure began with a practice session using approximately 3-5 files. Each coder scored the files independently, and recorded where they obtained the information. The coding sheets for each file were compared and contrasted. Both coders' explained the reason for choosing a particular score, and why they obtained the information from a particular source. For items that were difficult to score, the coders collaborated on a scoring system. The files coded in the practice sessions were not used in the data analysis. 82 After the practice session discussions the two coders resumed working independently on the files. However, they did share a data dictionary containing the operational definitions for specific variables. For example, the exact range for the number of residences or types of employment that would be considered a reflection of an unstable, moderately stable, or stable lifestyle were defined. These definitions were elaborated upon by the coders when they considered it necessary. The data collected after the practice sessions were analyzed using a clustering technique. A cluster analysis technique groups "entities into homogeneous subgroups on the basis of their similarities" (Lorr, 1983, pi). In this case, the entities are individuals who had contact with the legal system as a consequence of committing a sexual offense against a child. Clustering techniques can be contrasted with factor analysis techniques that focus on the similarities of variables, not entities. There are a number of reasons for using statistical techniques based on entities (Lorr, 1983). First, these techniques analyze data thought to be representative of numerous populations whose statistical parameters remain undiscovered. Second, by using these techniques anticipated clusters representing the populations can be tested, and unanticipated clusters can be detected. Finally, cluster analysis techniques are useful in making predictions. Prior to using a cluster analysis technique the potential causal variables were converted into comparable units. This was necessary, as variables with larger variances would have been given heavier weights than variables with smaller variances. Transforming variables into units of equal weights was accomplished by standardizing all of the variables (Lorr, 1983). Also prior to analysis the data was examined for outliers (i.e., over two standard deviations away from the mean). It was necessary to determine 83 whether there were errors in coding, or if the outliers denoted individuals who were not adequately represented in a sample (Everitt, 1980). No outliers of this type were found, although some errors in entering the data needed to be corrected. The next step was to choose a matrix and calculate the similarity between pairs of entities. A suitable matrix is usually based on one of four scales (i.e., nominal, ordinal, interval, and ratio). In this project, the data matrix contained a mixture of scales (i.e., nominal and interval). This is not an uncommon phenomenon when collecting data in the "real world" (Lorr, 1983). Romesburg (1984) suggested a number of strategies for dealing with this situation. The simplest solution is to ignore the problem by treating the nominal data like interval scales (e.g., an Euclidean distance measure can be used). According to Romesburg, the results do not appear to be negatively affected by using data from different types of scales when using a cluster analysis technique. When analyzing the data a hierarchical structural model, rather than a nonhierarchical scheme, was used for the following reasons (Lorr, 1983). First, nonhierarchical schemes performance have been rated poorly. Second, a hierarchical scheme is sensitive to developmental arrangements. Ward's minimum variance clustering method is a widely used hierarchical method (Romesburg, 1984). It is an agglomerative technique, as it starts with all of the entities forming their own cluster and ends with all entities in one cluster. Each step throughout this process merges two clusters that will increase the index E, or variance, the least amount. This method reduces the number of computations, while still giving a nearly optimal solution. The number of clusters for an optimal solution was based on previous research, with the restriction of 10 subjects per cluster. 84 To test the validity of the results obtained from this cluster analysis technique two measurements were employed. First, the clusters formed from the adolescent child molesters' data were compared with the clusters from the adult child molesters' data. The number of clusters should be the same for the two samples, or demonstrate convergent validity. Considering that each cluster can be viewed as an explanatory concept, similar types of clusters increase our confidence in psychologically based classification systems. Second, clusters derived from adolescent and adult child molesters' samples were examined in relationship to the descriptions of their offenses. Any similarities found in the types of offenses committed by adolescent or adult child molesters from the same type of clusters suggests that the psychological differences were manifested through their offenses (See Appendix C). 85 Chapter 6 RESULTS This chapter is divided into three major sections. The first section is a descriptive analysis of the child molesters' characteristics for both the adolescent and adult samples. The second section discusses the potential causal variables used in the cluster analysis technique, as well as the five-cluster solutions obtained for the adolescent and adult samples. The contrasting variables used to distinguish between the five-clusters solutions are discussed in the last section. 6.1 Child Molester Characteristics The average age of the adolescent child molesters at the time of committing their current offense was 14.46 years (SD=1.81), compared to adult child molesters whose mean age was 34.49 (SD=10.86). A descriptive analysis of the adolescent and adult child molesters' current non-sexual and sexual charges are listed in Table 4. For both samples there appeared to be more sexual, than non-sexual charges. The majority of charges were for sexual assault. This was in contrast to previous convictions where there were less sexual offenses than non-sexual offenses. The largest number of non-sexual offenses was for theft. 86 6.1.1 Background Information. Information was collected on the adolescent and adult child molesters' background, specifically their educational and employment achievements. The average grade level achieved by the adolescent child molesters was 7.8 (SD=1.55), compared to 10.23 (SD=2.04) for the adult molesters (See Table 4). A large majority of the adolescent child molesters had some previous work experience (43.0%), although current employment may have been disrupted as a consequence of their offenses. The adult molesters were often employed in semi-or unskilled positions (See Table 5). Table 5 lists additional background information collected on the adult child molesters, not applicable to adolescents. Inter-rater reliabilities for these descriptive analysis variables were completed on 20% of the adolescent and adult files (See Table 6). The files were randomly selected and independently coded. 87 Table 4. Descriptive Analysis for Adolescent and Adult Child Molesters Variable Adolescent Adult Age 14.46 34.49 Frequency Percentage Frequency Percentage Current Charges Non-Sexual Assault 3 Robbery/Weapons 1 Property 3 Obstruction 5 Theft 10 Drug/Alcohol _0 22 Current Charges Sexual Sexual Assault (SA) 82 S A, with Weapon 1 Aggravated SA 4 Invitation to Touch 12 Sexual Interference 22 Sexual Touching 12 Exposure, under 14 2 Gross Indecency 1 Indecent Assault (f) 0 Buggery 3 Indecent Assault (m) 0 Incest 2 Sexual Exploitation _1 142 13.6 4.5 13.6 22.7 45.5 0.0 99.9 57.7 0.7 2.8 8.5 15.5 8.5 1.4 0.7 0.0 2.1 0.0 1.4 07 100.0 4 0 0 5 0 _0 9 75 0 0 4 25 6 3 3 28 2 4 0 _1 151 44.4 0.0 0.0 55.6 0.0 0.0 100.0 49.7 0.0 0.0 2.6 16.6 4.0 2.0 2.0 18.5 1.3 2.6 0.0 0.7 100.0 Note: A molester may have multiple charges for their current offense. (table continues) 88 Table 4 (continued) Descriptive Analysis for Adolescent and Adult Child Molesters Variable Adolescent Adult Grade Level 7.84 10.23 Frequency Percentage Frequency Percentage Previous Convictions Non-Sexual Assault Robbery/Weapons Property Public Order Theft Drug/Alcohol 6 0 4 4 20 _ l 35 17.1 0.0 11.4 11.4 57.1 2.9 99.9 23 4 2 10 37 15 91 25.3 4.4 2.2 11.0 40.7 16.5 100.1 Previous Convictins Sexual Sexual Assault (SA) S A, with Weapon Aggravated SA Invitation to Touch Sexual Interference Sexual Touching Exposure, under 14 Gross Indecency Indecent Assault (f) Buggery Indecent Assault (m) 8 50.0 1 6.3 0 0.0 3 18.8 1 6.3 3 18.8 0 0.0 0 0.0 0 0.0 0 0.0 _Q _OQ 16 100.2 31 60.8 1 2.0 1 2.0 0 0.0 2 3.9 0 0.0 1 2.0 3 5.9 7 13.7 1 2.0 _4 JL8 51 100.1 Note: A molester may have multiple charges for their current offense. 89 Table 5 Descriptive Analysis of Adult Child Molesters Variable Frequency Percentage Marital Status: Single 32 32.0 Common-law 13 13.0 Married 34 34.0 Separated 12 12.0 Divorced 7 7.0 Unknown _2 2.0 100 100.0 No. of Biological Children: 0 43 43.0 1-4 52 52.0 5-8 _5 _^0 100 100.0 Employment: Unemployed 25 25.5 Part-time employment 16 16.3 Full-time employment 53 54.1 Retired _4 _AA 98 100.0 Employment Level: Professional 14 14.0 Skilled 34 34.0 Semi- or unskilled 52 52J) 100 100.0 90 Table 6 Reliabilities for the Descriptive Analysis Variables Inter-rater Reliabilities Variables Adolescent Adult % K % K Age (birthdate) 95 .90 100 1.00 Marital Status - 80 .75 No. of Children - 100 1.00 Previous Employment 80 .69 -Current Employment - 76 .60 Spearman Spearman Grade .96 .99 % % Current Charges Non-Sexual 85 95 Sexual 100 85 Previous Conviction Non-Sexual 80 100 Sexual 85 95 Note: K indicates a Kappa Coefficient 91 6.2 Cluster Analysis The potential causal variables selected for analysis from the adolescent and adult child molester files were based on practical considerations. Variables were included in the analysis when 80% of the files contained the information needed. In the event that less than 20% of the data was missing for a particular variable, the values were replaced with the mean for continuous data and considered not present for nominal data (i.e., a procedure followed by Forensic Psychiatric Services). Variables were discarded from the analysis if professionals conducting the assessments did not address the topic directly and consistently (e.g., whether the offender had close relationships), or were measured inconsistently over time (e.g., using different types of tests) for 40% or more of the sample. In the end, 23 of 40 potential causal variables were acceptable from the adolescent data (See Tables 7a and 7b), and 24 of 27 from the adult data (See Tables 7a and 7c). Thirteen of the variables were identical for the adolescent and adult samples (See Table 7a). As mentioned previously, twenty percent of the files were scored again by a different coder for each of the samples to test the reliability of the system. The files were randomly selected and independently coded. Reliabilities for all of the potential causal variables used in the cluster analysis are listed in Table 8. Percent agreements and kappa coefficients (i.e., percent agreement minus chance) (Cohen, 1960) were calculated for both the adolescent and adult child molesters' data with two exceptions. First, reliabilities were not calculated for the adolescent questionnaires (i.e., Cognitive Scale and UCLA 92 Empathy Scale), since the data was acquired from another study currently in progress at Youth Court Services. Second, the adult PCL-R total scores were calculated using intra-class (ICCi) and inter-class (ICC2) correlations. Reliabilities for the single ratings (ICCi=.90), as well as the double ratings (ICC2=95) were high. For both the adolescent and adult samples the kappa coefficients ranged from fair to excellent with the majority of the coefficients rated as excellent (Cicchetti & Sparrow, 1981). All of the potential causal variables used in the cluster analysis were analyzed to determine whether they contributed to the five-cluster solutions that were obtained for the adolescent and adult samples. The nominal variables were analyzed using a Chi Square test, and ordinal data were analyzed using Kruskal-Wallis tests. Kruskal-Wallis tests were also used to analyze interval data when an Anova test was inappropriate. Anova tests were considered inappropriate when the variances were not equal across clusters with unequal sample sizes (Howell, 1982). The data could not be transformed, since the means varied independently of the variances (Kirk, 1968). Using Kruskal-Wallis tests appeared to be the best solution to test the variables across clusters. As a consequence of the types of data used, post hoc analyses were not conducted. The variables that were significantly different across clusters for each sample are shown in Tables 7a, 7b, and 7c. "There is no generally acceptable empirical rule for choosing the correct number of clusters" (Luke et al., 1991, p. 158). Rather unique combinations of rules and interpretations of the data are used. For the adolescent sample the five-cluster solution was an obvious choice because the groups were easily distinguishable. After this point the clusters split-off into smaller groups (i.e., less than 10 cases per cluster). For the adult data the five-cluster solution was less distinguishable. Nevertheless, using fewer than five 93 clusters did not adequately divide up the sample (i.e., 65 cases in one cluster). Using more than five clusters resulted in the splitting-off of numerous groups (i.e., less than 10 cases per cluster). Overall, the five-cluster solutions for the adolescent and adult samples appeared to be homogeneous representations of the data. To examine the stability of the cluster solutions Luke, Rappaport, and Seidman (1991) recommended using a split-half test. Half of the child molesters' data was randomly selected, and re-analyzed using the same clustering technique as previously described. For the adolescent sample a four-cluster, rather than five-cluster, solution emerged. Clusters 1 (Stable), 3 (Disturbed Conduct), 4 (Pedophilic Tendencies), and 5 (Dysfunctional Families) appeared stable. However, Cluster 2 (Pedophilic) merged with other clusters. Cluster 2's lack of stability may be a consequence of the axis I diagnosis (e.g., pedophilia) variable not being randomly selected for the split-half test. Exclusion of an axis I diagnosis may also explain the reason for Cluster 1 (Mentally Unstable) merging with Cluster 2 (Mentally Stable) in the adult sample. Clusters 3 (Brain Damage/Head Injuries), 4 (Psychopathic Tendencies), and 5 (Dysfunctional Families) were easily distinguishable using the split-half test. Apparently these four clusters reflect stable psychological motivations for engaging in sexual activities with children. Each of the five-cluster solutions are discussed separately for the adolescent (See Figure 2) and adult (See Figure 3) samples. For a visual depiction of each cluster, the means of each potential causal variable were ranked on a scale from 1 to 5 (See Figures 4a-e and 5a-e: Legends are positioned after each figure set). The higher rankings indicate patterns of dysfunction, compared to lower rankings. Fieure 2. Tree of Adolescent Child Molester Cases 94 75 cases 28 cases 25 cases Cluster 1 47 cases Cluster 2 13 cases Cluster 3 15 cases Cluster 4 10 cases Cluster 5 15 cases 95 Figure 3. Tree of Adult Child Molester Cases 79 cases 65 cases 21 cases Cluster 1 25 cases Cluster 2 40 cases Cluster 3 14 cases Cluster 4 10 cases Cluster 5 11 cases 96 Table 7a Potential Causal Variables for the Five-Cluster Solution: Adolescent and Adult Molesters No. Variables Adolescent X 2 Adult X 2 1 Sexually abused as a child ^ 19.52** 1.40 2 Emotionally abused as a childc 22.90*** 12.06 3 Physically abused as a childc 15.02** 20.17*** 4 Neglected as a childc 11.92 25.23*** 5 Socialization with peers 9.88 5.91 6 Responsibility for offense 1.91 15.33** 7 Brain damage/head injuryc 11.65 55.25*** 8 Primary diagnoses/DSM, Axis I c 14.73** .26 9 Secondary diagnoses/DSM, Axis II c 38.56*** 6.07 10 Other diagnoses/DSM, Axis III0 30.64*** 10.37 11 Biological parents, alcohol involvement 13.30** 20.74*** 12 Biological parents, drug involvement ^ 6.86 18.37*** 13 Use of pornography 1.03 9.92 **g<01 ***p<001 Note: The error rate was not tested, since this is a descriptive analysis. KW indicates a Kruskal-Wallis Test C indicates a Chi-Square Test For all tests df= 4 97 Table 7b Potential Causal Variables for the Five-Cluster Solution: Adolescent Molesters No. Variables Adolescent X 2 F 14 Childhood diagnoses 10.46 -Penile plethysmograph, 15 Male stimuli 62.23*** -16 Female stimuli 68.17*** -17 Non-coerced stimuli ^ 69 95*** -18 Coerced stimuli ^ 66.96*** -19 Post-Pubescent stimuli ^ 60.38*** -20 Pre-Pubescent stimuli 67.30*** 21 Child stimuli ^ 62.70*** -22 Cognition Scale _ 2.75 23 UCLA Empathy Scale - 3.33** **D<01 ***D<001 Note: The error rate was not tested, since this is a descriptive analysis. KW indicates a Kruskal-Wallis Test F indicates an analysis of variance-Anova For all tests df = 4 Table 7c Potential Causal Variables for the Five-Cluster Solution: Adult Molesters No. Variables Adult X 2 14 Sexual activities 1.77 15 Sexual problems 4.04 16 Masturbation0 18.87*** 17 Close relationships1™' 2.61 18 Sexual Aggression Model k w 17.69*** 19 Intimate relationships k w 18.64*** 20 Concrete reasoningc 18.21*** 21 Intelligence 8.50 22 Mental health involvementkw 19.52*** 23 Transientkw 24 95*** 24 PCL-R total score k w 33.79*** **p_<01 ***p_<001 Note: The error rate was not tested, since this is a descriptive analysis. KW indicates a Kruskal-Wallis Test C indicates a Chi-Square Test For all tests df= 4 99 Table 8 Reliabilities for Potential Causal Variables for the Adolescent and Adult Molesters' Data Inter-rater Reliabilities Psychometric Variables Adolescent Adult properties of % K % K measurement Sexually abused as a child 100 1.00 95 .90 SR Emotionally abused as a child 75 .47 100 1.00 SR Physically abused as a child 90 .78 94 .87 SR Neglected as a child 90 .78 90 .78 SR Socialization 85 .70 80 .62 SR Responsibility for offense 90 .83 95 .93 SR Brain damage/head injury 85 .63 95 .78 SR DSM, Axis I 86 .75 90 .71 UK DSM, Axis II 88 .67 97 .73 UK DSM, Axis III 93 .75 97 .90 UK Biological parents, alcohol 95 .92 100 1.00 SR Biological parents, drugs 90 .78 100 1.00 SR Use of pornography 85 .77 90 .76 SR Childhood diagnoses 85 .78 - _ UK Penile plethysmograph, 80 .42 - - UK Cognitive Scale - - - - .76 UCLA Empathy Scale - - - - .84 Sexual activities - - 80 .52 SR Sexual problems - - 85 .57 SR Masturbation - - 95 .86 SR Close relationships - - 90 .85 SR Sexual Aggression Model - - 87 .77 SR Intimate relationships - - 90 .79 SR Concrete reasoning - - 100 1.00 SR Intelligence - - 100 1.00 SR Mental health involvement - - 80 .61 UK Transient - - 100 1.00 SR ICd ICC 2 PCL-R total score - - .90 .95 .85-.95 Note: K indicates a Kappa Coefficient SR= Self-Report UK = Unknown 100 6.2.1 Adolescent: Five-Cluster Solution The adolescent child molesters were categorized into five-clusters referred to as la to le, while the clusters for the adult child molesters are referred to as 2a to 2e. For each cluster in both samples only statistically significant differences will be discussed. Cluster la comprised the largest group (47.0%) of adolescent child molesters who were labeled "Stable" (See Figure 4a). These adolescents ranked lower than the average child molester for experiencing incidents of emotional and physical abuse as a child. As well, they ranked low on the number of psychiatric diagnoses that were given at the time of assessment for their current offense. The rest of the adolescents' scores in Cluster la fell within the average range. Overall, these adolescents appear relatively stable in comparison to their adolescent counterparts. The second cluster, comprising 13.0% of the adolescent sample, were labeled "Pedophilic" (See Figure 4b). All of the adolescents in this cluster, as well as the adolescents of Cluster le, had a primary diagnosis. The majority of these diagnoses were for Pedophilia as defined in the DSM Manuals under section 302.2. Unfortunately, their scores on the penile plethysmograph did not reflect this diagnosis. Instead, their scores indicated low levels of arousal to all the sexual stimuli presented. These adolescents also ranked low on the UCLA Empathy Scale. (Adolescents with a low ranking may exhibit empathy and moralistic behaviors, but seldom aggressiveness.) The adolescents in this cluster ranked low on experiencing sexual abuse as children and having biological parents who drank heavily. 101 Adolescent child molesters belonging in Cluster lc were labeled "Disturbed Conduct" (See Figure 4c). This cluster, comprising 15.0% of the sample, received the largest number of secondary and other types of diagnoses. The diagnoses frequently described disturbances of conduct, disturbances of conduct with developmental delays, or developmental delays. These diagnoses coincided with the most frequent primary diagnosis, Disruptive Behavior Disorder Not Otherwise Specified (section 312.9 of the DSM III-R Manual). Since there were reports that their biological parents were heavy drinkers, some of these adolescents may have been suffering from FAE/FAS: As a matter of fact, their parents ranked higher on this item than parents of adolescents from any other cluster. These adolescents also ranked high on variables relating to emotional, physical, and sexual abuse as a child. The findings suggest that these adolescents were raised in highly disturbed families. Their scores on the UCLA Empathy Scale suggest they experience problems being empathic to others. Like the adolescents in Cluster lb, their levels of sexual arousal using the penile plethysmograph were low. The remarkable features of Cluster Id (10.0) are the consistently low rankings of these adolescent child molesters on childhood experiences of emotional, physical, and sexual abuse (See Figure 4d). Even their biological parents were ranked low for abuse of alcohol. Unlike the adolescents from Cluster lc, these adolescents do not appear to be suffering from a disturbed upbringing. Nor do these adolescents appear to be suffering from a psychiatric disorder: only half of the adolescents received a primary, secondary, or other type of diagnosis. Of those who received a diagnosis, 3 adolescents were considered pedophilic and 2 had conduct disorders. Adolescents' in this cluster were ranked high, according to their scores on the penile plethysmograph (i.e., highly aroused 102 by the sexual stimuli) and UCLA Empathy Scale (i.e., lacking affective empathy). Overall, the profile suggests that these adolescents have "Psychopathic Tendencies." In strong contrast to Cluster Id, adolescent child molesters assigned to Cluster le ranked high on emotional, physical, and sexual abuse during childhood (See Figure 4e). Another variable indicating that these adolescents were from highly dysfunctional families is the elevated ranking of their biological parents as heavy drinkers. These adolescents were frequently given a primary, secondary, or another type of psychiatric diagnosis. The majority of their diagnoses were for Pedophilia or a Disruptive Behavior Disorder Not Otherwise Specified. These adolescents' ranked a 5 for sexual responses using the penile plethysmograph. However, their scores were ranked low on the UCLA Empathy Scale, indicating affective empathic responses to others. This group of adolescents was labeled "Dysfunctional Families," and comprised 15.0% of the adolescent child molester sample. 103 Figure 4a. Cluster 1 for Adolescent Data: Stable C h iid Moles ters A dolescent C luster 1 a 1 3 5 7 9 1 1 1 3 Potentia 1 C a u s a l Variables Figure 4b. Cluster 2 for Adolescent Data: Pedophilic C h iid Moles ters Adolescent C luster 1b 104 Figure 4c. Cluster 3 for Adolescent Data: Disturbed Conduct 5 .5 Chi ld Mole s t er s A dolescent Clus te r 1 c Figure 4d. Cluster 4 for Adolescent Data: Psychopathic Tendencies C h ild M o le s te rs A dolescent C luster 1 d 1 3 5 7 9 1 1 Potential C a u s a l Variables 13 15 17 19 2 1 105 ure 4e. Cluster 5 for Adolescent Data: Dysfunctional Families C h iid M o lesters C l u s t e r 1 e 1 3 5 7 9 11 1 3 1 5 1 7 1 9 2 1 2 3 Potent ia l C a u s a l Variables 1 Sexually abused as a child 2 Emotionally abused as a child 3 Physically abused as a child 4 Neglected as a child 5 Socialization with peers 6 Responsibility for offense 7 Brain damage/head injury 8 Primary diagnoses/DSM, Axis I 9 Secondary diagnoses/DSM, Axis II 10 Other diagnoses/DSM, Axis HI 11 Biological parents, alcohol involvement 12 Biological parents, drug involvement 13 Use of pornography 14 Childhood diagnoses Penile plethysmograph, 15 Male stimuli 16 Female stimuli 17 Non-coerced stimuli 18 Coerced stimuli 19 Post-Pubescent stimuli 20 Pre-Pubescent stimuli 21 Child stimuli 22 Cognition Scale 23 UCLA Empathy Scale 106 6.2.2 Adult: Five-Cluster Solution The adult child molesters in Cluster 2a comprised 25.0% of the sample, and were labeled "Mentally Unstable" (See Figure 5a). Reports indicated that many of these adults had previous contact with mental health professionals, as they were ranked 5 on this variable. Although their diagnoses at that time were unavailable, it is known that their current diagnoses were most often for disturbances of mood (40.0%) ranging from Neurotic Depression (section 300.4, DSM IIJ-R Manual) to Affective Psychosis (section 296.80, DSM III-R Manual). Other diagnoses frequently used were Pedophilia and Adjustment Reaction (section 309, DSM III-R Manual). There were rarely reports of these adults being physically abused or neglected as children. In adulthood they did not appear to have problems with intimate relationships. Nor did they appear to suffer from brain damage, head injuries, or engage in concrete reasoning. In regards to sexuality, their attitudes towards masturbation ranged from neutral to positive. In sharp contrast to Cluster 2a, the adult child molesters in Cluster 2b appeared to be "Mentally Stable" (See Figure 5b). None of the variables that were statistically significant in separating the clusters were ranked as a 4 or 5. Actually, the adults' scores from this cluster were ranked low on experiencing neglect as a child, and having biological parents that abused alcohol or drugs. Their scores were also ranked low on negative attitudes towards masturbation and the Sexual Aggression Model. Few incidents of brain damage or head injuries were reported, and there were seldom reports of engaging in concrete reasoning. They did not appear to have transient lifestyles, or be classifiable as psychopathic. As a matter of fact, they ranked lower than any other cluster on taking 107 responsibility for their current offense. Compared to adult child molesters in other clusters they had less contact with mental health professionals prior to their current offense. The "mentally stable" molesters in this cluster comprised the largest percentage of the adult sample (40.0%). Cluster 2c encompassed 14.0% of the adult child molester sample and is labeled "Brain Damage/Head Injuries" (See Figure 5c). Seventy-one percent of the adults in this cluster were reported to have suffered from some type of brain damage or head injuries. Of those, 70% were diagnosed as having permanent damage or severe injuries. Their diagnoses included, head tremors, cerebral palsy, dyslexia, narcolepsy, spinal meningitis, and two cases of epilepsy. Adults assigned to this cluster were rated higher than other clusters of child molesters for engaging in concrete reasoning, or having difficulties with abstract thoughts. Despite these problems, they ranked low on having contact with mental health professionals. These adults had difficulty taking responsibility for their current offense, but could not be characterized as having psychopathic tendencies. For example, they did not have a transient lifestyle. Regarding their own sexuality, they conveyed negative attitudes towards masturbation, denied, or excessively engaged in this activity. These adults were rated low on the following variables: physical abuse as a child, alcoholic parents, Sexual Aggression Model, and number of intimate relationships. There is evidence that the adult child molesters in Cluster 2d have "Psychopathic Tendencies," as indicated by their ranking of 4 on their total PCL-R score (See Figure 5d). Although a ranking of 4 does not indicate the highest mean, the means of the two highest clusters were very close (Cluster 2d, M = 21.55, SD = 8.67; Cluster 2e, M = 21.58, SD = 7.24). Adults in Cluster 2d ranked slightly higher on Factor 1 of the PCL-R (Cluster 2d, 108 M = 9.40, SD = 4.17; Cluster 2e, M = 7.82, SD = 3.92) and slightly lower on Factor 2, than the adults in Cluster 2e (Cluster 2d, M = 9.13, SD = 4.86; Cluster 2e, M = 10.37, SD = 3.17). Factor 1 measures the underlying personality characteristics of the psychopath, while Factor 2 focuses on the expression of these characteristics in an antisocial lifestyle. In terms of measuring psychopathy in child molesters, the Factor 1 scores are of primary importance. Consistent with a psychopathic profile the adults in Cluster 2d lived a transient lifestyle, including be involved in a large number of intimate relationships. There were reports that they engaged in concrete reasoning or had difficulties with abstract thought, but there were few reports of brain damage or head injuries. Furthermore, they did not appear to take responsibility for their current offense. Other characteristics of this group included, being physically abused or neglected as a child, having biological parents who were involved with alcohol and drugs, high scores on the Sexual Aggression Model, and negative attitudes or denying or excessively engaging in masturbation. Adults from Cluster 2d comprised 10% of the sample. Like Cluster le in the adolescent sample, the adult child molesters in Cluster 2e appeared to come from "Dysfunctional Families" (See Figure 5e) (11%). They ranked higher than adults from other clusters on experiencing physical abuse or neglect as a child. Perhaps, as a result of the abuse or neglect, they experienced brain damage or head injuries. One or both of their biological parents were ranked high for involvement with alcohol. They ranked high on the Sexual Aggression Model. As adults, they frequently engaged in intimate relationships, and transient lifestyles. As mentioned in the previous section, their lifestyles often included antisocial behaviors, as indicated by their Factor 2 and total PCL-R scores, but they appeared to take responsibility for their current offenses. 109 110 F i g u r e 5c. C l u s t e r 3 f o r A d u l t D a t a : Brain Damage/Head Injuries C h iid Moles ters 1 3 5 7 9 11 13 15 17 Potentia 1 C a u s a l Variables F i g u r e 5d. C l u s t e r 4 f o r A d u l t D a t a : Psychopathic Tendencies C h iid M o le s te rs I l l Figure 5e. Cluster 5 for Adult Data: Dysfunctional Families C h ild M o l e i t C luster 2e 1 3 5 7 9 11 13 15 17 19 21 23 P o t e n t i a l C a u s a l V a r i a b l e s 1 Sexually abused as a child 2 Emotionally abused as a child 3 Physically abused as a child 4 Neglected as a child 5 Socialization with peers 6 Responsibility for offense 7 Brain damage/head injury 8 Primary diagnoses/DSM, Axis I 9 Secondary diagnoses/DSM, Axis II 10 Other diagnoses/DSM, Axis III 11 Biological parents, alcohol involvement 12 Biological parents, drug involvement 13 Use of pornography 14 Sexual activities 15 Sexual problems 16 Masturbation 17 Close relationships 18 Sexual Aggression Model 19 Intimate relationships 20 Concrete reasoning 21 Intelligence 22 Mental health involvement 23 Transient 24 PCL-R total score 112 6.3 Contrasting Variables To validate the five-cluster solutions for the adolescent and adult samples the clusters were contrasted using additional variables. For these samples, the contrasting variables that were used described their current offense (See Appendix C l and C2 of the coding sheets). Three criteria were used to determine which variables from the appendix to choose. First, the information should only be obtained from police reports. Second, the variables needed to be typical of this type of offense (i.e., child molestation) as described in the literature. Third, as presented in the police reports the variables should reflect upon the psychological state of the offender, rather than the victim (e.g., variables pertaining to disclosure were not used). The remaining variables described the victim, sexual act and amount of contact during the offense. Victim variables included, age of victim, relationship of victim to offender, and sex of the victim(s). When there were multiple victims the information from the police report usually focused on one particular victim. If this did not occur the information from the most recent victim was used in the analysis. However, the sex of the multiple victims was considered and labeled homosexual, heterosexual, of bisexual for the last variable. The sexual act variables included substance abuse and type of act. The type of acts considered were fondling and oral sex. The amount of contact variables focused on the frequency and duration of the sexual acts. Reliabilities for the current offense variables were fair to excellent (See Table 9). Inter-rater correlation coefficients were calculated to measure reliabilities for the "age of the victim" variable. The Pearson correlation coefficient was high for the adolescent 113 sample. This was the appropriate statistic since the distribution was normal with a mean of 6.9 (SD=3.01). In contrast, the distribution of the adult sample was bimodal with the most frequent ages of the victims being 4 and 11. Since it was a bimodal distribution, a Spearman correlation coefficient was calculated. The coefficient was high. For the remaining variables, percent agreements and Kappa coefficients were calculated (See Table 9). Percent agreements between coders ranged from 80 to 100%, while the Kappa coefficients were .57 to 1.00 (i.e., fair to excellent) (Cicchetti & Sparrow, 1981). None of the victim or sexual act variables validated the adolescent five-cluster solution (See Table 10). However, an amount of contact variable (i.e., frequency) did distinguish between the clusters using a Chi square test, X 2 (4, 100) = 10.87, p < .05. Cluster lc had the highest percentage (53.0%) of adolescents who engaged the child in sexual activities more than once during his current offense (Cluster lb, 39.0%; Cluster la, 36.0%; Cluster le, 13.0%, and Cluster Id, 0.0%). For the adult sample the victim variables did not validate the five-cluster solution; however, variables from the sexual act and amount of contact sections did validate the solution. In the sexual act section there was a statistically significant difference between clusters on substance abuse X 2 (4, 100) = 13.91, p < .01, and fondling X 2 (4, 100) = 10.93, p < .05 using a Chi Square test. It was reported that 64.0% of the offenders in Cluster 2e had been under the influence of alcohol or drugs during the offense, compared to 52.0% in Cluster 2a, 40.0% in Cluster 2d, 21.0% in Cluster 2c, and 18.0% in Cluster 2b. Regarding the fondling variable, all of the offenders in Cluster 2d were reported to have engaged in this activity (60.0%, offender to victim; 40.0%, bi-directional) compared to 96.0% in Cluster 2a (4.0%, victim to offender; 63.0% offender to victim; 33.0%, bi-114 directional) 91.0% in Cluster 2e (100.0% offender to victim), 87.0% in Cluster 2b (3.0%, victim to offender; 50.0% offender to victim; 47.0%, bi-directional), and 77.0% in Cluster 2c (18.0%, victim to offender; 73.0% offender to victim; 9.0%, bi-directional). Cluster 2d also had the highest percentage of offenders who abused a child over an extended period (or duration) of time (20.0%, days; 10.0%, mos.; 70.0%, yrs.), compared to clusters 2b (23.1%, days; 76.9%, mos; 0.0%, yrs.), 2a (28.0%, days; 72.0%, mos.; 0.0%, yrs.), 2e (45.5%, days; 18.2%, mos.; 36.4%, yrs.), and 2c (57.1%, days; 21.4%, mos.; 21.4%, yrs.). 115 Table 9 Reliabilities for the Contrasting Variables Variables Adolescent Adult Victim Variables Pearson Spearman Age of Victim .99 .97 % K % K Relationship 100 1.00 85 .80 Sex of Victim(s) 100 1.00 100 1.00 Sexual Act % K % K Substance Abuse 100 1.00 85 .57 Type of Act Fondling 80 .73 85 .57 Oral Sex 85 .78 90 .83 Amount of Contact Frequency 90 .79 89 .73 Duration 90 .84 90 .84 Note: K indicates a Kappa Coefficient 116 Table 10 Contrasting the 5-Cluster Solutions for the Adolescent and Adult Child Molesters Variable Adolescent Adult Victim Variables df F df F Age of Victim 4 .66 4 1.34 df X 2 df X 2 Relationship1™ 4 1.21 4 5.37 Sex of Victim(s)tar 8 7.16 8 7.43 Sexual Act df X 2 df X 2 Substance Abusec 4 3.66 4 13.91** Type of Sexual Act Fondling1™ 4 3.25 4 10.93* Oral Sex1™ 4 4.75 4 6.17 Amount of Contact Frequency6 4 10.87* 4 4.13 Duration0 8 5.61 8 54.61*** Note: F indicates an analysis of variance-Anova KW indicates a Kruskal-Wallis Test C indicates a Chi-Square Test r 117 Chapter 7 DISCUSSION The purpose of this project was to develop a psychological classification system for male adolescent and adult child molesters using a cluster analysis technique. The five-cluster solutions obtained are discussed, as well as attempts to validate the cluster solutions using contrasting variables related to the child molesters' offenses. The clinical and investigative implications of these findings are considered. A cautious approach is suggested when using information from this exploratory project, because the findings may or may not be replicated in the future. 7.1 A Summary of the Findings Five-cluster solutions were discovered when using the adolescent and adult child molesters' samples (see Appendix D and E). One of the five solutions for each sample did not reflect any psychological motivations for committing sexual offenses. Instead, child molesters appeared "Stable" in the adolescent sample and "Mentally Stable" in the adult sample. Perhaps these molesters would have scored abnormally on variables not assessed in the current study. This is quite possible because many of the variables suggested by the theoretical perspectives could not be coded: either there was not enough information to code these variables or the information could not be consistently found in the files. The 118 three mental functions theory data was particularly sparse when describing the cognitive and volitional states of the molester, but information on the molester's affective states prior to or during their sexual offense was completely absent from the files. Information pertinent to third force theories needs to be obtained through interview, rather than file information. The interviewer would need to address issues relating to adolescent and adult child molesters' personal interests, needs, temperament, and experiences. Molesters labeled "Pedophilic" from the adolescent sample and "Mentally Unstable" in the adult sample appear to be motivated by a sexual orientation towards children. The adolescents were frequently given a diagnosis of "pedophilia," while the adults were often diagnosed with a mood disorder and seen by mental health professionals. Using Ward, Hudson, and Marshall's (1994) reasoning a connection can be made between pedophilia and mood disorders. An adult engaging in pedophilic fantasies uses a stream of cognitions to rationalize engaging in this sexual act. During the cognitive processes of planning and committing the act they experience a decrease in positive emotions and an increase in negative emotions. Unfortunately, there are no hard data to determine the degree of accuracy of this line of reasoning. Nor has it been determined whether a disturbance in mood is a reaction to a deviant sexual preference or a reaction to societies stigmatization of individuals who engage in sexual activities with children (Langevin, Paitich, Freeman, Mann, & Handy, 1978). Interestingly, the adolescents' and adults' affective and cognitive responses to others did appear normal in other ways. The adolescents' scores on the empathy scale indicated highly empathetic responses, moral altruism, and low levels of aggression. The adults did not report problems maintaining intimate relationships. 119 The adolescents labeled "Disturbed Conduct" and adults with "Brain Damage/Head Injuries" appeared to be suffering from physiological problems, yet the two profiles are quite different in other respects. The adolescent molesters were frequently diagnosed with disturbed conduct and/or developmental delays. They came from disturbed families where the parents abused alcohol. Some of these adolescents may be experiencing FAS or FAE. Contrary to expectations, FAS or FAE adolescents did not appear to be represented in the adult sample. Perhaps adolescents who experienced these symptoms were diagnosed as mentally handicapped. If diagnosed as mentally handicapped in British Columbia, they would be strictly supervised and treated in long-term foster homes. Thus, they would have less opportunity to offend. In contrast to the adolescent sample, the adults with brain damage or head injuries did not appear to come from disturbed families. Their biological parents were not described as heavy drinkers. Instead, the majority of these adults suffered from permanent brain damage or severe brain injuries that occurred or were diagnosed in adulthood. Considering the effects of brain damage or injuries, it is important to note that all of the adult molesters with "Brain Damage/Head Injuries" committed heterosexual or homosexual offenses~none of them appeared to be bisexual. Heterosexual and homosexual male offenders, according to Langevin, Wortzman, Wright, and Handy (1989), are not only experiencing "left hemispheric language-mediated problems, but also frontal lobe rigidity and perhaps impulsiveness" (p. 177). This is in contrast to bisexual male offenders who appear to be suffering from a right hemispheric impairment. These findings were based on the Halstead-Reitnan Neuropsychological Test Battery. Unfortunately, the authors use of Computer Tomography was not successful: a bone 120 located at the base of the brain interfered with the desired X ray scans. As an alternative approach, the researchers recommended examining brain structures using a Magnetic Resonance scan, and electroencephalograph and Positron Emission Tomography during sexual arousal. Any speculations as to the etiology of sexual abnormalities related to brain structures are dependent upon technological advancements and continued exploration of its structure and functions. Prior to interpreting the data from the adolescent and adult molesters with "Psychopathic Tendencies," it is important to recall the primary source(s) of the data collected for the adolescent and adult samples. The adolescent data on family background was derived from numerous outside sources (e.g., caregivers, school personnel, social workers, etc.), while the adult data on family background was largely based on the offender's reports to professionals (e.g., probation officers, psychologists, psychiatrists, etc.) Reflecting upon the fact that the personality characteristics of a psychopath include being deceitful and manipulative, the adult data should be interpreted cautiously. For instance, very few of the adolescent molesters with psychopathic tendencies were described as emotionally, physically, or sexually abused as children. However, the information from adult molesters with psychopathic tendencies frequently contained self-reports of physical abuse and neglect as children. They also reported that their biological parents were involved with alcohol and drugs. An explanation for the discrepancies between samples is that adults who report dysfunction within their families are attempting to instill sympathy from the interviewer. Alternatively, they may simply be demonstrating the deceptive traits characteristic of psychopaths. 121 In contrast to the male child molesters with psychopathic traits, both the adolescents and adults from "Dysfunctional Families" reported abuse and neglect during their childhood, and parents who were heavy drinkers. These child molesters' behaviors appeared disruptive during adolescence and antisocial as adults. The adults were also involved in numerous intimate relationships and transient lifestyles. Despite these difficulties within a social context, the adolescents appeared to be empathetic and the adults did take responsibility for their offense(s). 7.2 Validation of the Five-Cluster Solution Although validation of the five-cluster solutions using contrasting variables for the adolescent sample was weak, it was more convincing for the adult sample. Only one variable was statistically significant in distinguishing the adolescent clusters, that is, the frequency of contact offenders had with their victims. Adolescents with "Disturbed Conduct" were ranked highest on the amount of times (i.e., regularly or repeatedly) they abused their victims. It is suspected that many of these adolescents suffered from FAS or FAE, although other explanations are plausible. A number of typical behaviors associated with FAS/FAE may have contributed to this increase in the frequency of contact, including, a lack of inhibition, poor social judgment, excessive need for physical contact, and an inability to appreciate the consequences of their behaviors (Short & Hess, 1995; Streissguth, 1994). Perhaps the reason other clusters representing the underlying psychological motivations for these adolescents to offend were undifferentiated in terms of other contrasting variables is that these adolescents had not established consistent offense 122 patterns. Consider the implications of their physiological and psychological development in relationship to their expression of sexuality; specifically, adolescents' genitalia are maturing while they are exploring their own identity in relationship to others (Erikson, 1968). This is a period of confusion where "criminal incidents do not have the same fatal significance which they may have at other ages" (Erikson, 1968, p. 132). For the adult sample of child molesters none of the victim variables served as statistically significant ways of distinguishing the clusters; however, the sexual act (i.e., substance abuse and fondling) and amount of contact variables (i.e., duration) validated the five-cluster solution. The adult molesters from "Dysfunctional Families" had the highest percentage of substance abuse. These molesters were raised in families with biological parents who were heavy drinkers; therefore, substance abuse among these molesters might have been anticipated. They were also more likely to have been under the influence of alcohol or drugs when offending. Finkelhor (1986) believed that alcohol acts as a disinhibitor for acting out sexual impulses in males who are interested in children. Regarding the variable fondling, all of the molesters with "Psychopathic Tendencies" were reported to have fondled their victims. In half of these cases, fondling progressed to digitally penetrating the victim. The direction of fondling involved the offender touching the victim or reciprocal touching between the offender and victim. None of the cases involved only the victim fondling the offender. Despite the direction of these sexual activities, the child molesters with "Psychopathic Tendencies" frequently blamed their victim, others, or the situation for the offense(s). Only one molester blamed himself for not thinking through the situation—in the future he'd kill his victims. 123 It is intriguing that in neither sample did the victim variables serve to distinguish the clusters, like the sexual act and amount of contact variables. This suggests that motivations were not manifested through their choice of a victim in terms of age, gender, or their relationship to the victim. This finding is contrary to expectations implied by previously discussed classification systems, such as Finkelhor and William's (1994) system. They distinguished incest offenders by the age of their daughters at the time they first experienced being strongly aroused by them. If their daughter was under 10 years of age, they were labeled "Sexually Preoccupied." Fathers who were aroused by their daughters when they were 10 years of age or over were labeled "Adolescent Regressive." Although the authors reported some clustering by type, approximately one-third of the fathers in the sample were considered to be discordant types. Coding victim variables by the age of the victim when the offender was first aroused may be a better predictor than the victim's age at the time of the offense as used in this study. The former measure may reflect the offender's psychological state, while the latter measure could be affected by situational factors. Unfortunately, information on the child's age when the offender first felt sexual arousal to a minor was not available in the adolescent or adult files for the present study. Even though there appears to be some validation for the five-cluster solutions, it cannot be overemphasized that cluster analysis is an exploratory technique. These solutions should be viewed as working hypotheses. The investigative and clinical implications of these findings should be applied cautiously, if at all, until research progresses in this area. 124 7.3 Implications of the Five-Cluster Solutions 7.3.1 Clinical Implications The underlying psychological motivations for the various types of adolescent and adult child molesters appear to be very similar. Such heterogeneous groups require unique intervention strategies based on their motivations, rather than simply implementing treatment strategies according to settings (i.e., prison, community, or psychiatric hospitals) or distinctive forms of diversity (i.e., occupations or ethnic groups) (Marshall, Fernandez, Hudson, & Ward, 1998). The following is an attempt to apply current treatment strategies to four of the five-cluster solutions derived from the adolescent and adult samples. (The "Stable" and "Mentally Stable" clusters will not be discussed, because there is insufficient information with respect to their psychological motivations). As mentioned previously, the child molesters in the "Mentally Unstable" and "Pedophilic" clusters were frequently diagnosed as having a sexual orientation towards children. One treatment approach for molesters with a sexual orientation to children involves three phases (Gillies, Hashmall, Hilton, & Webster, 1992). The first phase consists of 90-minute psychoeducational meetings held weekly with the child molesters over a 3-month period. The meetings focus on the sequence of events leading up to their offenses, including affective states, fantasies, planning, and carrying-out the plan. Strong affective states can lead to deviant sexual fantasies, therefore, pedophiles learn to relax while under stress. Deviant fantasies are strengthened when masturbating to images of children, and using child pornography. Both activities must be avoided. Finally, the molester must be aware that plans can be carried-out by "seemingly unimportant 125 decisions," such as walking through a park where children are playing. The second phase continues to take a psychoeducational approach, but the purpose of the group meetings is to be supportive of individual members. The third phase continues this same approach at a deeper and more personal level. In addition, this psychoeducational approach provides other forms of information (e.g., legal, medical, sexual). Throughout all phases of treatment the staff encourage a non-confrontational approach to emphasize that they are against the offender's actions, but supportive of the offender as a person. Evaluations of this approach are currently being undertaken (Gillies, Hashmall, Hilton, & Webster, 1992). Gillies et al.'s (1992) program for pedophiles may also be beneficial to child molesters with deficits in abstract reasoning due to "Brain Damage/Head Injuries." Males with borderline intelligence were not excluded from Gillies et al.'s (1992) program and apparently benefited from participating in the groups. Another study by Lindsay, Neilson, Morrison, and Smith (1998) described a treatment program in the community, specifically for male child molesters with learning disabilities. Their cognitive therapy program focused on changing attitudes towards offending behaviors in two ways. First, child molesters were assigned to groups based on the gender of their victim. Each group was lead by male and female therapists who focused on offending behaviors in general to elicit the molesters' attitudes towards child sexual abuse. Second, child molesters were shown slides and pictures of interactions between children and men (e.g., a girl sitting on a man's lap while he reads her a story) to examine their cognitions. A cognition that supports offending behaviors is challenged (e.g., the child is being sexually provocative, or leading the man on) while a socially acceptable cognition is supported (e.g., the child wants to be read a story). According to the authors it is not uncommon for the molester to become 126 angry at this point because he is forced to re-examine his own belief system and take responsibility for his behaviors. A program evaluation suggested that this approach was successful when the offenders were followed up at least 4-years later. Unfortunately, cognitive behavioral approaches may be less successful with molesters with severe cognitive deficits (e.g., FAS children). For these cases, Coleman and Haaven (1998) recommend "utilizing environmental contingencies and operant-based techniques such as response cost, time out, and differential reinforcement of other competing behaviors" (p. 273). Less success is anticipated with males exhibiting "Psychopathic Tendencies." These males have a need for novelty and excitement, while lacking the ability to tolerate routines (Hare, 1993). Treatment programs may have little effect on this type of offender (Ogloff, Wong, & Greenwood, 1990), or even make him worse (Rice, Harris, & Cormier, 1992). For example, group therapy sessions are not effective for males with psychopathic tendencies as they tend to impose their own agenda on the group (Hare, 1993). As a consequence, alternative treatment plans were developed by Hare and a panel of experts (Hare, 1996). They designed a cognitive-behavioral program to teach offenders pro-social ways of meeting their own desires or needs. This program called for tight controls or supervision while being incarcerated and after release into the community. "Dysfunctional Families" is the label applied to adolescent and adult child molesters who came from abusive or neglectful families, and the adult child molesters appeared to be creating their own dysfunctional families by engaging in numerous intimate relationships, transient lifestyles, and antisocial behaviors. Even though Giarttetto's (1978) Child Sexual Abuse Treatment Program (CSATP) was designed for father-127 daughter incest cases, his use of humanistic principles as premises for treatment may be applicable to this type of child molester. The major premises of CSATP are that individuals with high self-concepts will engage in healthy marital relationships, and healthy marital relationships are the means to building homeostasis in the family. Treatment procedures for this type of child molesters need to focus on individual, marital, family, and group counseling when applicable. Adhering to humanistic principles, the atmosphere during the counseling sessions is compassionate and supportive. An evaluation of the CSATP program reported a zero recidivism rate for over 500 families that received at least ten hours of treatment and were formally terminated from the program. Other treatment strategies frequently employed can be described as "individualistic," because strategies are adjusted to the needs of each molester. This approach is often used with adolescents (Groth, Hobson, Lucey, and Pierre, 1981; Worling, 1998). The draw back to individualistic approaches is the in-depth and time-consuming assessment period. For example, Groth et al.'s strategy was to assess the offender's experiences of adversities in early life, adjustment problems, chronic emotional problems, difficulties in social relationships, defects in self-regard, sexual concerns, attitudes and values, maladaptive defenses, prognosis, and other compounding problems. Even after problem areas are identified, the therapist needs to be aware of similar patterns in other molesters to anticipate their responses to therapeutic interventions. 7.3.2 Investigative Implications Lanning's (1992) system, designed for investigators rather than clinicians, is divided into preferential (i.e., fixated) and situational (i.e., regressed) molesters with each 128 type of molester divided into 3 or 4 sub-types according to their patterns of behavior. According to Lanning, the preferential child molesters (i.e., seduction, introverted, and sadistic) typically have the largest number of victims, but comprise the smallest group of offenders. Consequently, the adolescent and adult samples did not appear to contain any introverted or sadistic child molesters. The adolescent molesters labeled "Pedophilic" and adult molesters labeled "Mentally Unstable" could be classified under the subtitle seduction. Seductive pedophiles identify with children, and know how to talk and listen to them. They focus in on particular types of victims based on age and gender. Frequently they have collections of child pornography or erotica. Our research suggests that these males had previous contact with mental health professionals for psychiatric or psychological problems. Despite mental health problems, these offenders are capable of presenting themselves as empathetic or concerned citizens (e.g., caring). Lanning's .. categories for the situational child molester were comparable to clusters labeled "Stable," "Disturbed Conduct," "Psychopathic Tendencies," and "Dysfunctional Families" for the adolescent sample, and "Mentally Stable," "Brain Damage/Head Injuries," "Psychopathic Tendencies," and "Dysfunctional Families" for the adult sample. Situational child molesters do not necessarily have a preference for children, but engage in sexual activities for a variety of reasons. The reasons are revealed through the following sub-types. The "Mentally Stable" adult child molesters are similar to Lannings' regressed category. Lanning described these males as looking for an easily available sexual substitute for an adult partner to coerce into sexual activities. The substitute may even be one of their children. No clear pattern for collecting pornography has been recorded. This study indicated that these men appear mentally stable, but may be experiencing stress 129 within their intimate relationships. The adolescents who appeared relatively "Stable" compared to their adolescent counterparts may also be regressed offenders. Considering that both these clusters comprised a large percentage of the samples (i.e., adolescent 47% and adult 40%), more information regarding their psychological motivations is needed. Molesters labeled "Disturbed Conduct" from the adolescent sample and "Brain Damage/Head Injuries" from the adult sample fell under Lannings' inadequate category. According to Lanning, physiological handicaps contribute to these males feeling like social misfits, so they exploit non-threatening victims (e.g., children or elderly) out of feelings of insecurity or curiosity. Their pornographic collections are more apt to be pictures of adults, rather than children. Data from the present study suggests that these males are encountering problems with sexual performance: they may be experiencing low arousal levels, or behavioral problems related to masturbation. The morally indiscriminate molester, according to Lanning, uses and abuses everyone in his path. This use and abuse of others may be of a sexual or non-sexual context, often involving lying, cheating, and stealing. When the sexual abuse of a child occurs, it is simply a reaction to an available opportunity, "Why not?" (Lanning, 1987, p. 6). This child molesters' approach to committing sexual offenses often involves force, luring, or the manipulation of acquaintances or strangers. Often these molesters' collection of adult pornography included detective magazines and sadomasochistic acts. Child pornography collections may include pre-pubescent and pubescent children. It is remarkable how similar Lanning's description of the morally indiscriminate molester is to Hare's (1993) description of psychopaths. "Lying, deceiving, and manipulation are natural talents for psychopaths" (Hare, p. 46), and "Psychopaths view people as little more than 130 objects to be used for their own gratification" (p. 44). Thus, the molesters labeled as exhibiting "Psychopathic Tendencies" in the adolescent and adult samples can be labeled as morally indiscriminant using Lanning's system. In the present study, child molesters with psychopathic tendencies are characterized by concrete, rather than abstract mental orientations. They would be found in relatively unskilled jobs (e.g., factory workers) or engaging in illegal activities (e.g., theft). It is interesting to speculate whether Lanning's sexually indiscriminate molesters have the same behavior patterns as the molesters from "Dysfunctional Families." The sexually indiscriminate males were described as sexual experimenters without any real preference for children as sexual partners. When they are sexually involved with children they engage them in previously existing activities (e.g., group sex). They are likely to have pornographic collections containing a wide variety of sexual materials. These molesters are discriminating in other types of behaviors (i.e., non-sexual) they engage in. The adolescent molesters from "Dysfunctional Families" appeared sexually indiscriminate in terms of their high responses on the penile plethysmograph. The adults from "Dysfunctional Families" engaged in a number of intimate relationships, also suggesting a sexually indiscriminate response. Contrary to expectations, the adults from "Dysfunctional Families" engaged in antisocial acts, and lead transient lifestyles. Thus, the distinction between being discriminating in daily life and indiscriminate sexually was not easily distinguishable, as implied by Lanning's description of this category. Despite some discrepancies, the investigative implications of Lanning's system coincided remarkably well with the implications of the five-cluster solutions from the adolescent and adult samples. This finding was completely unanticipated. Lanning's 131 categories are not psychologically based, nor did Lanning intend for this system to be used by mental health professionals. It was intended to be an investigative tool for law enforcement officers. One final word of caution in the investigation of child sexual abuse cases. When the FBI Behavioral Science Unit applied Lanning's system to offenders who had murdered children, they found the situational molesters were the most dangerous—particularly the males classified as morally indiscriminate or inadequate. If Lanning's morally indiscriminate molesters are actually molesters with psychopathic tendencies, warning of the dangerousness of psychopathic sex offenders should be carefully considered (Hare, 1996). Unfortunately, there is a lack of information available on the dangerousness of inadequate molesters, or molesters suffering from "Brain Damaged/Head Injuries." 7.4 Limitations and Future Directions of Research A limitation of this study is that the five-cluster solutions for the adolescent and adult child molester samples were only based on data from the referrals to Forensic Psychiatric Services5. The sample of adolescents were assessed or treated through Youth Court Services which provides services to open and secure centers for youth in the lower mainland, and to youth throughout the province by the traveling clinic. The adult sample consisted of male child molesters who were being assessed, treated, or supervised through a Community Services clinic in Vancouver. The Vancouver clinic is the largest adult clinic in the province. The clinic serves high risk and Not Criminally Responsible-Mentally 132 Disordered (NCR-MD) clients residing in the community. The clusters derived from the adolescent and adult samples may not generalize to other settings, or additional clusters may be required. Also, the number and type of clusters may vary according to the methodology utilized. Archival file information, collected during this study, contained extensive background information from various sources (e.g., parents, school, police). Nevertheless, the file information did not address all of the essential features of each theoretical position. Future research could focus on essential theoretical features in more depth by conducting interviews with the child molesters, and using information contained in their files. The large number of variables obtained could be reduced using a discriminant analysis statistic, prior to the cluster analysis technique. The importance of a large sample size cannot be overly stressed, as this was a limitation of the current project. A large sample is needed to detect a small effect size with a high degree of certainty when validating the results obtained from the cluster analysis. Other ways of extending this psychological classification system involve a more in-depth analysis of the five-cluster solutions. Four of the five clusters for both samples apparently reflect stable psychological motivations for engaging in sexual activities with children. Only the clusters based on psychiatric diagnoses of pedophilia were unstable when these diagnoses were eliminated from the analysis. Each cluster appears to represent a different type of child molesters' motivations that could be examined from various theoretical perspectives. Not all of the theories would be expected to apply to a particular type of molester. Nevertheless, a multi-theoretical approach is advantageous as it 5 While undertaking this research project Youth Court Services changed jurisdictions by moving from the 133 represents child molesters as complex beings with multiple response patterns across their life-span (Staats, 1983). Ministry of Health to the Ministry for Children and Families. 134 REFERENCES Abel, G.G., Gore, D.K., Holland, C.L., Camp, N , Becker, J.V., & Rathner, J. (1989). The measurement of the cognitive distortions of child molesters. Annals of Sex Research. 2. 135-153. Abel, G., Mittelman, M., & Becker, J. (1985). Sexual offenders: Results of assessment and recommendations for treatment. In M. Ben-Aron, S. Hucker, & C. Webster (Eds.), Clinical criminology: The assessment and treatment of criminal behaviour (pp. 191-205). Toronto, ON: M&M. American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Barbaree, H.E., Seto, M.C., Serin, R.C., Amos, NX., & Preston, D.L. (1994). Comparisons between sexual and nonsexual rapist subtypes: Sexual arousal to rape, offense precursors, and offense characteristics. Criminal Justice and Behaviour. 21(1). 95-114. Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective. Journal of Social and Personal Relationships. 7, 147-178. Bartholomew, K., & Horowitz, L.M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology. 61(2), 226-244. 135 Becker, J.V., Hunter, J.A., Stein, R.M., & Kaplan, M.S. (1989). Factors associated with erection in adolescent sex offenders. Journal of Psychopathology and Behavioral Assessment. U(4), 353-362 Becker, J.V., & Quinsey. (1993). Assessing suspected child molesters. Child Abuse and Neglect. 17, 169-174. Blashfield, R.K., & Draguns, J.G. (1976). Toward a taxonomy of psychopathology: The purpose of psychiatric classification. British Journal of Psychiatry. 129. 574-583. Bowlby, J. (1969). Attachment and loss. Vol.1. Attachment. New York, NY: Basic Books. Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. New York, NY: Basic Books. Bradford, J.M.W., & Gratzer, T.G. (1995). A treatment for impulse control disorders and paraphilia: A case report. Canadian Journal of Psychiatry. 40. 4-5. Bradford, J.M.W., Bloomberg, D, Bourget, D. (1988). The heterogeneity/homogeneity of pedophilia. Psychiatric Journal of the University of Ottawa. 14(4), 217-226. Brandt, R.B. (1959). Ethical theory: The problems of normative and critical ethics. Englewood, NJ: Prentice-Hall. Brown S. (1994). Sexual aggression and psychopathy among convicted rapists: Static. dynamic, and motivational precursors. Unpublished master's thesis, Carleton University, Ottawa, Ontario, Canada. Caplan, G.M. (1982, January). Sexual exploitation of children: The conspiracy of silence. Police Magazine. 46-51. Carlson, V., Cicchetti, D., Barnett, D., & Braunwald, K. (1989). Disorganized/disoriented attachment relationships in maltreated infants. Developmental Psychology. 25(4), 525-531. Cicchetti, D.V., & Sparrow, S.A. (1981). Developing criteria for establishing interrater reliability of specific items: Applications to assessment of adaptive behavior. American Journal of Mental Deficiency. 86(2). 127-137. Cleckley (1976). The mask of sanity (5th ed.). St. Louis, MO: Mosby. Cohen J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement. 20(1), 37-46. Cohen, M., Seghorn, T.E., & Calmas, W. (1969). Sociometric study of the sex offender. Journal of Abnormal Psychology. 74(2), 249-255. Coleman, E., & Haaven, J. (1998). Adult intellectually disabled sexual offenders. In W.L. Marshall, Y.M. Fernandez, S.M. Hudson, & T. Ward (Eds.), Sourcebook of treatment programs for sexual offenders (pp. 273-285). New York, NY: Plenum Press. Coleman, I. (1982). Family intimacy and chemical abuse: The connection. Journal of Psychoactive Drugs. 14, (1-2), 153-158. Coles, E.M. (1982). Clinical psychopathology. Boston, MA: Routledge & Kegan Paul. Collins, N.L. & Read, S.J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology. 58(4). 644-663. 137 Conran, T., Neugebauer, A., & Wrzesinski-Conran, M. (1989). A case of a "sex offender": Clinical reflections. Journal of Strategic and Systemic Therapies. 8(4), 33-42. Conte, J.R. (1991). The nature of sexual offenses against children. In CR. Hollin & K. Howells (Eds.), Clinical approaches to sex offenders and their victims (pp. 11-34). New York, NY: John Wiley & Sons. Correctional Service Canada (1991, August). Basic facts about Corrections in Canada 1991. Ottawa: Creative Services. Delin, B. (1978). The sex offender. Boston, MA: Beacon. Duncan, P., & Hern, A. (1980, Fall). Patterns in sexual offenses. Police Journal. 11-13. Erikson, Erik H. (1968). Identity youth and crisis. New York, NY: W.W. Norton & Company Inc. Erickson, W.D., Walbek, N.H., & Seely, R.K. (1987). The life histories and psychological profiles of 59 incestuous stepfathers. Bulletin of the American Academy of Psychiatry and the Law. 15, 349-357. Erickson, W.D., Walbek, N.H., & Seely, R.K. (1988). Behavior patterns of child molesters. Archives of Sexual Behavior. 17, 77-86. Everitt, B. (1980). Cluster analysis. New York, NY: Halstead Press. Finkelhor, D. & Araji, S. (1986). Explanations of pedophilia: A four factor model. The Journal of Sex Research. 22(2), 145-161. Finkelhor, D., & Williams, L.M. (1994, December). Toward a typology of incestuous fathers. Durham, NH: University of New Hampshire. Flor-Henry, P., Lang, R.A., Koles, Z.J., & Frenzel, R.R. (1991). Quantitative EEF studies of Pedophilia. International Journal of Psychophysiology. 10. 253-258. Freud, S. (1946). Totem and taboo: Resemblances between the psychic lives of savages and neurotics (A.A. Brill, Trans.). New York, NY: Vintage. (Original work published 1918). Freud, S. (1962). Three essays on the theory of sexuality (J. Strachey, Trans.). New York, NY: Avon. (Original work published 1915). Freud, S. (1964). The standard edition of the complete psychological works of Sigmund Freud. (J. Strachey, Trans.). London: Hogarth. (Original work published 1893-1899). Freund, K., Heasman, G., Racansky, I.G., & Glancy, G. (1984). Pedophilia and heterosexuality vs. Homosexuality. Journal of Sex & Marital Therapy. 10(3), 193-200. Freund, K., & Kuban, M. (1994). The basis of the abused abuser theory of pedophilia: A further elaboration on an earlier study. Archives of Sexual Behavior. 23(5), 553-563. Freund, K., & Watson, R.J. (1992). The proportions of heterosexual and homosexual pedophiles among sex offenders against children: An exploratory study. Journal of Sex & Marital Therapy. 18(1), 34-43. Freund, K., Watson, R., & Dickey, R. (1990). Does sexual abuse in childhood cause pedophilia: An exploratory study. Archives of Sexual Behavior. 19(6), 557-568. Gaffney, G.R., & Berlin, F.S. (1984). Is there hypothalamic-pituitary-gonadal dysfunction in paedophilia? British Journal of Psychiatry. 145. 657-660. 139 Gaffney, G.R., Lurie, S.F., & Berlin, F.S. (1984). Is there familial transmission of pedophilia? Journal of Nervous and Mental Disease. 172(9). 546-548. Gebhard, P.H., Gagnon, J.H., Pomeroy, W.B., & Christenson, C.V. (1965). Sex offenders: An analysis of types. New York, NY: Harper & Row/Hoeber. Giarretto, H. (1978). Humanistic treatment of father-daughter incest. Journal of Humanistic Psychology. 18(4), 59-76. Gillies, L.A., Hashmall, J.M., Hilton, N.Z., & Webster, C D . (1992). Relapse prevention in pedophiles: Clinical issues and program development. Canadian Psychology. 33(2), 199-210. Gold, J.R. (1996). Key concepts in psychotherapy integration. New York, NY: Plenum Press. Goldstein, M.J. (1973). Exposure to erotic stimuli and sexual deviance. Journal of Social Issues. 29(3), 197-219. Goldstein, M., Kant, H., Judd, L., Rice, C , & Green, R. (1971). Experience with pornography: Rapists, pedophiles, homosexuals, transsexuals, and controls. Archives of Sexual Behavior. 1(1), 1-15. Gordon, R., & Verdun-Jones, S.N. (1983). Ethics and ethical dilemmas in the treatment of sex offenders. In S.N. Verdun-Jones, & A. A Keltner (Eds.), Sexual aggression and the law (pp. 74-96). Burnaby, BC: Criminology Research Centre. Gottfredson, D.M. (1987). Prediction and classification in criminal justice decision making. In D.M. Gottfredson & M. Tonry (Eds.). Prediction and classification: Criminal Justice decision making (pp. 1-20). Chicago, JJL: University of Chicago. 140 Greenspan, EX. , & Rosenberg, M. (1996). Martin's annual criminal code-1997 [ISSN 0527-7892]. Aurora, ON: Canadian Law Book Inc. Groth, A.N. (1979). Men who rape: The psychology of the offender. New York, NY: Plenum Press. Groth, A.N., Hobson, W.F., & Gary, T.S. (1982). The child molester: Clinical observations. In J. Conte & D. Shore (Eds.), Social work and child sexual abuse (pp. 120-144). New York, NY: Howorth. Groth, N.A., Hobson, W.F., Lucey, K.P., & Pierre, J.S. (1981). Juvenile sexual offenders: Guidelines for treatment. International Journal of Offender Therapy and Comparative Criminology. 25, 265-275. Hammer, E.F. (1957). A psychoanalytic hypothesis concerning sex offenders: A study by clinical psychologic techniques. Journal of Clinical and Experimental Psychopathology. 18(2), 177-184. Hammer, E.F. (1968). Symptoms of sexual deviation: Dynamics and etiology. The Psychoanalytic Review. 55(1), 5-27. Hare, R.D. (1990, September). Manual for the revised psychopathy checklist. Vancouver, BC: University of British Columbia. Hare, R.D. (1993). Without conscience. Toronto, ON: Pocket Books. Hare, R.D. (1996). Psychopathy: A clinical construct whose time has come. Criminal Justice and Behavior. 23(1), 25-54. Hart, S.D., Forth, A.E. & Hare, R.D. (1990). Performance of criminal psychopaths on selected neuropsychological tests. Journal of Abnormal Psychology. 99(4), 374-379. Hart, S.D., Hare, R.D., & Harpur, T.J. (1992). The Psychopathy Checklist: Overview for researchers and clinicians. In J. Rosen & P. McReynolds (Eds.), Advances in psychological assessment. Vol. 7. New York: Plenum. Herman, J., & Hirschman, L. (1977). Father-daughter incest. Signs: Journal of Women in Culture and Society. 2, 735-756. Hindman, J. (1988, July/August). Research disputes assumptions about child molesters. National District Attorneys Association Bulletin. 7(4), 1-3. Horley, J., & Quinsey, V.L. (1994). Assessing the cognitions of child molesters: Use of the semantic differential with incarcerated offenders. The Journal of Sex Research. 31(3), 171-178. Horley, J., Quinsey, V.L., & Jones, S. (1995, June). Child molesters' beliefs, attitudes, and values. Paper presented at the meeting of the Canadian Psychological Association, Charlottetown, PEL Horwood, H. (1987, January 28). T am not nuts.' The Province, p. 8. Howell, D.C. (1982). Statistical methods in psychology. Boston, MA: Duxbuty Press. Hucker, S., Langevin, R., Wortzman, G., Bain, J., Handy, L., Chambers, J., & Wright, S. (1986). Neuropsychological impairment in pedophiles. Canadian Journal of Behavioural Science. 18(4), 440-448. Hudson, S.M., Ward, T., and Marshall, W.L. (1992). The abstinence violation effect in sex offenders: A reformulation. Behaviour Research and Therapy. 30(5), 435-441. Hunter, J.A., Jr., Becker, J.V., Kaplan, M., & Goodwin, D.W. (1991). Reliability and discriminative utility of the Adolescent Cognitions Scale for juvenile sexual offenders. Annals of Sex Research. 4, 281-286. 142 Kirk, R.E. (1968). Experimental design: Procedures for the behavioral sciences. Belmont, CA:Wadsworth. 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. 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, NY: Plenum Press. Knight, R.A., Rosenberg, R., & Schneider, B.A. (1985). Classification of sexual offenders: Perspectives, methods, and validation. In A.W. Burgess (Ed.), Rape and sexual assault (pp. 222-293). New York, NY: Garland. Kraemer, W. (Ed.). (1976). The forbidden love: The normal and abnormal love of children. London: Sheldon. Krafft-Ebing, R. (1924). Psvchopathia sexualis (R.F. Rebman, Trans.). New York, NY: Physicians & Surgeons Book Co. (Original work published 1886). Laflen, B., & Sturm, W.R. (1994). Understanding and working with denial in sexual offenders. Journal of Child Sexual Abuse. 3(4), 19-36. Lang, R.A., & Frenzel, R.R. (1988). How sex offenders lure children. Annals of Sex Research. 1. 303-317. Langevin, R. (1983). Sexual strands: understanding and treating sexual anomalies in men, (pp. 7-29). Hillsdale, NJ: Lawrence Erlbaum. 143 Langevin, R., & Lang, R. (1988). Incest offenders: A practical guide to assessment and treatment. [Brochure]. Etobicoke, ON: Juniper Press. Langevin, R., Paitich, D., Freeman, R, Mann, K., & Handy, L. (1978). Personality characteristics and sexual anomalies in males. Canadian Journal of Behavioural Science. 10(3), 222-238. Langevin, R., Wortzman, G., Wright, P., & Handy, L. (1989). Studies of brain damage and dysfunction in sex offenders. Annals of Sex Research. 2, 163-179. Lanning, K.V. (1986). Child molesters: A behavioral analysis for law enforcement (2nd ed.). In R.R. Hazelwood & AW. Burgess (Eds.), Practical aspects of rape investigation: A multi-disciplinary approach (pp. 1-112). New York, NY: Elsevier. Lanning, K. (1987). Child molesters: A behavioral analysis (2nd ed.). Arlington, VA: National Center for Missing and Exploited Children. Lanning, K. (1992, December). Child molesters: A behavioral analysis (3ri ed.). Arlington, VA: National Center for Missing and Exploited Children. Lanyon, R.I., (1986). Theory and treatment in child molestation. Journal of Consulting and Clinical Psychology. 54(2), 176-182. Lerman, H. (1986). A mote in Freud's eye: From psychoanalysis to the psychology of women. New York, NY: Springer. Lindsay, W.R., Neilson, C.Q., Morrison, F., & Smith, A.H.W. (1998). The treatment of six men with a learning disability convicted of sex offenses with children. British Journal of Clinical Psychology. 37, 83-98. Looman, J. (1995). Sexual fantasies of child molesters. Canadian Journal of Behavioural Science. 27(3), 321-332. Lorr, M. (1983). Cluster analysis for social scientists: Techniques for analyzing and simplifying complex blocks of data. San Francisco, CA: Jossey-Bass. Luke, D.A., Rapport, J., & Seidman, E. (1991). Setting phenotypes in a mutual help organization: Expanding behavior setting theory. American Journal of Community Psychology. 19(1), 147-167. Mair, K.J. (1993). The nature of the act. British Journal of Criminology. 33(2), 267-275. Malcolm, P.B., Andrews, D.A., Quinsey, V.L. (1993). Discriminant and predictive validity of phallometrically measured sexual age and gender preference. Journal of Interpersonal Violence. 8(4), 486-501. Marshall, W.L. (1988). The use of sexually explicit stimuli by rapists, child molesters, and nonoffenders. Journal of Sex Research. 25(2), 267-288. Marshall, W.L. (1989). Invited essay: Intimacy, loneliness, and sexual offenders. Behaviour Research and Therapy. 27(5), 491-505. Marshall, W.L. (1993). The role of attachments, intimacy, and loneliness in the etiology and maintenance of sexual offending. Sexual and Marital Therapy. 8(2), 109-121. Marshall, W.L., & Christie, M.M. (1981). Pedophilia and aggression. Criminal Justice and Behavior. 8(2), 145-158. Marshall, W.L., Fernandez, Y.M., Hudson, S.M., & Ward, T. (Eds.). (1998). Sourcebook of treatment programs for sexual offenders. New York, NY: Plenum Press. Masson, J.M. (1984a, January). A challenge to the foundations of psychoanalysis: Freud and the seduction theory. The Atlantic Monthly, pp. 33-60. Masson, J.M. (1984b). The assault on truth: Freud's suppression of the seduction theory. Toronto, ON: Farrar, Straus and Giroux. Masson, J. (1985) (Ed. & Trans.). The complete letters of Sigmund Freud to Wilhelm Fliess 1887-1904. Cambridge, MA: Belknap Press of Harvard University Press. McGuire, F.J., Carlisle, J.M., & Young, B.G. (1965). Sexual deviations as conditioned behaviour: A hypothesis. Behaviour Research and Therapy. 2(3), 185-190. Mehrabian, A., & Epstein, N. (1972). A measure of emotional empathy. Journal of Personality. 40, 525-543. Millon, T. (1996). Disorders of Personality: DSM-IV and Beyond (2nd ed.). New York, NY: John Wiley & Sons. Nemes, I. (1992, Winter). The relationship between pornography and sex crimes. Journal of Psychiatry and Law. 459-481. Nutter, D.E., & Kearns, M.E. (1993). Patterns of exposure to sexually explicit material among sex offender, child molesters, and controls. Journal of Sex and Marital Therapy. 19(1), 77-85. Ogloff, J., Wong, S., & Greenwood, A. (1990). Treating criminal psychopaths in a Therapeutic Community program. Behavioral Sciences and the Law. 8(2), 181-190. Orr, B.Y. (1991). Male adolescent sex offenders: A comparison of two treatment approaches. Journal of Child and Youth Care. Special Issue, 87-101. O'Shaughnessy, R., Hare, R.D., Gretton, H , & McBride, M. (1994). [Psychopathy in adolescent sex offenders]. Unpublished raw data. Oxford, (1984). The compact edition of the oxford English dictionary. (Vol. 2). London: Oxford University. Paris, F.A. (1992). Child molester's retrospective accounts of their acts and subsequent reasoning on ethical issues. Unpublished master's thesis, University of British Columbia, Vancouver, British Columbia, Canada. Prentky, R.A., & Knight, R.A. (1986). Impulsivity in the lifestyle and criminal behavior of sexual offenders. Criminal Justice and Behavior 13(2). 141-164. Rada, R.T. (1978). Sexual psychopathology: Historical survey and basic concepts. In R.T. Rada (Ed.), Clinical aspects of the rapist. New York, NY: Grune & Stratton. Reber, A.S. (1987). Dictionary of psychology. New York, NY: Penguin. Regestein, Q.R., & Reich, P. (1978). Pedophilia occurring after onset of cognitive impairment. Journal of Nervous and Mental Disease. 166(11). 794-798. Rice, M.E., & Harris, G.T. (1992). A comparison of criminal recidivism among schizophrenic and nonschizophrenic offenders. International Journal of Law and Psychiatry. 15, 397-408. Rice, M.E., Harris, G.T., & Cormier, CA. (1992). An evaluation of a maximum security therapeutic community for psychopaths and other mentally disordered offenders. Law and Human Behavior. 16(4). 399-412. Rice, M.E., Harris, G.T., & Quinsey, V.L. (1990). A follow-up of rapists assessed in a maximum-security psychiatric facility. Journal of Interpersonal Violence. 5(4), 435-448. Rice, M.E., Quinsey, V.L., & Harris, G.T. (1991). Sexual recidivism among child molesters released from a maximum security psychiatric institution. Journal of Consulting and Clinical Psychology. 59(3), 381-386. Robinson, D.N. (1979). Systems of modern psychology: A critical sketch. New York, NY: Columbia University. Romesburg, H.C. (1984). Cluster analysis for researchers. Toronto, ON: Lifetime Learning. Rothbard, J.C., & Shaver, P.R. (1994). Continuity of attachment across the life span. In M.B. Serling & W.H. Berman (Eds.), Attachment in adults: Clinical and developmental perspectives (pp. 31-71). New York, NY: Guilford. Sandfort, T.G.M. (1984). Sex in pedophilic relationships: An empirical investigation among a non-representative group of boys. Journal of Sex Research. 20, 123-142. Scott, M.L., Cole, J.K., McKay, S.E., Golden, C.J., & Liggett, K.R. (1984). Neuropsychological performance of sexual assaulters and pedophiles. Journal of Forensic Sciences. 29(4), 1114-1118. Serin, R.C. (1993). Diagnosis of psychopathology with and without an interview. Journal of Clinical Psychology. 49(3), 367-372. 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., Barbaree, H.E., & Serin, R.C. (1994). Comparison of low-fixated and high-fixated child molesters. Canadian Psychology. 35(2), 51. Sgroi, S.M. (1978)^  Child sexual assault: Some guidelines for intervention and assessment. In A.W. Burgess, A.N. Groth, L L . Holmstrom, & S.M. Sgroi (Eds.) Sexual assault of children and adolescents (pp. 129-142). Lexington, MA: D.C. Heath. 148 Short, R.H., & Hess, G.C. (1995). Fetal alcohol syndrome: Characteristics and Remedial Implications. Developmental Disabilities Bulletin. 23(1), 12-29. Simon, B. (1992). "Incest-see under oedipus complex": The history of an error in psychoanalysis. Journal of the American Psychoanaltyic Association. 40(4), 955-988. Skinner, H.A. (1977). 'The eyes that fix you': A model for classification research. Canadian Psychological Review. 18, 142-151. Sneath, P.H.A., & Sokal, R.R. (1973). Numerical Taxonomy. San Francisco: W.H. Freeman. Sorensen, T., & Snow, B. (1991). How children tell: The process of disclosure in child sexual abuse. Child Welfare. 70, 3-15. Staats, AW. (1983). Psychology's crisis of disunity: Philosophy and method for a unified science. New York, NY: Praeger. Stagner, R. (1988). A history of psychological theories. New York, NY: Macmillan. Steidman, B.T., Marshall, W.L., Hudson, S.M., & Robertson, P.J. (1994). An examination of intimacy and loneliness in sex offenders. Journal of Interpersonal Violence. 9(4), 518-534. Streissguth, A.P. (1994). A long-term perspective of FAS. Alcohol. Health and Research World. 18(1). 74-81. Szasz, T. (1980). Sex by prescription. New York, NY: Anchor Press. Szasz, T. (1991). The medicalization of sex. Journal of Humanistic Psychology. 31(1), 34-42. Ward, T., Hudson, S.M., & Marshall, W.L. (1994). The abstinence violation effect in child molesters. Behaviour Research and Therapy. 32(4), 431-437. Ward, T., Hudson, S.M., & Marshall, W.L. (1995). Cognitive distortions and affective deficits in sex offenders: A cognitive deconstructionist interpretation. Sexual Abuse: A Journal of Research and Treatment. 7(1), 67-83. Weiss, G., Hechtman, L., Perlman, T., Hopkins, J., & Wener, A. (1979). Hyperactives as young adults: A controlled prospective ten-year follow-up of 75 children. Archives of General Psychiatry. 36, 675-681. Williams, L.M., & Finkelhor, D. (1990). The characteristics of incestuous fathers: A review of recent studies. In W.L. Marshall, D.R. Laws, & H.E. Barbaree (Eds.), Handbook of sexual assault: Issues, theories, and treatment of the offender (pp. 231-255). New York, NY: Plenum Press. Wilson, P. (1981). The man they called a monster: Sexual experiences between men and boys. Cassell, Australia: Cassell. Wong, S. (1988). Is Hare's psychopathy checklist reliable without the interview? Psychological Reports. 62, 931-934. Worling, J.R. (1998). Adolescent sexual offender treatment at the SAFE-T Program. In W.L. Marshall, Y.M. Fernandez, S.M. Hudson, & T. Ward (Eds.), Sourcebook of treatment programs for sexual offenders (pp. 273-285). New York, NY: Plenum Press. Wormith, J.S. (1986). Assessing deviant sexual arousal: Physiological and cognitive aspects. Advances in Behaviour Research and Therapy. 8(3), 101-137. Youth Court Services (1997). The sexual response meter (penile plethysmograph) [Brochure]. Burnaby, BC: Author. 151 APPENDIX: A l A PSYCHOLOGICAL CLASSIFICATION SYSTEM BASED ON CHILD MOLESTERS' MOTIVATIONS Adolescent Files. Coding Sheet for Background Information on the Child Molester Subject ID no.7: Coder: Current date: _ _ _ / _ _ / _ yr. mo. day Information on the Child Molester. Birthdate _ _ / _ _ _ _ / _ yr. mo. day Education: • Highest level attained (years) Description . Employment: • Currently employed/unemployed or previously employed/unemployed (Circle one) Explain Information pertaining to child molester's biological parents (e.g., education, employment, and amount of contact with their child): • mother • father 6 If information is unavailable, please record "u.a." on the form for each item. 7 A list will be constructed relating file numbers to subject ID numbers. This list will be destroyed upon completion of the project. 152 Previous charges or convictions (Underline whether it is charges or convictions, try to obtain charges): • Non-sexual charges or convictions # type code . type code . type code . • Sexual charges or convictions # type code . type code . type code . Offenses not previously charged for (e.g., under the age of 12 or admitted committing) # type code . type code . type code . Current charges or convictions (current): • Non-sexual charges and convictions # type code . _ type code . type code . • Sexual charges and convictions # type code . type code . type code . Drug Involvement: Alcohol Involvement: Other forms of substance abuse? Describe 153 APPENDIX: B l A PSYCHOLOGICAL CLASSIFICATION SYSTEM BASED ON CHILD MOLESTERS' MOTIVATIONS Adolescent Files: Coding Sheet for Potential Causal Variables1 ,8 Subject ID no.9: Coder: Current date: / / yr. mo. day Coding of Potential Causal Variables: 1. Was the offender sexually abused as a child? If the offender was sexually abused as a child or adolescent (Code for multiple incidents, when necessary), a. age at the time of the offense: • under 14 years of age (3 point) • 14 years to under 18 (2 point) • unknown, but abused (1 point) b. relationship to his offender: • biological mother (8 points) • biological father (7 points) • guardian or step-parent (6 points) • family member (e.g., brother) (5 points) • relative (e.g., uncle, grandfather) (4 points) • acquaintance (e.g., mother's boyfriend, babysitter) (3 points) • stranger (2 points) • unknown (1 point) c. sex of his offender: • male (1 point) • female (2 points) Note source of information . (1) 8 If information is unavailable, please record "u.a." on the form for each item. 9 A list will be constructed relating file numbers to subject ID numbers. This list will be destroyed upon completion of the project. 154 2. Are there reports that the offender was emotionally abused as a child? • Yes (1 point) • No (0 points) Note source of information . (1) 3. Are there reports that the offender was physically abused as a child? • Yes (1 point) • No (0 points) Note source of information . (1) 4. Are there reports that the offender was neglected as a child? • Yes (1 point) • No (0 points) Note source of information . (1) 5. During childhood/adolescence did the offender have close relationships with his • mother (1 point) • father (1 point) • another adult (1 point) • sibling(s) (1 point) • peer(s) (1 point) • other (1 points) Note source of information . (1) 6. Sexual aggression model: • parents used punitive forms of discipline (no=0, yes=l) • exposed to aggressive, drunken male models (no=0, yes=l) • drinking of alcohol during offense (no=0, yes=l) • associates sexual arousal with sexual aggression (no=0, yes=l) Note sources of information . (1) 7. Reported levels of: • intimacy (self-revealing) (low=0, high=l) • loneliness, a loner (low=0, high=T) • hostility towards girls/women (low=0, high=l) Note source of information . (1) 8. Adult attachment style (if neither + or - , please note): • Image of self (positive=0, negative=l) • Image of others (positve=0, negative=l) Note source of information . (1) 155 9. Participation in peer groups: • low sociability (3 points) (i.e., few, if any close friends) • moderate sociability (2 points) • high sociability (1 point) (i.e., enjoys people, active in sports) Note source of information . (1) 10. Cognition Scale: • score . (1) 11. Child molester's description of self (Incomplete Sentence Test, question 37): • I • (l=positive, 0=neutral, -l=negative). (1) 12. Child molester's description of boys (Incomplete Sentence Test, question 7): • Boys • (l=positive, 0=neutral, -l=negative) . (1) 13. Child molester's description of girls (Incomplete Sentence Test, question 40): • Most girls . • (l=positive, 0=neutral, -l=negative). (1) 14. Responsibility • Child molester denies he committed the current offense (3 points) • Child molester minimizes the offense (e.g., did not hurt child) (2 points) • Child molester admits committing the offense, but relinquishes responsibility for his actions (e.g., personal history, alcohol, or stress) (1 point) • Accepts responsibility, experiences guilt (0 points) Note source of information . (1) 15. The current offense: • Reason(s) given for committing the offense . . • (1) • Offender's affective reaction to committing the offense. • (1) 16. Are there reports that the child molester engages in concrete reasoning? • yes (1 point) • no (0 points) Note source of information . (1) 17. Reports of brain damage or head injuries (yes=l point, no=0 points): Explain • (1) 156 18. Penile Plethysmograph Report (1st report, if conclusive). Average arousal for: a. male subjects % b. female subjects % c. non-coerced subjects % d. coerced subjects % e. post-pubescent subjects % f. pre-pubescent subjects % g. child subjects %. (7) 19. PCL-R scores: • Factor 1 • Factor 2 • Total score . (1) 20. IQ Index (Order varies): • performance • verbal • full . (1) 21. List categories for the: • primary diagnoses DMS . ICD DMS . ICD DMS . ICD • secondary diagnoses DMS . ICD DMS . ICD DMS _ . ICD Other diagnoses DMS . ICD DMS . ICD DMS ICD Describe any sexual problems (e.g., fetishes, high or low sex drive, or medical problems involving the genitalia) 157 22. Approximate number of placements # • List Approximate number of schools # • List Transiency • unstable placements and schooling (2 points) • unstable placements or schooling (1 point) • stable placements and schooling (0 points) Note source of information . (1) 23. Reports of either of the offender's biological parents being heavy drinkers? • mother (yes=l, no=0) • father (yes =1, no=0) Note: Did mother drink during pregnancy? . (1) 24. Reports of either of the offender's biological parents being addicted to drugs? • mother (yes=l, no=0) • father (yes =1, no=0) Note: Did mother take drugs during pregnancy? . (1) 25. During childhood/adolescence was the offender ever diagnosed with. • FAS/FAE (yes=l, no=0) • Conduct disorder (yes=l, no=0) • Attention deficit disorders (yes=l, no=0) Note source(s) of information . (1) 26. Exposure to pornography (e.g., books, magazines, movies, strip clubs, etc.) • during pubescence/adolescent (soft core=l point; hard core=2 points) • prior to committing the offense (soft core=l point; hard core=2 points) Describe Note source(s) of information (1) 158 27. Fear ofNegative Evaluation (Watson & Friend, 1969) • score . (1) 28. Temperament or personality measures • Basic Personality Inventory Denial (Den) Impulse Expression (ImE) Social Introversion (Soi) Alienation (Am) • The Jesness Inventory Denial Scale (Den) Manifest Aggression Scale (MA) Social Anxiety Scale (SA) Value Orientation Scale (VO) 29. Minnesota Multiphasic Personality Inventory (MMPI, MMPI-II, MMPI-A): Raw Scores T-Scores Psychopathic Deviant (Pd) Schizophrenia (Sc) Social Introversion (Si) (3) 30. Sympathetic/empathic measure • UCLA Empathy Scale (score) . (1) Comments Raw Scores T-Scores Raw Scores T-Scores ZZ Z Z (4) 31. Internal/External Locus of Control: • (score) . (1) Comments 32. Record interpretations of the House-Tree-Person Projective Drawing Test 159 APPENDIX: Cl A PSYCHOLOGICAL CLASSIFICATION SYSTEM BASED ON CHILD MOLESTERS' MOTIVATIONS Adolescent Files: Coding Sheet for Current Sexual Offense 10 Subject ID no. 11. Coder: Current date: / / yr. mo. day Information on the Victim(s): Number of victims: 1=1-3, 2=4-6, and 3=7 or more. Type of victims: l=children, 2=children and peers. Sex of victims: l=males, 2=females, and 3=males and females. Most recent victim that is a child: 1. Ages: Age of offender (at the time of the offense). Age of victim (at the time of the offense). Age difference (at the time of the offense) 2. Sex of offender . Sex of victim Circle one: homosexual/heterosexual act. 3. Relationship to offender • stranger • acquaintance • relative • sibling/half-sister or brother • other Was the offender in a position of authority (e.g., babysitting)? Circle yes or no. If information is unavailable, please record "u.a." on the form for each item. 1 1 A list will be constructed relating file numbers to subject ID numbers. This list will be destroyed upon completion of the project. 160 Description of the Current Sexual Offense: Date(s) of the sexual offense(s): _ _ / _ _ / _ to / / _ yr. mo. day yr. mo. day Approximate time to Place the offense occurred 1. Prior to the sexual offense: • planned or fantasized about the offense • did not plan or fantasize about the offense 2. Reports of substance abuse during the offense: • under the influence of alcohol or drugs • not under the influence of alcohol or drugs type 3. Seduction, technique: • affectionate (e.g., cuddling, comforting) • playful • grab/take advantage (e.g., commit acts while child is sleeping) • manipulate/bribe • threats • physical force • physical force, with a weapon 4. Type of sexual activity: (Circle direction of the activity: victim to offender, offender to victim, or both.) • verbal contact (e.g., request) Circle one: v to o / o to v / both • visual contact (e.g., exhibitionism) Circle one: v to o / o to v / both • physical contact (e.g., fondling) Circle one: v to o / o to v / both • oral sex Circle one: v to o / o to v / both • simulating intercourse Circle one: v to o / o to v / both • penetration, digital Circle one: v to o / o to v / both • penetration, vaginal or anal Circle one: v to o / o to v / both 5. Violence causing: • no physical harm • temporary physical harm • permanent physical harm 161 6. Frequency, the sexual abuse occurred • once • a few times • repeatedly/regularly 7. Duration, the sexual abuse lasted • day(s) • month(s) • years(s) 8. Activities came to the awareness of a(n): • father figure • mother figure • sibling • relative • acquaintance • teacher or other professional Explain the sequence of events 9. If the victim disclosed was it: • accidental • purposeful Explain Comments: 162 APPENDIX: A2 A PSYCHOLOGICAL CLASSIFICATION SYSTEM BASED ON CHILD MOLESTERS' MOTIVATIONS Adult Files: Coding Sheet for Background Information on the Child Molester12 Subject ID no.13: Coder. Current date: _ _ / _ _ / _ yr. mo. day Information on the Child Molester. Birthdate _ _ / _ _ / _ _ yr. mo. day Marital Status (Updated CPI Record) l=single, 2=common-law, 3=married, 4=separated, 5=divorced, 9=unknown Number of biological children . Education: • Highest grade level attained in high school (years) Describe any certificates, diplomas, or degrees completed since high school: Employment: • Employed/unemployed full-time prior to the offense (Circle one) • Employed/unemployed part-time prior to the offense (Circle one) Describe previous types of employment: Hobbies or interests: 1 2 If information is unavailable, please record "u.a." on the form for each item. 1 3 A list will be constructed relating file numbers to subject ID numbers. This list will be destroyed upon completion of the project. 163 Information pertaining to child molester's biological parents (i.e., employment, and amount of contact with their child): • mother deceased? • father deceased? • parent's relationship Previous convictions (Use additional sheets if necessary): • Non-sexual convictions # type code_ type code_ type code • Sexual convictions # type code type code_ type code_ Number of charges, no convictions # type code type code_ type code_ Current charges or convictions (current): • Non-sexual charges and convictions # type code type code type code Sexual charges and convictions # type code type code type code_ 164 Are there reports that drug involvement has been a problem for the offender in the past (i.e., abuse or dependency)? Yes/No (Circle one) Past Drug Involvement: Was drug involvement a problem for the offender at the time of the offense (i.e., abuse or dependency)? Yes/No (Circle one) Drug Involvement at time of the Offense: Are there reports that alcohol involvement has been a problem for the offender in the past (i.e., abuse or dependency)? Yes/No (Circle one) Past Alcohol Involvement: Was alcohol involvement a problem for the offender at the time of the offense (i.e., abuse or dependency)? Yes/No (Circle one) Alcohol Involvement at time of the Offense: 165 APPENDIX: B2 A PSYCHOLOGICAL CLASSIFICATION SYSTEM BASED ON CHILD MOLESTERS' MOTIVATIONS Adult Files: Coding Sheet for Potential Causal Variables14 Subject ID no.15: Coder: Current date: _ _ / _ _ / _ yr. mo. day Coding of Potential Causal Variables: 1. Are there reports that the offender was sexually abused as a child (no=0 points)? If the offender was sexually abused as a child or adolescent (Code for multiple incidents, when necessary on additional sheets), a. age at the time of the offense: • under 14 years of age (3 points) • 14 years to under 18 (2 points) • unknown, but abused (1 point) c. relationship to his offender: • biological mother (8 points) • biological father (7 points) • guardian or step-parent (6 points) • family member (e.g., brother) (5 points) • relative (e.g., uncle, grandfather) (4 points) • acquaintance (e.g., mother's boyfriend, babysitter) (3 points) • stranger (2 points) • unknown (1 point) c. sex of his offender: • male (1 point) • female (2 points) • (1) 1 4 If information is unavailable, please record "u.a." on the form for each item. 1 5 A list will be constructed relating file numbers to subject ID numbers. This list will be destroyed upon completion of the project. 166 2a. Sexual History: The offender learned about sexual activities... 2b. Sexual Activities: If there are reports of the offender ever willingly engaged in the following activities circle each one (2 points each). If there are reports that the offender ever had fantasies or sexual urges to perform the following activities underline each one (1 point each). 1. Exhibitionism 2. Frotteurism (touching or rubbing against a non-consenting person) 3. Pedophilia 4. Sexual masochism (own suffering is sexual exciting) 5. Sexual sadism (victim's suffering is sexual exciting) 6. Incest (sexual intercourse with a person related by blood—sister, daughter) 7. Anal intercourse (circle whether it was a male/female partner) 8. Bestiality 9. Fetishism 10. Transvestic fetishism 11. Voyeurism 12. Sexual activities with prostitutes 2c. Reported sexual orientation (circle one): heterosexual/homosexual children/adults Describe any sexual problems (e.g., fetishes, deviant fantasies, high or low sex drive, or medical problems involving the genitalia) 3. Masturbation,: • Denies or excessively engages in masturbation (1 point) • Negative attitude towards masturbation (1 point) • (1) 4. Are there reports that the offender was emotionally abused or emotionally deprived (circle) as a child? • Yes (1 point) • No (0 points) • (1) 5. Are there reports that the offender experienced corporal punishment or was physically abused (circle) as a child? • Yes (1 point) • No (0 points) • (1) 167 6. Are there reports that the offender was neglected or abandoned (circle) as a child? • Yes (1 point) • No (0 points) • (1) 7. The offender has had a close and supportive relationship (i.e., securely attached, not avoidant or ambiguously attached) with his: • Male caregiver (1 point) • Female caregiver (1 point) • Another adult (1 point) • Sibling(s) (1 point) • Peer(s) (1 point) • Other (1 point) . • (1) 8. Sexual Aggression Model. There are reports that the offender... • experienced punitive forms of discipline (no=0, yes=l) • was exposed to aggressive, drunken male models (no=0, yes=l) • drank alcohol during the current sexual offense (no=0, yes=l) • associates sexual arousal with sexual aggression (no=0, yes=l) • (1) 9. Reported levels of: • intimacy (e.g., open and self-revealing during the interview) (low=l, high=0) • loneliness (i.e., described as a loner in a report or assessment) (low=0, high=l) • hostility towards girls/women (low=0, high= 1) • (1) 10. Number of relationships over 6 months (e.g., girlfriend, wife) • 5 or more steady/long-term relationships (3 points) • 3-4 steady/long-term relationships (2 points) • 1-2 steady/long-term relationships (1 point) • 0 steady/long-term relationships (0 points) 11. Adult attachment style within romantic relationships (if neither + or -, please note): • Image of self as a partner (positive=0, negative=l) • Image of male/female (circle) partner(s) (positve=0, negative=l) • (1) 168 12. Socializes with other adults: • low sociability (3 points) (e.g., an introvert with few, if any close friends or socializes with children/elderly) • moderate sociability (2 points) • high sociability (1 point) (e.g., an extrovert who enjoys being with people, participating in sports or other activities) • (1) 13. Responsibility • Child molester denies he committed the current offense (3 points) • Child molester minimizes the offense (e.g., did not hurt child) (2 points) • Child molester admits committing the offense, but relinquishes responsibility for his actions (e.g., personal history, alcohol, or stress) (1 point) • Accepts responsibility, experiences guilt (0 points) • (1) 14. The current offense: • Reason(s) given for committing the offense . (1) • Offender's affective reaction to committing the offense. • (1) 15. Are there reports that state specifically that the child molester engages in concrete reasoning, or has difficulties with abstract thought? • yes (1 point) • no (0 points) . • (1) 16. Reports of brain damage or head injuries (permanent damage or sever injuries=l point; temporary damage or mild injuries=0 points): Explain , , • (1) 17. PCL-R scores (see scoring sheets): • Factor 1 • Factor 2 • Total score . (1) 169 18. Evaluation of Intelligence (Raven's Standard Progressive Matrices,) FROM PSYCHOLOGISTS ASSESSMENT, DO NOT OPEN ENVELOPE: • Above average range • High average range • Mid average range • Low average range (1 point) • Below average range (2 points) Note exact wording, if different from the above labels 19. Diagnoses (Personality Assessment Inventory, or psychiatrists assessment): FROM PSYCHOLOGISTS/PSYCHIATRISTS ASSESSMENT DO NOT OPEN ENVELOPE: • Clinical Disorders or other conditions that may be a focus of clinical attention DMS DMS DMS • Personality Disorders/Mental Retardation DMS DMS DMS General Medical Conditions DMS DMS DMS 20. Any previous contact with mental health professionals? • Yes (2 points=assessments, treatments, hospitalizations, etc.) • Yes, briefly (1 point=military or other screening procedure, etc.) • No (0 points) 170 21. Approximate number of residents # Approximate number of occupations # Transiency • Unstable residences (a large number of residences or no fixed address) and sporatic employment (or unemployed for a long period) (2 points) • Moderately stable residences (e.g., various locations, but lives in own residence) and employment (e.g., variety of jobs, but always employed) (1 point). Or a stable residence or employment (1 point) • Stable residence and employment (e.g., predominately living in the same location and working within the same occupation) (0 points) (1) 22. Reports of either one of the offender's biological parents being involved with alcohol (i.e., abuse or dependency)? • mother (yes=l, no=0) • father (yes =1, no=0) Did mother drink during pregnancy? . (1) 23. Reports of either one of the offender's biological parents being involved with drugs (i.e., abuse or dependency)? • mother (yes=l, no=0) • father (yes =1, no=0) Did mother take drugs during pregnancy? . (1) 24. Exposure to pornography (e.g., books, magazines, movies, strip clubs, etc.) • during pubescence/adolescent (soft core=l point; hard core=2 points) • during adulthood (soft core—1 point, hard core=2 points) • prior to committing the offense (soft core=l point; hard core=2 points) (1) Describe 171 25. Offender's personal needs or interests were not satisfied prior to the current offense(s): • economical (e.g., unemployed) (yes=l, no=0) • emotional (e.g., reports of offender being lonely, isolated) (yes=l, no=0) • social relationships (e.g., low levels of sociability with adults) (yes=l, no=0) • intimate relationships (e.g., did not have a sexual partner) (yes=l, no=0) (1) 172 APPENDIX: C2 A PSYCHOLOGICAL CLASSIFICATION SYSTEM BASED ON CHILD MOLESTERS' MOTIVATIONS Adult Files: Coding Sheet for Current Sexual Offense16 Subject ID no.11: Coder: Current date: _ / / yr. mo. day Information on the Victim(s): Number of victims: 1=1-3, 2=4-6, and 3=7 or more. Type of victims: l=children, 2=children and peers. Sex of victims: l=males, 2=females, and 3=males and females. Most recent victim that is a child: 1. Ages: Age of offender (at the time of the offense). Age of victim (at the time of the offense). Age difference (at the time of the offense) 2. Sex of offender . Sex of victim Circle one: homosexual/heterosexual act. 3. Relationship to offender • stranger • acquaintance • relative • sibling/half-sister or brother • other Was the offender in a position of authority (e.g., babysitting)? Circle yes or no. If information is unavailable, please record "u.a." on the form for each item. 1 7 A list will be constructed relating file numbers to subject ID numbers. This list will be destroyed upon completion of the project. 173 Description of the Current Sexual Offense: Date(s) of the sexual offense(s): _ / _ _ / _ to _ / _ _ / _ _ yr. mo. day yr. mo. day Approximate time to Place the offense occurred 1. Prior to the sexual offense the child molester: • planned or fantasized about the offense • did not plan or fantasize about the offense 2. Reports of substance abuse during the offense the child molester: • was under the influence of alcohol/drugs • not under the influence of alcohol or drugs type 3. The child molester lowered the victim's inhibitions through the use of • cigarettes • alcohol • drugs • pornographic magazines • pornographic videos 4. Seduction, technique: • affectionate (e.g., cuddling, comforting) • playful • grab/take advantage (e.g., commit acts while child is sleeping) • manipulate/bribe • threats • physical force • physical force, with a weapon 5. Type of sexual activity: (Circle direction of the activity: victim to offender, offender to victim, or both.) • verbal contact (e.g., request) Circle one: v to o / o to v / both • visual contact (e.g., exhibitionism) Circle one: v to o / o to v / both • physical contact (e.g., fondling) Circle one: v to o / o to v / both • oral sex Circle one: v to o / o to v / both • simulating intercourse Circle one: v to o / o to v / both • penetration, digital Circle one: v to o / o to v / both • penetration, vaginal or anal Circle one: v to o / o to v / both 174 6. Violence causing: • no physical harm • temporary physical harm • permanent physical harm 7. Frequency, the sexual abuse occurred • once • a few times • repeatedly/regularly 8. Duration, the sexual abuse lasted • day(s) • month(s) • year(s) 9. Activities came to the awareness of a(n): • father figure • mother figure • sibling • relative • acquaintance • teacher or other professional Explain the sequence of events 10. If the victim disclosed was it: • accidental • purposeful Explain Comments: 175 ADDITIONAL SHEET PREVIOUS CONVICTIONS: • Non-sexual convictions # type code_ type _code_ type code_ type code_ type code type code_ type code_ type code_ type code_ • Sexual convictions # type code_ type code_ type code_ type code_ type code_ type code_ type code_ type code_ type code_ Number of charges, no convictions #_ type_ code_ type code_ type code_ type code_ type code_ type code_ Current charges or convictions (current): • Non-sexual charges and convictions # type^  code_ type code_ type code_ type code_ type code_ type code_ Sexual charges and convictions # typse code_ type code_ type code_ 176 ADDITIONAL SHEETS FOR OFFENDER'S EXPERIENCES, OF CHILD SEXUAL ABUSE Was the offender sexually abused as a child? If the offender was sexually abused as a child or adolescent (Code for multiple incidents, when necessary), a. age at the time of the offense: • under 14 years of age (3 points) • 14 years to under 18 (2 points) • unknown, but abused (1 point) b. relationship to his offender: • biological mother (8 points) • biological father (7 points) • guardian or step-parent (6 points) • family member (e.g., brother) (5 points) • relative (e.g., uncle, grandfather) (4 points) • acquaintance (e.g., mother's boyfriend, babysitter) (3 points) • stranger (2 points) • unknown (1 point) c. sex of his offender: • male (1 point) • female (2 points) Affective deficiencies, empathy score Deviant arousal levels, penile plethysmograph Biological parents, alcohol involvement Other diagnoses/DSM, Axis III Secondary diagnoses/DSM, Axis II Primary diagnoses/DSM, Axis I Physically abused as a child Emotionally abused as a child Sexually abused as a child • 1* If « **• Si Si < « I S a. f ! I I I r § r % r r I Cluster 1 Stable i 1 r 1 \ ES & I I % | | f? " High K CD a. High High High High High High | | | High High \ 1 i 1 1 | sr % High High High High High High High i f f TJ P S 1 CD 3 I 8-ry1 IS. I 8 2. IS o 3 8. po cn B a-i i o I 1 2. I 8 us A3 cl-' 3 I 2 Fro 3" - I SIS* Si I 1 ES B- Kg, Kg. Kg. B- <g. Kg. Kg. oo 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items