Correctional Service Canada
Symbol of the Government of Canada

Common menu bar links

Sexual Homicide and Paraphilias: The Correctional Service of Canada’s Experts Forum 2007

Warning This Web page has been archived on the Web.



Sexual Killers and Post Mortem Sexual Interference Offenders: Assessment, Treatment and Risk Management

Adam J. Carter, Ruth E. Mann and Helen C. Wakeling
Her Majesty’s Prison Service England and Wales

Address for correspondence:
HM Prison Service
Chromis Team Intervention Group
Cleland House, room 104 Page St.
London, England SW1P-4LN
Telephone: 020-7217 5075
Fax: 020-7217 5871



Establishing what part a sexual element played in the motivation of a killing and determining whether a killing was sexually motivated is a difficult task for correctional staff responsible for the assessment of sexual killers. When the perpetrator has engaged in sexual assault or non-assaultive behaviour (such as masturbation) following the death, the possibility of necrophilia as a motivating factor needs to be considered. This chapter will review the evidence for psychological treatment needs related to sexual killing and post mortem sexual interference. We used two well known dynamic risk factor tools as a framework to establish treatment needs for sexual killers and post mortem sexual interference offenders. We aimed to establish which treatment needs were potentially relevant to these offenders by examining the extant literature and two relevant data bases. Finally, we considered treatment implementation issues, again drawing on existing literature and our own clinical experience. This chapter concludes with recommendations for best practice and future research.



Sexual Killers and Post Mortem Sexual Interference Offenders: Assessment, Treatment and Risk Management

In this chapter, we aim to identify the treatment needs of Sexual Killers (SKs) and Post Mortem Sexual Interference Offenders (PMSIOs). We then draw on this information to make suggestions about appropriate treatment and management. While rare, SKs are a persistent and aberrant feature of society, surprisingly the extent to which researchers and correctional professionals understand this type of offence is limited. There is not even common agreement on the definition of this behaviour, which is necessary if we are to progress in detection, research, assessment, and treatment.

There are a number of difficulties in establishing that an offence is a sexual killing (Grubin, 1994), starting with problems with the manner in which it is recorded. If the sexual aspect of a crime is not easily recognizable, it can be recorded as “unknown motive” (Folino, 2000). Canada is the only country that specifically records data on sexual killings (Schlesinger, 2004) although there have been concerns that the definition used in Canada’s Criminal Code is not broad enough to accurately capture this crime (Roberts & Grossman, 1993). Although these aforementioned difficulties mean that the actual figure could be higher, it was estimated that sexual homicide accounted for 4% of the total homicides in Canada between 1974-1986 (Roberts & Grossman, 1993). In the United Kingdom in 2003, the sexual homicide rate was estimated at 6% of all homicides (Beech, Fisher, & Ward, 2005).

Whatever the actual figure, within Her Majesty’s Prison Service establishments holding life sentence prisoners, assessment of potential and actual SKs is a regular occurrence as the vast majority of these offenders can expect to be eligible for parole at some point. Prison staff may have to carry out assessments to determine whether the killing was sexual, often in a context of the offender’s denial (Folino, 2000). Assessors must identify the triggers and antecedents to the crime and quite possibly recommend or provide treatment for these perpetrators.

Determining that a Killing is Sexual

When some kind of sexual act is attached to a killing, it is usually deemed a sexual killing (Folino, 2000). The intended or actual sex could be attached prior to, during or after the killing (Porter, Woodworth, Earle, Drugge, & Boer, 2003) and in some cases could occur in two or at all three stages. As well as establishing if and when sex was attached, the motivation for the killing is another clinically important piece of information to determine if it should be deemed sexual. If a sexual act was intended but not actually carried out, only the perpetrator’s openness will tell us that the killing was in fact sexual.

Generally, when there is a fusion of sex and aggression, (i.e. when both occur simultaneously or in very close succession) forensic staff is likely to reach conclusions that a killing was sexual, although the nature of the association between sex and aggression is still extremely varied. Malmquist (1996) suggested a “Working breakdown” of sexual killings. In rape killings, the homicide takes place during the course of a “sexual act” and the death is not “part of a ritualized attack”. Second, the lust killer is “One who has made a vital connection between sexual gratification and violence” (Holmes, 1991, p. 67). In some cases of lust killing, the act of murder becomes “the equivalent of coitus” (p. 174) and replaces actual sexual activity (Podolsky, 1965). Malmquist stated that when a victim or a witness is killed, to help the perpetrator escape detection this should be considered, Killings after a sexual act to destroy evidence. Bartholomew, Milte and Galabally (1975) and Malmquist have questioned whether killings after a sexual act to destroy evidence meets the criteria for a sexual killing. However, West (1987) believed killing to eliminate a witness to rape would count as a sexual killing, although the rationale for this is not provided.

Sadism, the paraphilia most often associated with sexual killing (Brittain, 1970; Dietz, Hazelwood, & Warren, 1990; Langevin, Ben-Aron, Wright, Marchese, & Handy, 1988) is believed to be most prevalent within the lust killing category, as well as perpetrators who engage in post mortem sexual interference acts with the body, including mutilation of sexual body parts. Malmquist (1996) recognised that these headings do not clearly categorise all sexual killings. Distinguishing cases discretely can be problematic as there is so much variation in the manner and type of sexual behaviour, actual or intended that can be attached to the killing. For example, anger on the part of the perpetrator because the victim struck out during a sexual assault could trigger the killing, but the sexual assault itself could have been a sadistic rape. Beauregard and Proulx (2002) used cluster analysis to determine profiles in the offending process of SKs. They used the terms anger killers and sadistic killers to label the two profiles that emerged from their analysis. Although the anger killers experienced significantly more anger than the sadistic killers prior to the crime, there was some cross-over. Problems with anger could therefore be appropriate treatment targets for offenders from either offence profile. Because sex and killing can be attached in different ways and for different reasons, any definition of sexual killing adopted by staff performing forensic assessments should effectively identify possible cases where there is a sexual motivation or element to the offence. If this is the case, further assessment will often be required to confirm and/or establish the nature of the motivation and the nature of the sexual element.


For the purpose of this chapter we have defined SKs as homicide offenders that meet at least one of the following criteria; the perpetrator has disclosed that he had killed with a sexual motive or element to the killing, or there was evidence of sexual behaviour prior to or during the killing, or clothes were disturbed for reasons that could not be explained by simply moving the body1.

We have defined PMSIOs as homicide offenders whose offence contained at least one of the following characteristics, the perpetrator disclosed that he had sexually assaulted the victim after killing them, there was evidence from a pathologist of post mortem sexual behaviour, the perpetrator had disclosed post mortem sexual behaviour2, there was evidence of sex with an unconscious or dead victim or the perpetrator disclosed since conviction that they had sexually assaulted the victim after killing them. The criterion, “Police suspected post mortem sexual behaviour” was not included because of the possibility that it would result in an increased number of false cases. Table 1 shows how many cases met the inclusion criteria for PMSIOs from our first data base (See later).

Table 1. — Cases meeting the inclusion criteria for PMSIO.

Necrophilia is defined as a sexual attraction to corpses. We wish to emphasise that most PMSIOs are probably not necrophiles. Necrophilia is a sub-set of PMSIO offending. Necrophilia can also occur without a homicide taking place, as in the case of offenders who are employed in capacities that give them access to corpses. In this chapter, we have restricted ourselves to discussing offenders who committed homicide and subsequently sexually interfered with the body. Hence, the uses of the broader term Post Mortem Sexual Interference Offenders (PMSIOs), rather than the more specific term necrophilia.

Within research on SKs, rarely has any distinction been made between SKs who kill to eliminate the only witness to a crime and killers for whom the sex and the killing have a strong association. Despite this lack of distinction, research on SKs to date shows that they share more similarities than differences with other offender groups (Beech et al., 2005; Gratzer & Bradford, 1995; Langevin et al., 1988). On this basis, SKs have been treated alongside non-homicide sexual offenders within HM Prison Service and approximately 50% to 60% of the total number of known sexual killers within HM Prison Service are believed to have participated in sex offender treatment programs to address sexual aspects of their crimes (Oliver, Beech, Fisher, & Beckett, 2007). Prior to considering the merits of this approach, we will first consider the treatment needs of SKs and PMSIOs.

Identifying the Treatment Needs of SKs and PMSIOs: Information Sources

In the first stage of this chapter, we review the existing literature and our data base information to determine the most likely treatment needs of SKs and PMSIOs. We defined a treatment need as a stable psychological characteristic that appears relevant to the particular type of offending committed. In effect, this is a similar definition to the widely understood concept of a dynamic risk factor. However, the (relatively) small number of SKs and PMSIOs, the low rate of release, and the stringent monitoring that such offenders usually receive, precludes any empirical study large enough to determine whether these psychological characteristics are associated with increased criminal reconviction. Our information sources include a literature review of both types of offending, a data base of the characteristics of 100 SKs, and a data base of psychometric test scores, measuring dynamic risk factors for sexual offending, including data on some of the SKs (n = 19) and PMSIOs (n = 17) in the first data base.

Our first data base holds information coded from 100 life sentence prisoners who met the above definition of a sexual killer (n = 66) or PMSIOs (n = 34). The mean age of the SKs and PMSIOs was 27.9 and 25.3 years respectively. The ages ranged from 16 to 50 years for SKs and 17 to 43 years for PMSIOs. Overall, 55 of the perpetrators’ victims were considered to be strangers, while 45 were considered to have known the perpetrator prior to the offence. The victims were all adult females aged 14 years or above. Three of the perpetrators had been convicted of two sexual killings while 97 had been convicted of a single killing. Ninety-one of the perpetrators were born in the United Kingdom, 3 in Europe and 6 elsewhere.

Broadly, the coding criteria looked for the presence or absence of characteristics relevant to the following: childhood and family background of the perpetrator; schooling; experiences of trauma and problem behaviour; adult characteristics of the perpetrator such as criminal behaviour, drug and alcohol use, psychiatric contact, work and relationship status; characteristics of the perpetrator’s victim including their living circumstances and relationship to the perpetrator; the offence characteristics, including access to the victim, crime scene information; and post offence information, including apprehension and prosecution. The data collection allowed for additional coding options, such as whether the coder suspected that an item was possibly present in the absence of firm evidence. Table 2 shows a summary of factors from this data base reported on in this chapter.

Ten of the 100 cases were selected randomly and scored independently to estimate inter-rater reliability. There were 531 item codings that were considered for each of the ten cases randomly selected. Cohen’s kappa was used to determine reliability with Fleiss’ (1981) criteria employed to assess the level of agreement. Kappas between .4 and .6 are considered fair, kappas between .6 and .75 are good and kappas above .75 are excellent. The inter-rater agreements ranged from a total percentage agreement of 71.9%, kappa = .46 (fair) to total percentage agreement of 84.2%, kappa = .69 (good). Overall, the cases were split equally between good and fair kappas.

Our second data base held at Her Majesty’s Prison Service Headquarters consists of information on all those who have completed a Sex Offender Treatment Program (SOTP). This data base holds SOTP participants’ demographic details, offence details, and pre and post treatment psychometric scores. A battery of psychometric measures is administered to all SOTP participants before and after treatment to identify risk factors and monitor progress in treatment. The psychometric scores of all those in the first data base who had completed a SOTP were extracted from this data base.

Table 2. — Childhood, adult and crime factors post mortem sexual interference sexual killers (PMSI SKs) vs. non post mortem sexual interference sexual killers (Non PMSI SKs).

* p < .05

To interpret the psychometric scores of the two groups of interest, we compared the two groups against each other, and also compared them against a “normative” group of low risk offenders as measured by Risk Matrix 2000 (Thornton, Mann, Webster, Blud, Travers, Friendship et al., 2003), untreated, sexual offenders, the vast majority of whom had not committed homicide. The normative group mainly comprised older, incestuous or first-time offenders, without a previous criminal record. This group has limitations as a normative group, but in our view it is meaningful to be able to say on which characteristics a sexual killer, or a PMSIO, differs significantly from the average low risk sexual offender (Thornton et al., 2003). If a sexual killer or a PMSIO scores worse than an average untreated low risk sexual offender (for the purpose of this chapter, “worse” means more than one half of a standard deviation away from the mean in the undesired direction), it is suggested that this variable may represent a particularly problematic area for that offender4. Graphs in Appendix 1 show the average scores for 19 SKs and 17 PMSIOs on a range of psychometric measures, compared to the average score for 644 adult male (mean age 48.6 years) low risk sexual offenders. This low risk group is treated as the normative group for comparison purposes. On the graphs, the normative scores for each psychometric measure are converted to Z scores and then standardized so that 50 is always the mean score of the normative group for each measure, and each 10 points away from the mean represents one standard deviation. For comparison purposes, medium, high and very high risk offenders’ mean scores are also provided.

When discussing treatment recommendations, we have also drawn heavily on our experience in treating SKs, both in standard sex offender programs where they are mixed with non-homicide offenders, and in a specialist sexual murderer treatment unit (Clarke & Carter, 2000).

Treatment Needs of SKs and PMSIOs

As a framework for our information review, we chose to search for evidence of the widely accepted dynamic risk factors for sexual offending as outlined by the two most well-known dynamic risk factor tools: SARN (a variant of Structured Risk Assessment; Thornton, 2002) and STABLE-2000 (Harris & Hanson, 2000). During the course of our information review, several additional potential treatment needs emerged. In the review that follows, terminology from both SARN and STABLE-2000 has been adopted (these are generally heavily overlapping systems but which use slightly different language) in order that our conclusions may be accessible to staff working in both British and Canadian jurisdictions.

Sexual Preoccupation

Sexual killers

There is little direct reference to sexual preoccupation in the literature on sexual killing. Prentky, Burgess, Rokous, Lee, Harman, Ressler et al. (1989) found that 25% of the serial killers they studied disclosed compulsive masturbation. In addition, an interest in pornography has been found in a number of SK studies. Dietz et al. (1990) reported that 53% of their sample of sadistic offenders, which included SKs, kept pornography and 27% bondage items. Ressler, Burgess, Hartman, Douglas and McCormack (1986) reported frequent use of pornography by 38% of their sample of SKs, who were predominantly serial perpetrators, while 81% maintained some interest in pornography. This high pornography use could be indicative of sexual preoccupation. Blanchard (1995) reported that sexual preoccupation was one of the topics that frequently came up in his interviews with SKs. MacCulloch, Snowden, Wood and Mills (1983) reported a “substantial increase in masturbatory activity” (p. 25) when sadistic offenders in their sample, including SKs, switched from non aggressive fantasies to those of a sadistic nature. Briken, Habermann, Kafka, Berner and Hill (2006) investigated the relevance of paraphilia related disorders (PRDs) in a sample of 161 SKs. They looked for the presence of compulsive masturbation, promiscuity, pornography/telephone dependence or severe desire incompatibility5. Using these criteria, they found PRDs to be present in just over half their sample. In addition, those SKs who were considered to have paraphilias as well as PRDs were significantly more likely to be given a diagnosis of compulsive masturbation and pornography dependence than SKs with only paraphilias.

Unfortunately we do not have the relevant psychometric data on this variable. The relevant psychometric scale, the Sexual Preoccupations subscale of the Multiphasic Sex Inventory (MSI, Nichols & Molinder, 1984), was introduced into our test battery after our current sample underwent assessment. In summary, the literature indicates that sexual preoccupation is evident, by the presence of PRDs, for a proportion of SKs, particularly if they have a diagnosis of paraphilia. Sexual preoccupation would seem to be a treatment need for a sub-set of SKs.

Post mortem sexual interference offenders

The literature notes that hypersexuality is a feature of some PMSIOs (Dimock & Smith, 1997). Similarly, Rosman and Resnick (1989) gave the example of “the need to perform limitless sexual activity” as a “less commonly reported motive” for sexual interference with a corpse. Unfortunately, as described above, we do not have the relevant psychometric data on this variable. With little hard data, it is not possible to conclude that sexual preoccupation is clearly a feature of PMSIOs. We therefore advocate that this issue continue to be explored.

Sexualized Violence including Sadism

Sexual killers

As stated earlier, sadism is the paraphilia most often associated with SKs. MacCulloch et al. (1983) established that 81.3% of the 16 psychopathic, personality disordered patients (seven of whom had killed) in a special hospital had formed fantasies that matched all or some elements of the index offence. Both at the time of the offence and preceding it, the patients had typically “been masturbating to fantasies of sequences of behaviour which included rape, flagellation, anaesthesia, torture and killing” (p. 23). Five of the 13 cases who formed these sadistic fantasies killed their victim. Brittain (1970) proposed that cruelty causes excitement for SKs, and could be drawn from sources such as books or fantasy. A release of sexual pressure is the goal of the great “sexual emotion” they glean from this excitement. Briken et al. (2006) found that sexual sadism (using DSM-IV criteria) was the most prevalent paraphilia (37.3%) in a sample of 161 SKs. Langevin et al. (1988) reported significantly higher rates of sadism (from puberty or earlier) among 13 SKs when compared to a group of 13 sexual aggressors who had not killed or a group of 13 non-sex killers. In addition, they found that phallometric testing indicated sadism for sexual killers at a greater rate than in these comparison groups. Firestone, Bradford, Greenberg and Larose (1998) reported significantly greater levels of sadism, using DSM diagnosis, in their sample of 17 homicidal child molesters compared to 35 child molesters who had not killed. Yarvis (1995) found that his group (n = 10) of rapist/murderers had “an extraordinarily high prevalence of sexual sadism” (p. 418) in comparison to murderers and rapists.

Our sexual killer data base indicated that 41.2% of non-PMSIO sex killers had attacked their victim with what seemed to be a sexual intention6, as opposed to 22.7% of PMSIO SKs. Signs of sexual intention to stab wounds7 and sexual intention to cutting/incision8 wounds for non-PMSIO sex killers were 10.6% and 12.1% respectively as opposed to 2.9% and 0% respectively on these items for PMSIO.

Strangulation has been associated with sadism. Schlesinger (2004) proposed, “The primacy of strangulation in sexual homicide, previously reported by Revitch and Schlesinger (1989), occurs because the offender can “control” the length of time necessary to cause death and concomitantly increase his gratification” (p. 239). Our data base showed that SKs took the lives of their victims by ligature or manual strangulation 37.9% of the time, although strangulation was involved in 61.6% of cases. Unfortunately we do not have relevant psychometric data on this variable. The relevant psychometric sub-scales from the Multiphasic Sex Inventory (MSI, Nichols & Molinder, 1984) were introduced into our test battery after our current sample underwent assessment. In summary, although formal diagnosis has not always been employed, sadism has been shown to be present in a number of studies of SKs and would seem to be a common feature for a large number of SKs.

Post mortem sexual interference offenders

Sadism is usually taken to refer to a sexual interest in causing pain or humiliation to others. It could therefore be argued that sadists require a living victim in order to gain sexual pleasure, as a dead victim no longer experiences pain or humiliation. Sexualized violence is a somewhat broader term, referring to the capability for sexual arousal in the context of violence. The question of whether or not PMSIOs are likely to be sadistic has been said to be a “topic of considerable disagreement” (Hucker & Stermac, 1992, p. 243). It seems that true necrophiles are not necessarily likely to be sadistic – in fact, they often choose to hug, fondle, kiss and sleep with their victims. However, some true necrophiles have been reported to show sadistic features. Rosman and Resnick (1989) reported on the largest study to date of PMSIOs. In their sample 52% of the true necrophiles were thought to be sadistic. Non-necrophiliac PMSIOs are probably more likely to be sadistic. Such offenders have been termed “necrosadists”, “lust murderers”, or “erotophonopiliacs” (Purcell & Arrigo, 2006). But even non-necrophiliac PMSIOs are not necessarily sadistic – this group also includes opportunistic offenders and those with a transitory attraction to a corpse.

Our data base showed that PMSIOs killed their victim through ligature or manual strangulation 50.0% of the time and strangulation was involved in the offence 73.5% of the time. If strangulation is indeed an indicator of sadism, this suggests a high proportion of PMSIOs are likely to have sadistic interests. In summary, sadism is probably present in a sub-set of PMSIOs, but PMSIO may also occur because of true necrophilia, opportunism, or transitory attraction to a corpse. Obviously it is a particular challenge in any individual case to ascertain whether sadism – defined as long-standing presence of masturbatory fantasy about violence, killing, pain, fear and humiliation – is a feature of the PMSIO.

Other Offence-Related Sexual Interests (paraphilia)

Sexual killers

SKs have been reported to suffer from multiple paraphilias. Brittain (1970) described how the sadistic killer often had a history of voyeurism and cross-dressing. A paraphilic interest in peeping, obscene phone calls or indecent exposure has been found in a number of studies of sadistic offenders whose samples have included SKs. Dietz et al. (1990) reported paraphilic interest in 20.0% of their sample, Gratzer and Bradford (1995) reported 42.0%, while Warren, Hazelwood and Dietz (1996) indicated 45.0%. Langevin et al. (1988) reported voyeurism in 54.0%, exhibitionism in 23.0% and toucherism and frottage, in 31.0% in their sample of 13 SKs. Ressler et al. (1986) reported a predominance of voyeurism (71.4%) as well as autoerotic practices (78.6%) in their sample of serial SKs. Two studies noted the presence of known cross-dressing; Dietz et al. reported 20.0% and Gratzer and Bradford, 39.8%. Firestone et al. (1998) found that 22.9% of their sample of homicidal sex offenders displayed atypical paraphilias and paedophilia. Briken et al. (2006) reported that just over half of their sample of SKs showed evidence of DSM-IV paraphilia. In our SK data base, we observed that both non-PMSIO SKs and PMSIO SKs had reported rates of about 23.0% evidence of paraphilia9. In summary, paraphilia would seem to be a treatment need for a small sub-set of SKs.

Post mortem sexual interference offenders

True necrophilia is defined as a Paraphilia not otherwise specified in DSM-IV-TR. That is, necrophilia is recognized as a paraphilia in its own right, but does not come to professional attention often enough to warrant full diagnostic criteria. Other than sadism the literature does not associate any other particular paraphilia with PMSIO. PMSIOs who reveal a conscious, long-standing, sexual attraction to corpses are few. Even among non-sadistic PMSIOs who have full intercourse with corpses, Rosman and Resnick (1989) reported more common motives for PMSIO being intimacy or reunion with the victim. These motives might be classified as involving transitory rather than longstanding attraction to corpses. In conclusion, paraphilia in the form of necrophilia could be considered a treatment need for a sub-set of PMSIOs.

Hostility toward Women (adversarial sexual beliefs; beliefs that women are deceitful)

Sexual killers

In terms of motivation for sexual killing, hostility toward women has often been suggested as a risk factor (Langevin et al. 1988). Brittain (1970) described how the sadistic murderer, “has a fear of adult contacts, both social and sexual, and some even have an active hatred of all women” (p. 200). The research to date would not however suggest that hostility toward women is predominant for all SKs. Langevin et al. found more of the SKs in their sample were angry at the world (38.0%) than at women specifically (23.0%). Beech et al. (2005) found that 79.0% of the 28 SKs they studied showed evidence of having a dangerous world implicit theory (or schema) whereby they see the world as dangerous and react to this by being dominant and chastising those they perceive to have caused them harm. A sub-group of these perpetrators (n = 8) who were characterized as subscribing to this schema, generally “Reported that their motivation to offend was grievance driven because of anger and resentment toward women” (p. 1381).

In our sexual killer data base, 18.2% of the non-PMSIOs SKs were coded as having evidence of grievance toward females (compared to 26.5% of PMSIOs). Our psychometric data showed that SKs’ average score on a scale measuring the extent to which women are seen as deceitful was close to the normative score for low risk sexual offenders. In summary, although the psychometric data did not indicate SKs see women as deceitful, considering the indications from other sources of information, hostility toward women would seem to be a possible motivation for sexual murder and hence, a treatment need for a small sub-set of SKs.

Post mortem sexual interference offenders

Holmes (1991) reported that “many necrophiliacs are insensitive to others and have a great hatred for women” (p. 60) although no data were given in support of this assertion. Our psychometric data showed that PMSIOs’ average score on a scale measuring the extent to which women are seen as deceitful was close to the normative score for low risk sexual offenders. As mentioned above, PMSIOs were more likely to have been coded as having evidence of grievance toward women than were SKs (26.5% compared to 18.2%). We suggest this is an area worthy of further study.

Sexual Entitlement

Sexual killers

There is little reference to sexual entitlement in the literature on SKs. One notable exception is Beech et al. (2005), who found that 43.0% of their sample of 28 SKs ascribed to entitlement implicit theories. The concept of entitlement included believing they could procure sex on the basis they were male but was not limited to entitlement thinking in the sexual arena. Our psychometric data did not show sexual entitlement to be a notable feature of SKs compared to a normative mean score for low risk sexual offenders. Currently this would not appear to be a feature of SKs, although given the finding by Beech et al., general entitlement thinking would warrant further exploration with larger sample sizes.

Post mortem sexual interference offenders

Sexual entitlement has not been reported as a relevant issue for PMSIOs in the literature. Our psychometric data did not show sexual entitlement to be a notable feature of PMSIOs when compared to low risk sexual offenders in general.

Personal Inadequacy

Sexual killers

Brittain’s (1970) clinical description of the sadistic murderer painted the picture of a rather pitiable individual, distinct and remote from others who suffered insecurity and perceived himself inferior, including sexually inferior to other men and unable to relate to people. Blanchard (1995) found that the SKs he interviewed described beliefs that were a variant of those proposed by Carnes as being evident for almost all “sex addicts” and included, “I am basically a bad, unworthy person” and “No one would love me as I am” (1983, cited in Blanchard, 1995, p. 64). It is not clear how many of the SKs that Blanchard interviewed could be considered serial killers. Schlesinger (2004) provides case examples of acute catathymic10 sexual killings “triggered by sexual inadequacy” (p. 138). Grubin (1994) found that 34.0% of the 21 SKs he studied had experienced a recent loss of self-esteem prior to the killing.

In our psychometric data base, the SKs revealed self-esteem scores that were very close to the medium risk sex offender group. This construct includes feeling ashamed and unhappy about yourself. However, the mean score was not at a level where it would be considered a problematic feature of SKs. The SKs also revealed significantly greater emotional loneliness at the time of the offence. This construct included feeling that no-one shares or understands your feelings or values your worth as a person. On this measure, the SKs’ scores were significantly higher than the low risk sex offender group norm. Taken together, personal inadequacy would seem to be a treatment need for a sub-set of SKs.

Post mortem sexual interference offenders

Schlesinger (2004) presented compelling case examples that illustrate this feature of PMSIOs. In one case, an offender who felt inadequate and angry at being taunted by a victim could only penetrate her once she was lifeless. Another offender is described as having “dominant features of insecurity, inadequacy, and low self-esteem… tremendous self-hatred and inferiority … and [a fear] of appearing stupid to others” (p. 141). Tardif, Daaylva and Nicole (2007) also presented a PMSIO case example with strikingly similar features, who’s “life was a constant stream of frustration, stemming from conflict with women, low self-esteem and feelings of rejection”. This offender first killed “when a prostitute with whom he was to have sex ridiculed him in front of a taxi driver, calling him an easy client” (p. 226). In further support of the potential relevance of this psychological feature of PMSIOs, Rosman and Resnick (1989) reported that 12.0% of their true necrophiles were attempting to gain self-esteem by expressing power over a homicide victim.

In our psychometric data base, the PMSIOs did not reveal particularly low self-esteem, but they did report significantly greater emotional loneliness at the time of their offence than the average untreated low risk offender. This construct includes feeling that no-one shares or understands your feelings or values your worth as a person. On this measure, the PMSIOs’ scores were almost one standard deviation above the low risk sex offender group norm, and were in fact higher than the very high risk sex offender group mean. This represents a very elevated score for emotional loneliness suggesting that this is a significant problem for PMSIOs. Taken together, this would indicate that personal inadequacy is a treatment need for a sub-set of PMSIOs although further research is needed to establish if it is relevant to the majority of these perpetrators.

Grievance Thinking (negative emotionality)

Sexual killers

Brittain (1970) described the clinical features of the sadistic murderer whose inadequacies were evident from his poor ability to relate socially to others. This left him feeling insecure and set apart from people. He frequently experienced impotency on the occasions he did have sexual relationships with the opposite sex. Grubin (1994) found that SKs were significantly more likely to “bottle up” their anger “before exploding, perhaps reflecting a tendency for overcontrol” (p. 625). Beech et al. (2005) found that grievance was the principal motivation for a minority of the 28 SKs in their study (28.6%), stemming from feelings of anger and resentment toward women. Our data base showed that 16.7% of the non-PMSIOs were coded as having evidence of general grievance compared to 23.5% of PMSIOs SKs. In summary, grievance thinking would seem to be a relevant treatment need for a small sub-set of SKs.

Post mortem sexual interference offenders

Grievance thinking has not been reported as a relevant issue for PMSIOs in the literature. Our psychometric data did not show negative rumination to be a notable feature of PMSIOs compared to low risk sexual offenders in general.

Lack of Emotional Intimacy (intimacy deficits)

Sexual killers

Brittain (1970) described the sadistic killer as socially isolated and sexually inexperienced. The scarcity of social and sexual contact with their gender of preference was pronounced in eight of the 13 SKs in Langevin et al.’s (1988) study. Grubin (1994) found that SKs were significantly less likely to have had intimate relationships with women, and were significantly more likely to have had no sex partner in the year of the offence than a comparison group of rapists who had not killed. Oliver et al. (2007), reported that sexual murderers (n = 58) were significantly less likely to be in a relationship at the time of the offence than a comparison group of rapists (n = 112). In addition to this, Briken et al. (2006), reported that only 26.7% were in a partnership at the time of the killing (n = 161) and a further 72.1% had never married.

In our data base, only 25% of SKs were coded as being married and with their wife at the time of the offence and a further 19.7% were living alone at the time. In addition, 51.5% were coded as being single, having never married or not having had a relationship lasting more than two years. In summary, lack of emotional intimacy would appear to be a treatment need for SKs.

Post mortem sexual interference offenders

There are some suggestions in the literature that this variable may be of relevance for this group. Rosman and Resnick (1989) reported that “the most common motive of true necrophiles was to possess an unresisting and unrejecting partner” (p. 158), with the second most common motive being reunion with a romantic partner (e.g. a man who has intercourse with the body of his deceased wife). Burg (1982) noted a number of cases who had undergone a serious rejection experience in early life, and suggested this could lead to the reasoning that a dead sexual partner “could not object to my company” (Brill, 1941, as cited in Burg, 1982, p. 246).

In our sexual killer data base, we observed that only 3 (8.8%) of the PMSIOs were married at the time of their offence, compared to 17 (25.8%) of the non-PMSIO SKs. In addition, 64.7% were coded as never having been married or lived with someone for at least two years. In our psychometric data base, the PMSIOs did report significantly greater emotional loneliness at the time of their offence than other groups. This construct includes feeling that no-one shares or understands your feelings or values or your worth as a person. On this measure, the PMSIOs scores were almost one standard deviation above the low risk sex offender group norm, and were in fact higher than the very high risk sex offender group mean. Furthermore, PMSIOs reported significantly lower levels of empathic concern for others, and poorer perspective taking, than low risk sexual offenders in general. On both measures, the PMSIOs showed more problematic scores than the very high risk sexual offender group. Taken together, these scores indicate quite severe problems with the kinds of traits that enable positive social or intimacy experiences.

Lifestyle Impulsiveness (general self-regulation problems)

Sexual killers

Although Ressler, Burgess and Douglas (1988) described an organized and disorganized sexual killer, the latter acting “impulsively under stress, finding a victim usually within his own geographic area” (p. 130), impulsiveness has not otherwise been reported as a relevant issue for SKs in the literature. Grubin (1994) reported no significant differences on a measure of impulsivity between SKs and rapists who had not killed. In addition, he did not report that this was a factor that characterized SKs.

The sexual killer data base showed that SKs used a weapon 28.8% of the time and of these, 84.2 % brought the weapon with them. While this could be an indication that the majority of perpetrators did not premeditate their crimes, it is also possible that there was another explanation (e.g. they favoured strangulation and could use either their hands or a ligature found at the crime scene). In summary, while impulsiveness would warrant further research, currently this would not seem to be a particular treatment need for SKs.

Post mortem sexual interference offenders

Lifestyle impulsiveness has not been reported as a relevant issue for PMSIOs in the literature. Our psychometric data did not show impulsivity to be a notable feature of PMSIOs compared to low risk sexual offenders in general, but they were significantly more impulsive than the non-PMSIO SK. The sexual killer data base indicated that the PMSIOs were very unlikely to have premeditated their homicide, with only 2.9% of PMSIOs taking a weapon to the crime scene, (compared to 24.2% of the non-PMSIO SKs) although, as discussed above, there could be another explanation for this. In summary, on current evidence this would not seem to be a treatment need for PMSIOs.

Poor Cognitive Problem Solving

Sexual killers

This has not been reported as a relevant issue for SKs in the literature. In our sexual killer data base, we observed similar rates of evidence of suicide or self-harm for non-PMSIO SKs and PMSIO SKs (22.7% and 23.5% respectively). In summary, poor problem solving would benefit from further research. At present it is possibly a treatment need for a small sub-set of SKs.

Post mortem sexual interference offenders

Dimock and Smith (1997) noted alcohol abuse as a recurring feature of many PMSIO case studies. Other than this, there has been little discussion in the literature about poor problem solving in PMSIOs. Unfortunately in our psychometric data base there was not a measure of problem solving ability. However, we observed that 50.0% of the PMSIO group had been coded as having an alcohol problem, compared to only 25.0% of the non-PMSIO SKs. If alcohol abuse is considered an indicator of poor problem-solving, this would indicate that PMSIOs may well have deficits in this area although this area would benefit from further research.

Poor Emotional Control

Sexual killers

Revitch (1965) proposed that acts of sadism were typically carried out by “either overt or latent psychotics with poor control and explosive breaks with reality” (p. 644) and as mentioned above, Grubin (1994) suggested that explosions of anger followed over-control of anger. Langevin et al. (1988) suggested that the numerous suicide attempts identified in their sample of SKs indicated unstable emotions. Schlesinger (2004) proposed that in catathymic crisis homicides, the “victim triggers underlying emotionally charged conflicts” (p. 137).

The psychometric data base revealed that SKs’ scores on a measure of impulsivity were almost two standard deviations below the low risk sex offender norm. Taken together, it would appear that the emotional control pattern for a sub-set of SKs seems to be typically over-controlled, with intermittent explosive outbursts.

Post mortem sexual interference offenders

This has not been specifically identified as a relevant issue for PMSIOs in the literature. However, the case studies cited above (see personal inadequacy section) suggest that, in a context of self-hatred, certain rejection experiences can trigger an overwhelming tide of emotion and in this state of mind a post mortem sexual interference offence can be committed. Our psychometric data base did not contain a measure of emotional control although it indicated that PMSIOs scores on a measure of impulsivity were not problematic when compared with low risk sexual offenders. It is not possible to conclude whether or not emotional dysregulation is a feature of PMSIOs and this area would benefit from further research, but their general apparent lack of premeditation, not apparently explained by impulsivity, indicates that emotional dysregulation should remain a credible hypothesis.

Social Isolation (general social rejection/loneliness)

Sexual killers

Social isolation has been identified as a characteristic of SKs in a number of studies. Grubin (1994) identified both social and emotional isolation across the lifespan as distinct features of a sexual murder group in comparison to a group of adult rapists. Milsom, Beech and Webster (2003) reported that SKs had significantly higher levels of loneliness during adolescence compared to a group of adult rapists. MacCulloch et al. (1983) found evidence that a group of sadistic offenders, including SKs, had difficulties “relating” to their favoured sex from early childhood. Compos and Cusson (2007) found that about half of the SKs in their study (n = 41) reported social isolation during adulthood (46.7%). Kennedy, Hoffman and Haines (1947) examined the case of serial killer William Heirens and found he experienced loneliness as a child. The data base showed that 38.2% of non-PMSIO were considered loners who did not socialize. Taken together, it would seem that social isolation is a feature of SKs.

Post mortem sexual interference offenders

This has not been addressed in the literature to date. In our data base, 40.0% of PMSIOs were considered loners who did not socialize. We discuss elsewhere that PMSIOs lack some of the essential skills for successful relationships, such as perspective taking and concern for others. On this basis, we suggest that social isolation is an area worthy of further consideration.

Lack of Concern for Others

Sexual killers

Langevin et al. (1988) found that five of the 13 SKs in their study failed to convey any feeling for their victim, while five did express feelings of guilt (the remainder expressed self concern or felt positive about the offence). One killer reported feeling positive about the sense of achievement he got from committing the crime. Tardif et al. (2007) emphasized the ability of sadistic murderers to dehumanise their victims. Marshall and Hucker (2006) described the same feature. Both Tardif et al. (2007) and Marshall and Hucker (2006) concluded that a major treatment target for sadists is to increase the extent to which they see other people as fellow human beings.

In our psychometric data base, SKs reported significantly lower levels of empathic concern for others than the low risk sex offender norm suggesting this may be a problematic area for SKs. Taken together, lack of concern for others would appear to be a feature for a large sub-set of SKs and an important target for treatment although it would benefit from further research in this area with larger sample sizes.

Post mortem sexual interference offenders

This has not previously been reported as a relevant issue for PMSIOs. As noted above, PMSIOs reported significantly lower levels of empathic concern for others, and poorer perspective taking, than low risk sexual offenders in general. On both measures, the PMSIOs showed more problematic scores than the very high risk sexual offender group. Lack of concern for others is therefore credibly hypothesized to be a treatment need for some PMSIOs.


Sexual killers

The literature suggests that psychopathy is a feature of SKs (Langevin et al., 1988; MacCulloch et al., 1983). Brittain (1970) described a number of characteristics of the sexual killer that could be considered psychopathic traits (e.g. he does not regret his offences, emotionally flat, and does not show emotions about his offences). However, Proulx and Sauvêtre (2007) rightly warn, “the few empirical studies that suggest that sexual murderers are psychopaths all suffer from methodological limitations” which include “psychometric instruments that are inadequate for the comprehensive evaluation of psychopathy” (p. 55). Firestone et al. (1998) found that sexual homicide offenders had significantly higher levels of psychopathy, as measured by the Psychopathy Checklist Revised (PCL-R, Hare, 1991) than non-homicidal child molesters (total score of 28.7 and 16.6 respectively), particularly on factor 1, which rates interpersonal and affective functioning (12.8 and 7.9). Porter et al. (2003) rated the majority (87.7%) of their sample of SKs as “moderate to high” on the PCL-R (where moderate was 20-29 and high 30 and above). They concluded that, “Not only are psychopathic offenders disproportionately more likely to engage in sexual homicide, but, when they do, they use significantly more gratuitous and sadistic violence” (p. 467).

The sexual killer data base indicated that 22.7% of non-PMSIO SKs were considered psychopathic as opposed to 17.7% of PMSIO SKs11. While further research is needed to replicate Porter et al.’s (2003) findings, particularly within the UK, taken together these studies indicate that psychopathy is a feature of a sub-set of SKs.

Post mortem sexual interference offenders

This has not been reported as a relevant issue for PMSIOs in the literature. There are no data on levels of psychopathy found in PMSIOs and this area requires research.

Brain Damage

Sexual killers

Langevin et al. (1988) found abnormalities in 40.0% (n = 10) of their SKs where killing was combined with sexual excitement. These abnormalities were generally in the right temporal horn 30.0% (n = 10). While this finding is consistent with a study of sadists (Hucker, Langevin, Dickey, Handy, Chambers, & Wright, 1988), these are small proportions within a very small sample of SKs. Further work is needed to establish if brain abnormalities contribute to either sadism (Proulx, Blais, & Beauregard, 2007) or sexual killing per se. Briken, Habermann, Berner and Hill (2005) analysed 166 SKs psychiatric court reports and found that 30.0% of these men had evidence of heterogeneous brain abnormalities. The findings also revealed that those SKs with brain abnormalities experienced more behavioural problems in childhood and were diagnosed with a greater number of paraphilias then SKs without brain abnormalities. They concluded that the findings, “Suggest the importance of a precise neurological and psychological examination of this specific offender group not only to evaluate responsibility but also for treatment and risk assessment” (p. 1207).

Among the 24.2% of SKs that suffered a head injury during childhood, 14.7% showed evidence of this having caused lasting damage. Taken together, brain damage would appear to be a feature for a sub-set of SKs although more research is needed to clarify what kind of brain damage.

Post mortem sexual interference offenders

This has not been reported upon in the literature on PMSIOs and requires further research.

Personality Disorder

Sexual killers

SKs have been found to have narcissistic features (Brittain, 1970; Dietz et al., 1990), schizoid, (Brittain, 1970), borderline (Geberth & Turco, 1997) and antisocial personality disorder and have been found to be distinguished from non sex killers and non-homicidal sexually aggressive men on the basis of antisocial diagnosis (Langevin et al., 1988). Yarvis (1995) found evidence of antisocial personality disorder in 90.0% of the rape/murderers in his study. Proulx and Sauvêtre (2007) reported “anti-social (35.7%), borderline (28.6%) and narcissistic (25%) disorders” (p. 60) in a sample of 40 non-serial sexual killers. They suggested that overlap between anti-social personality disorder, narcissism and psychopathy should be noted (Proulx & Sauvêtre, 2007). In summary, personality disorders could be relevant for a sub-set of SKs.

Post mortem sexual interference offenders

Rosman and Resnick (1989) reported that 56.0% of their total sample of PMSIOs had been diagnosed with a personality disorder. Of the sub-groups of PMSIOs who had killed, over 80.0% had been diagnosed with a personality disorder. Unfortunately Rosman and Resnick did not report the nature of the personality disorders diagnosed. While this would seem to be a relevant treatment need for a sub-set of PMSIOs, it would benefit from further research.

Treatment Need Analysis Conclusions

There are limitations to the data used in this study. The retrospective nature of the information, that the data had been collected with reliance on files containing reports and entries that are not consistently written, and that the inter-rater reliability of the report writers was unknown (Briken et al., 2006). The information used in the second data base is from a small sample and, as commented upon previously, there was an absence of relevant psychometric scales for some of the variables considered. While a warning on the methodological limitations of the data collection is warranted, we are able to reach preliminary conclusions about the treatment needs of these two types of killers.

Sexual killers

The main treatment needs that emerged from our multiple information sources were:

  • Sadism
  • Personal inadequacy
  • Lack of emotional intimacy
  • Social and emotional isolation

In some cases, sexual preoccupation, paraphilias, hostility toward women, grievance thinking, poor emotional control (overcontrol with intermittent outbursts of anger), brain damage, personality disorder and entitlement thinking may additionally be an issue.

Some areas that appear to be treatment needs but would benefit from further research were:

  • Poor problem solving
  • Lack of concern for others
  • Psychopathy12

Potential treatment needs have been inadequately researched and information was not available in our data base. These include:

  • Sexual entitlement
  • Lifestyle impulsiveness
  • Grievance thinking
Post mortem sexual interference offenders

The main treatment needs that emerged from our multiple information sources were:

  • Lack of emotional intimacy
  • Poor emotional control
  • Low concern for others and poor perspective taking ability

In some cases, additionally or alternatively, sexual preoccupation, and necrophilia may be an issue.

Some areas that appear to be treatment needs but would benefit from further research were:

  • Sadism
  • Sexual preoccupation
  • Extreme personal inadequacy (self-hatred)
  • Necrophilia and other paraphilias
  • Poor cognitive problem solving
  • Lack of concern for others
  • Personality disorder

Some potential treatment needs have been inadequately researched and information was not available in our data bases. These include:

  • Hostility toward women
  • Social isolation
  • Psychopathy
  • Brain damage13

Oliver, Beech, Fisher and Beckett (2007) proposed that future research to determine differences between rapists and sexual murderers on dynamic risk factors could, if they were to be identified, require changes to treatment approaches for SKs. We would propose that there already exists a basis to amend existing programs for sexual offenders to make them more applicable to SKs. These changes are needed along with an emphasis on staff training, implementation, integrity and support issues.

Treatment Involvement

The existing literature is minimal and divided. Tardif et al. (2007) pointed out that “current sexual offender treatment programs have not been developed with sexual murderers in mind” (p. 222) and concluded that “There is a real danger in extrapolating positive treatment results from other groups of sexual offenders to sexual murderers”. On the other hand, Beech et al. (2005) reported quite considerable benefits observed in sexual murderers who participated in Her Majesty’s Prison Service’s Sex Offender Treatment Program alongside non-homicidal sexual offenders14. In addition, Beech et al.’s (2005) study of sexual murderers’ implicit theories (ITs) reported, “The finding that no new and distinct ITs were found in the sexual murderers” indicated “that they are not qualitatively different from rapists in terms of the underlying schemas they have about the world” (p. 1385).

In our view, it is possible to amend and re-focus traditional cognitive behavioural sex offender treatment programs to make them more relevant to the needs of sexual murderers. Our literature review and data suggest that out of the range of currently-accepted treatment targets for sexual offenders, a fairly identifiable sub-set of targets seem to apply to sexual murderers. Our experience in treating sexual murderers in a regular sex offender treatment program further indicates that as long as the program offers flexibility to focus on particular areas as needed, a different program does not seem to be necessary. The overlap in treatment needs between sexual killers and non-homicide sex offenders is substantial, and of course it is also the case that non-homicide sex offenders are a heterogeneous group. Marshall and Hucker (2006) suggested six important adaptations that could be made to a typical sex offender program in order for it to be suitable for sadistic offenders. Their suggestions make excellent sense with respect to sexual murderers and hence, we repeat them here:

  1. When offenders describe their offending – the typical starting point for most sex offender treatment programs – ensure they do not provide the details of the sexual, violent and sadistic elements of their offending.
  2. Challenge every expression of beliefs that favour violence.
  3. Adapt victim empathy work to focus on the victim’s post-offence suffering, rather than examining their suffering during the offence. Opinions differ about the value or danger of discussing with a sadist how their victim suffered during the offence. However, Marshall and Hucker made a strong case that the sadist was not intending to exert control outside of the offence itself, so discussion or evocation of the victim’s post offence suffering is likely to reduce the offender’s dehumanization of the victim15.
  4. To particularly focus skills-building work on skills that would meet the offender’s need for control in a pro-social way.
  5. To encourage greater emotional expression, as sadists tend to over-regulate their emotions.
  6. To make fantasy modification an essential element of treatment.

Tardif et al. (2007) also considered special elements of treatment for sadistic sexual murderers. They particularly recommended that treatment should not focus on sexual motivations such as sadism until later stages of incarceration. Earlier in a prison sentence, interventions should focus on three key objectives: (1) to humanise all interpersonal relationships; (2) to encourage acceptance of the real world in preference to the previously-preferred fantasy world; and (3) to develop self-esteem that is not oriented toward omnipotence (i.e. not dependent on achieving domination over others). For angry, rather than sadistic, sexual murderers, an additional short-term objective is to offer support in dealing with the emotional consequences of the offence; and in the long term, to teach the offender to better manage situations where he feels narcissistically affronted, particularly involving women.

Beech et al. (2005) suggested that the key to successfully treating sexual murderers is to recognize the different possible motivations for the homicide, and to adapt treatment accordingly. Grievance-motivated offenders should take a treatment pathway that emphasizes anger control and schema-management. Sexually motivated offenders should focus on their views of women as sexual objects. Sadistic offenders need, as a priority, behavioural modification work to improve their engagement in healthy sexual fantasies, and to learn strategies to control (or ideally, eliminate) sexualized violence fantasies.

In this chapter, we have only considered the psychological needs of sexual murderers. Marshall and Hucker (2006) strongly argue that psychological intervention alone is insufficient. Medication, such as anti-androgens (e.g. cyproterone acetate), testosterone-reducing medication (e.g. medroxyprogesterone acetate) or SSRIs (e.g. fluoxetine), should augment psychological treatment. For those SKs or PMSIOs who suffer from sexual preoccupation, sadism or necrophilia in particular, pharmacological therapy should most definitely be considered (see Bradford, this volume).

Assessment and Management

Beech et al. (2005) revealed a range of implementation issues that likely affect the success of therapy with this difficult client group.16

  1. The needs of female therapists should be carefully monitored and considered. Male therapists may need additional training in recognizing and responding to attempts to covertly dominate female staff and in presenting as excellent male role models. Female therapists reported experiencing both overt and covert hostility and sexualization from sexual murderers in treatment. The expressions of hostility are at their most covert when in the presence of male therapists, who often missed the expressions entirely. Tardif et al. (2007) noted a similar phenomenon with this particular client group, where “Typically, therapists… feel intruded upon and violated with regard to their personal boundaries – in short, a victim-aggressor relationship is initiated” (p. 219).
  2. Senior management support is essential. For specialist treatment to be successfully implemented, the senior managers in a correctional service must agree that the resource should transcend local and regional considerations.
  3. Treatment should be delivered and supported by multi-disciplinary teams, equally representing both genders.
  4. Treatment should start early in the incarceration period, and managers should be prepared to expect that in many cases, treatment may need to be repeated before “the penny drops” (reaching an understanding about something they have struggled to comprehend or have misunderstood for a period of time). As Tardif et al. (2007) note, sexual murderers often do poorly in treatment because they are “so invested in their position of omnipotent aggressor that they remain completely opposed to the idea of exploring their inner world” (p. 219). However, appearances can be otherwise: “Sexual murderers with good cognitive abilities are quite capable of showing successful integration of treatment” (p. 224). This appearance of psychological healthiness can be due to faking in some cases, or in others can reflect the fact that many murderers do not show signs of pathology prior to committing a homicide.
  5. Perhaps for the reasons described in the above point, treatment should take place only where objective assessment processes are available. Self-report, including psychometric questionnaires, are too subject to distortion. Phallometric testing, polygraphy, and some newer methodologies are all possible approaches.

In addition to the recommendations above, we would suggest that there are a number of other steps that could usefully be taken to help ensure traditional programs are applicable for SKs.

Determining if a killing was sexually motivated or had a sexual element is crucial for planning, for making assessments and implementing treatment. The difficulty often experienced when trying to establish if a killing was sexually motivated or had a sexual element could be helped by ensuring that staff who work with these men have an opportunity to develop their knowledge in this area. Allowing staff to focus on sexual killing as a core part of their forensic practice could enable them to more quickly develop skills in these areas. Research resources should also be made available to research SKs sub-groups (e.g. those who victimize children, those who victimize men, those who kill following a rape to get rid of the witness to the crime). This research information could then be used to educate staff, helping their ability to fulfil their specialist roles and help inform assessment issues, the planning and development of treatment pathways.

A sound understanding of the issues relevant to sexual killing is necessary for staff undertaking assessments of progress in treatment including the perpetrators ability to generalize skills learned outside of the treatment room. In terms of staff undertaking risk assessments on SKs, to help ensure objectivity, staff should avoid undertaking this role if they have been directly involved in the treatment of the individual concerned. All staff working with this client group must develop some level of knowledge or expertise in this area. This could be met through providing specialist training on relevant areas tailored to the expected level of their involvement. Staff not directly involved in delivering treatment would not require training in facilitation but could benefit from some training on the client group as a whole (e.g. the extent that social isolation is an issue in their lives, that they present as hostile to women, presence of a large number of psychopathic traits). All of this would help staff effectively engage with prisoners, understand problematic behaviour and recognize areas that would be difficult for these men to address and change.

Psychopathy is a relevant issue for a sub-set of SKs and it would be expected that psychopaths would be present in treatment programs for these men. Strategies for dealing with difficult and psychopathic prisoners need to be well thought through, robust and carefully implemented to prevent a staff versus prisoner culture emerging. The litigious behaviour of many life sentence prisoners, amplified by requiring them to undertake a range of assessments and challenging treatments, can put pressure on staff and disrupt therapeutic alliances. Transparent, open working relationships can, to a certain extent, reduce offenders’ desire to take legal redress. Equally important would be that staff are trained and able to deal with the kinds of problems and pressures that can arise when working with SKs, many of whom are being asked to undertake work related to sex offending but whom are serving an index offence for murder or manslaughter without a sexual conviction. Ideally, legal issues and correspondence should be the responsibility of someone not directly involved in treatment so that these functions can be kept separate. The tension that could arise when staff are being asked to balance an open, collaborative and therapeutic approach with prisoners while remaining objective and guarding against susceptibility to being manipulated needs to be recognized, discussed and monitored. This could be particularly felt by staff who have a disciplinary function as well as a therapist role.

Given the limited existing research with SKs, and that acting on some of the issues and recommendations we have outlined could be challenging, we would suggest that a forum or network is established for staff in this field to share understanding, get support, advice and develop best practice. An important question is whether sexual murderers should be integrated into regular sex offender treatment programs, or whether (and to what extent) they should be treated separately. There are three possible options, all of which we consider to be acceptable, but with each having their own advantages and disadvantages:

  • Treat sexual killers in regular sex offender treatment programs alongside other non-homicide sex offenders,
  • Treat sexual killers separately from non-homicide sex offenders, but locate them together with non-homicide offenders,
  • Locate and treat sexual killers separately.

We would suggest that given the research to date on SKs that they share more similarities than differences with rapists and, they can be placed in groups with other SKs and/or rapists.

Clarke and Carter (2000) argued there were advantages in locating SKs together to undertake treatment. Table 3 outlines advantages and disadvantages of a specialist unit for SKs which we will now discuss.

Advantages of a Specialist Unit for SKs

As covered above, the assessment of SKs poses a number of difficulties. The provision of a specialist unit for SKs would lend itself well to the development of expertise and knowledge about the assessment and treatment of these perpetrators, which could be disseminated to other forensic settings. Running groups in specialized units with programs adapted for SKs would present an opportunity to quickly act upon experience and feedback, to improve treatment delivery, and to incorporate emerging research with this client group. It would also provide an ideal environment for staff to develop specialist roles in assessment and treatment. Staff support and positive therapeutic alliances are important ingredients in successful interventions (Beech et al., 2005). A dedicated unit would help ensure that awareness and training would be available for all staff working with SKs. Practically, such a unit would be well placed to provide specialized assessments such as penile plethysmography. As staff develop expertise in assessment and treatment of this client group, it would be hoped that they would feel more confident and able to defend their work and deal with the pressures that can arise. Sending out a message that a service is providing an opportunity for SKs to get treatment and meet targets for parole via a discreet unit could counter some of the scepticism that SKs, their families and legal teams can encounter.

Table 3. — Undertaking assessment and treatment with SKs in a specialist facility: Possible advantages and disadvantages


Disadvantages of a Specialist Unit for SKs

A dedicated unit for SKs in a particular establishment needs to be supported nationally by the correctional system. Otherwise the practical difficulties in terms of transferring prisoners in and out of a unit in a timely fashion to ensure that assessment, treatment and progression targets are reached could be compromised. If prisoners are to be expected to move away from families and friends to undertake treatment, a realistic time frame is necessary to prevent frustration or ill feeling on the individuals’ part. Allowing prisoners breaks from treatment to go to an establishment nearer family for accumulated visits would only work efficiently if transfer and return was organized promptly with the support of prisons nationally. A dedicated unit with larger numbers of SKs and potentially, greater numbers of psychopathic individuals could increase the likelihood of both staff and prisoners being seen as elitist within the prison. The concentration of this client group in one place and the potential pressures this work brings could increase the likelihood of staff verses prisoner culture developing. The hearing process for lifers, where prison staff are often called to defend their reports in adversarial parole board panels, is time consuming and demanding and again can hinder therapeutic alliances.

The litigious behaviour of life sentence prisoners mentioned above, could be increased by housing these men together. Our suggestions on open working relationships and separating therapeutic input from involvement with legal responsibilities as much as possible would be important to maintain a therapeutic and constructive environment.

Whether or not SKs are assessed and treated in a discrete unit, we would suggest that the issues outlined above could usefully be considered for interventions with these men. In our experience, it is quite common to find that men who we would consider to have committed a sexually motivated killing have not been convicted of any sexual offence, often because this aspect of the crime was not been pursued by the courts. Asking these men to consider participating in a Sex Offender Treatment Program when they are not serving a sentence for a sexual offence, and can be in denial of any sexual aspect, can work against constructive engagement and result in the involvement of their legal representatives. Treatment providers will need to determine whether there should be a policy that participants must accept the sexual element of their offence prior to coming into the course or whether the group could be used to facilitate this kind of admission. If the latter stance were taken, strategies to help offenders out of denial would need to be considered in addition to creating an environment where participants are able and feel safe to talk about all aspects of sexual killing including post mortem interference.

For those offenders diagnosed with a personality disorder, we would suggest some kind of psycho-educational work aimed at raising personality disorder awareness. This could assist prisoners to develop insight into personality disorder, how it has possibly impacted their life, relevance to offending and to undertake work to lead a more fulfilling and offence free life.

Provision also needs to be made for the responsivity needs of these individuals and again, a discrete unit does lend itself to a more innovative and flexible approach. For example, psychopathic offenders who are prone to boredom and in need of stimulation may benefit from being placed in smaller groups, so that they get through the course more quickly and are not expected to focus on other prisoners’ needs for long periods of time. Ensuring that sessions do not last for more than an hour to facilitate concentration could also make dynamics easier to manage. These measures could of course be taken in the absence of a specialist unit.

Step Down Provision

Consideration for step down provision is paramount to both ensuring that prisoners are supported and encouraged to try new skills and different ways of thinking as well as helping staff to establish if new skills to deal with problematic and risky behaviours are being generalized. Simply looking for the absence of problematic behaviour, particularly given that some of these SKs will be overcontrolled or manipulative, is probably insufficient. The presence of positive attitudes and behaviours to deal with unhelpful thinking and behaviour should also be looked for. For example, for perpetrators who disclose hostile attitudes to women, any indication that they are developing positive attitudes to women both in a professional or personal capacity that are demonstrated across time and during periods of stress would be important indicators for risk assessment reports. In addition, some of the treatment needs identified in this chapter could be important areas to monitor after the prisoner has been released on parole, such as how they are dealing with alcohol use or taking measures to ensure that they do not become isolated. Of course, problems in these areas would act as signals to intervene and in certain circumstances, could trigger return to prison.

Consideration given to progression establishments will generally involve the offender moving through the system via decreasing levels of security until release. Progression and community-based staff should receive awareness training about SKs and related issues as described above. A receiving establishment where there were no staff with specialist knowledge on SKs should be avoided if we are to safely progress individuals and meet public protection responsibilities.

Community Management

Staff involved in the supervision of SKs and PMSIOs released on parole should receive training to work with these men as we have suggested above and would also benefit from having a forum to discuss concerns. Our observation that sexual murderers and PMSIOs may be particularly affected by social and emotional isolation emphasizes the need for a high level of social support to be provided after release. Statutory agencies and supervision arrangements usually do not extend to offering a social life, and the offender may doubt the level of personal commitment in the support being offered. For this reason, we strongly support initiatives such as Circles of Support and Accountability (COSA), and note the very promising outcome data of COSA in reducing reconviction rates among the most notorious, high risk cases (Wilson, Picheca, & Prinzo, 2005).

Future Research

We have identified a number of dynamic treatment needs as being under researched. Additional research would bring increased understanding to this crime and ultimately improve interventions. Further work on identifying the sub-groups of SKs could prove fruitful in delineating dynamic treatment needs and tailoring treatment to suit the spectrum of motivations for the fusion of sex and killing. In this chapter we have considered PMSIOs as a sub-group of SKs and while similarities exist, some potential differences in treatment needs have emerged.

We are aware that it is always very difficult to establish why an individual actually killed someone and in what way sex was attached. This is a question that should be considered more frequently, particularly with SKs who have engaged with the therapeutic process. The possible role of loneliness in the genesis of the antecedents to sexual killing warrants further exploration and whether it has a role in the escalation of behaviour to sexual killing should be examined. MacCulloch et al. (1983) uncovered such a dearth of social interaction and sexual encounters for the majority of the sexual killers in their study that, “sexual fantasy life and behavioural try-outs were their only source of sexual arousal” (p. 25). The possibility that loneliness contributes to the development of sadistic fantasies, development of paraphilias and hostile attitudes toward women could be a helpful next step to understand and inform the treatment of sexual killers. Ressler et al. (1988) have developed their understanding of sexual killing by in depth interviews with SKs, predominantly serial killers. Although there are limitations to this approach, we would suggest that it would be helpful to ask and collect information from prisoners about what, if anything, could have been done to prevent them from committing these crimes. Our experience in working with these prisoners suggests that this is an often overlooked but potentially helpful source of information. For this reason, even single case studies can usefully add to overall knowledge of assessment and treatment of such offenders.


We have considered the treatment needs of SKs and PMSIOs and conclude that the dynamic treatment needs identified in SARN and STABLE-2000 are relevant to both groups. In addition, we have made some suggestions about the way in which cognitive behavioural sex offender treatment programs can be adapted to be run with SKs. If possible, implementing programs within a specialist unit might provide the best environment to do this although providing a limited range of sites as opposed to mixing SKs with overall Sex Offender Treatment Programs would help to support staff working with these perpetrators and help ensure effective assessment and treatment for SKs within general custodial settings. In conclusion, SKs represent much of what is so challenging for forensic psychologists and treatment staff. Forensic staff are charged with trying to understand crimes, help inform assessments, develop interventions, and give opinions on risk that contribute to decisions concerning progression and release. The terrible crimes and consequences of SKs demand that a co-ordinated and research-driven approach be taken to improve and disseminate understanding of this crime.


We would like to thank Trish Wincote and David Thornton for their helpful comments on an earlier draft of this chapter. We would also like to thank Clive Hollin for his assistance in developing the ideas discussed in this chapter. The views expressed in this chapter are those of the authors and do not necessarily represent the official position of the Ministry of Justice and Her Majesty’s Prison Service.


Bartholomew, A. A., Milte, K. L., & Galabally, F. (1975). Sexual murder: Psychopathology and psychiatric jurisprudential considerations. Australian and New Zealand Journal of Psychotherapy, 34, 20-25.

Beauregard, E., & Proulx, J. (2002). Profiles in the offending process of nonserial sexual murderers. International Journal of Offender Therapy and Comparative Criminology, 46(4), 386-399.

Beech, A. R., Fisher, D., & Ward, T. (2005). Sexual murderers’ implicit theories. Journal of Interpersonal Violence, 20, 1366-1389.

Blanchard, G. T. (1995). Sexually addicted lust murderers. Sexual Addiction and Compulsion, 2, 62-71.

Bradford, J. M. W. (manuscript in preparation). The biomedical treatment of sexual sadism and associated conditions.

Briken, P., Habermann, N., Berner, W., & Hill, A. (2005). The influence of brain abnormalities on psychosocial development, criminal history and paraphilias in sexual murderers. Journal of Forensic Sciences, 50, 1204-1208.

Briken, P., Habermann, N., Kafka, M. P., Berner, W., & Hill, A. (2006). The paraphilia-related disordered: An investigation of the relevance of the concept in sexual murderers. Journal of Forensic Sciences, 51, 683-688.

Brittain, R. (1970). The sadistic murderer. Medicine, Science and the Law, 10, 198-207.

Burg, B. R. (1982). The sick and the dead: The development of psychological theory on necrophilia from Krafft-Ebing to the present. Journal of the History of the Behavioral Sciences, 18, 242-254.

Clarke, J., & Carter, A. J. (2000). Relapse prevention with sexual murderers. In D. R. Laws, S. M. Hudson, & T. Ward (Eds.), Remaking relapse prevention with sex offenders (pp.389-401). London: Sage Publications, Inc.

Campos, E., & Cusson, M. (2007). Serial killers and sexual murderers. In J. Proulx, E. Beauregard, M. Cusson, & A. Nicole (Eds.), Sexual murderers: A comparative analysis and new perspectives (pp. 107-122). John Wiley & Sons Ltd.

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

Dimock, J., & Smith, S. (1997). Necrophilia and anti-social acts. In L. B. Schlesinger & E. Revitch (Eds.), Sexual dynamics of anti-social behaviour (2nd ed) (pp. 241-251). Charles C Thomas Publisher Ltd.

Firestone, P., Bradford, J. M., Greenberg, D. M., & Larose, M. R. (1998). Homicidal sex offenders: Psychological, phallometric, and diagnostic features. Journal of American Academy Psychiatry and Law, 26(4), 537-552.

Fleiss, J. L. (1981). Statistical methods for rates and proportions (2nd ed). New York: John Wiley & Sons.

Folino, J. O. (2000). Sexual homicides and their classification according to motivation: A report from Argentina. International Journal of Offender Therapy and Comparative Criminology, 44(6), 470-750.

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

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

Grubin, D. (1994). Sexual murder. British Journal of Psychiatry, 165(5), 624-629.

Hare, R. D. (1991). The Hare psychopathy checklist-revised manual. Toronto: Multi-Health Systems.

Harris, A. J. R., & Hanson, R. K. (2000). STABLE-2000. Unpublished manuscript. Department of the Solicitor General Canada.

Holmes, R. M. (1991). Sex crimes. California: Sage Publications, Inc.

Hucker, S. J., Langevin, R., Dickey, R., Handy, L., Chambers, J., & Wright, P. (1988). Cerebral damage and dysfunction in sexually aggressive men. Annals of Sex Research, 1, 33-47.

Hucker, S. J., & Stermac, L. (1992). The evaluation and treatment of sexual violence, necrophilia, and asphyxiophilia. Psychiatric Clinics of North America, 15, 703-719.

Kafka, M. P. (2000). The paraphilia-related disorders. Nonparaphilic hypersexuality and sexual compulsivity/addiction. In S. R. Leiblum & R. C. Rosen, (Eds), Principles and practice of sex therapy (pp. 471-503). New York: Guildford Press.

Kennedy, F., Hoffman, H. R., & Haines, W. H. (1947). A study of William Heirens. American Journal of Psychiatry, 104, 113-121.

Langevin, R., Ben-Aron, M. H., Wright, P., Marchese, V., & Handy, L. (1988). The sex killer. Annals of Sex Research, 1, 263-301.

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

Malmquist, C. P. (1996). Homicide: A psychiatric perspective. Washington, DC: American Psychiatric Press, Inc.

Marshall, W. L., & Hucker, S. J. (2006). Severe sexual sadism: Its features and treatment. In R. McAnulty & M. Burnette (Eds.), Sex and Sexuality (pp. 227-250). Praeger Perspectives.

Milsom, J., Beech, A. R., & Webster, S. D. (2003). Emotional loneliness in sexual murderers: A qualitative analysis. Sexual Abuse: A Journal of Research and Treatment, 15, 285-296.

Nichols, H. R., & Molinder, I. (1984). The Multiphasic Sex Inventory Manual. Tacoma, WA: Nichols & Molinder.

Oliver, C. J., Beech, A. R., Fisher, D., & Beckett, R. (2007). A comparison of rapists and sexual murderers on demographic and selected psychometric measures. In J. Proulx, E. Beauregard, M. Cusson, & A. Nicole (Eds.), Sexual murderers: A comparative analysis and new perspectives (pp. 159-173). John Wiley & Sons Ltd.

Podolsky, E. (1965). The lust murderer. Medico-Legal Journal, 33, 174-178.

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

Prentky, R. A., Burgess, A. W., Rokous, F., Lee, A., Harman, C., Ressler, R. et al. (1989). The presumptive role of fantasy in serial sexual homicide. American Journal of Psychiatry, 146, 887-891.

Proulx, J., Blais, E., & Beauregard, E. (2007). Sadistic sexual offenders. In J. Proulx, E. Beauregard, M. Cusson, & A. Nicole (Eds.), Sexual murderers: A comparative analysis and new perspectives (pp. 107-122). John Wiley & Sons Ltd.

Proulx, J., & Sauvêtre, N. (2007). Sexual murderers and sexual aggressors: Psychopathological considerations. In J. Proulx, E. Beauregard, M. Cusson, & A. Nicole (Eds.), Sexual murderers: A comparative analysis and new perspectives (pp. 51-69). John Wiley & Sons Ltd.

Purcell, C. E., & Arrigo, B. A. (2006). The psychology of lust murder: Paraphilia, sexual killing, and serial homicide. San Diego, CA: Elsevier Academic Press.

Ressler, R. K., Burgess, A. W., & Douglas, J. E. (1988). Sexual homicide: Patterns and motives. New York: Lexington.

Ressler, R. K., Burgess, A. W., Hartman, C. R., Douglas, J. E., & McCormack, A. (1986). Murderers who rape and mutilate. Journal of Interpersonal Violence, 1, 273-287.

Revitch, E. (1965). Sex murder and the potential sex murderer. Diseases of the Nervous System, 26, 640-648.

Revitch, E., & Schlesinger, L. B. (1989). Sex murder and sex aggression. Springfield, IL: Charles C Thomas.

Roberts, J. V., & Grossman, M. G. (1993). Sexual homicide in Canada: A descriptive analysis. Annals of Sex Research, 6, 5-25.

Rosman, J. P., & Resnick, P. J. (1989). Sexual attraction to corpses: A psychiatric review of necrophilia. Bulletin of American Academy of Psychiatry and Law, 17, 153-163.

Schlesinger, L. B. (2004). Sexual murder: Catathymic and compulsive homicides. Boca Raton, FL: CRC Press.

Tardif, M., Daaylva, B., & Nicole, A. (2007). Psychotherapeutic and psychodynamic issues with sexual murderers: A comparison of rapists and sexual murderers on demographic and selected psychometric measures. In J. Proulx, E. Beauregard, M. Cusson, & A. Nicole (Eds.), Sexual murderers: A comparative analysis and new perspectives (pp. 213-228). John Wiley & Sons Ltd.

Thornton, D. (2002). Constructing and testing a framework for dynamic risk assessment. Sexual Abuse: A Journal of Research & Treatment, 14, 139-153.

Thornton, D., Mann, R., Webster, S., Blud. L., Travers, R., Friendship, C. et al. (2003). Distinguishing and combining risks for sexual and violent recidivism. In R. Prentky, E. Janus, & M. Seto (Eds.), Understanding and managing sexually coercive behaviour (pp. 225-235). Annals of the Academy of Sciences.

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

West, D. J. (1987). Sexual crimes and confrontations: A study of victims and offenders. Brookefield, VT: Gower.

Wilson, R. J., Picheca, J. E., & Prinzo, M. (2005). Circles of support and accountability: An evaluation of the pilot project in South-Central Ontario. Research report R-168, Ottawa, Ontario: Correctional Service of Canada.

Yarvis, R. M. (1995). Diagnostic patterns among three violent offender types. Bulletin of the American Academy of Psychiatry and Law, 23, 411-419.

Appendix 1

Graphs showing the average scores for 19 SKs and 17 PMSIOs on a range of psychometric measures, compared to the average score for 644 adult male (mean age 48.6 years)

Graph 1. — Offence Supportive Attitudes Domain: Hi Scores ( .55) indicate Treatment Need

Click image below for larger view.

Graph 1. — Offence Supportive Attitudes Domain: Hi Scores (.55) indicate Treatment Need

Graph 2. — Relationships Domain (Graph 1): Hi Scores ( .55) indicate Treatment Need

Click image below for larger view.

Graph 2. — Relationships Domain (Graph 1): Hi Scores ( .55) indicate Treatment Need

Graph 3. — Relationships Domain (Graph 2): Low Scores (.45) indicate Treatment Need

Click image below for larger view.

Graph 3. — Relationships Domain (Graph 2): Low Scores (.45) indicate Treatment Need

Graph 4. — Self Management Domain: Hi Scores ( .55) indicate Treatment Need

Click image below for larger view.

Graph 4. — Self Management Domain: Hi Scores ( .55) indicate Treatment Need

Mr. Carter’s Discussion

Arrigo: I am interested in what some would call the least restrictive environment, the least invasive form of intervention. At what point does one make a determination about step-down levels of care or release? Eventually, people have to resocialize and recommunalize. What kinds of things do you do in your unit as a basis to say, “Ok, this client can make this transition?”

Carter: We expect them to demonstrate learning in how they interact with staff. If they have an anti-authoritarian approach to staff before they come to the unit, and they start to be more engaging and therapeutic with staff, that demonstrates learning. Often their relationship with their probation officer is an important indication that they are working constructively. When people start to progress closer to release, they must have an opportunity to work in placements within the community. That is an important indication of how well they are generalizing learned skills.

Arrigo: To what extent do our interventions enable people to act out, particularly when they are troubled and vulnerable, in comparison to the extent to which our interventions enable and empower them to tap into their competencies. In my experience, we create interventions that indeed tap into people’s limitations and pathologies and to some we create asylum-like conditions.

Carter: We are not setting out in treatment to make someone who is going to necessarily be the person you want to live next door to or sit and talk to on the bus. All we are really trying to do is to stop them being at risk for committing a serious offence. We need to be careful not to instil our values and so forth on to these individuals.

Proulx: In your model the sadistic offenders are trapped in their vicious cycle with mental disorder, social isolation, and low self-esteem and they cannot get out by themselves. In treatment, we do not focus only on deficits, we also focus on the skills that they have to develop to get out of that cycle. We give them opportunities to improve their lives so they will not have to engage in sexual aggression as the only way to cope or feel alive. In a treatment program, we deal with risk factors, but we also focus on the development of skills to improve quality of life.

Carter: All our programs run with a Good Lives approach. In the extended program, we do an exercise where they compare the “new me” to the “old me”. They are working towards trying to meet their needs in a prosocial way that leads them away from violence.

Schweighofer: Unfortunately the nature of our work tends to pull us toward pathology. In supervising staff, one of the constant refrains is “What can we say that is positive about this man? What can we emphasize in this report about what he is doing well?” The importance of remaining mindful of the strengths and explicitly commenting on them is something too often overlooked.

Arrigo: We are so drawn to the pathology, harm, and violence that it becomes the definition that we attach to that person which then forecloses other ways of looking at the individual that might preclude alternative forms of intervention. To what extent do we find ways to encourage, enable and empower this person to tap into the strengths that he may have?

Schweighofer: Increasingly, with the ascendance of the Good Lives model, that way of thinking is achieving increased emphasis.

Pagé: The Good Lives Model has fantastic things to offer, but we do not want to forget relapse prevention as part of the model. The combination of both can help us to enhance treatment and make it more effective for the clients.

Bradford: Mr. Carter, have any of your guys been released? What are the follow-up and rates of recidivism?

Carter: We do not have statistics on the effectiveness of SOTP with Sexual Killers. However, in terms of further reconvictions, between 2000 and 2007 there were approximately 1700 life sentence prisoners released on life licence. I know of only 3 cases (from a total of about 30 reconvictions for serious offences) where there was a murder committed by a life licence prisoner and I understand in only one of these cases was the killing considered to be sexual. This killing was not committed by someone on life licence for murder or manslaughter. These figures have not been published and should be treated as an estimate but may act as an indication of future murders committed by life sentence prisoners. A proportion of these offenders would have taken a Sexual Offender Treatment Program, it is not possible to draw conclusions about the efficacy of treatment from these figures.

Perkins: In terms of child molesters, rapists, and sexual killers, who should be together in group therapy? I wondered if you have had experience with mixtures of types of offenders together, and what are your observations on that?

Carter: We mix people together, and that is mainly because they had to complete treatment before a hearing or by a given time, so we had to get people into programs. Inevitably, we had a mixture of people who had killed adults and children, males and females. I think having that dynamic wasn’t a bad thing. If anything, it created different discussions and you would find that someone who was initially resistant to answering questions would feel more comfortable asking questions about someone else’s offences because they were different from theirs. You could feed that back to them and really force the fact that they were giving insights and asking questions. In my experience, the mix has been very helpful and something I would support.

Looman: Mr. Carter, with regard to mixing different types of offenders in the same group, in our program, we’ve had at least three guys who have been post-mortem sexual interference in our regular groups and they’ve disclosed, not in gory detail, but they’ve disclosed the nature of their offences in the group. I wouldn’t use the word “well-received” but it was accepted, it was surprising how little reaction that got. One of the guys in particular was a repulsive person to begin with, in terms of his mannerisms, hygiene, etc, and he got almost no heat from group for the kind of offence he committed.

Carter: I think a key consideration is how you set up the group. We bring a graduate from the previous group into the new group. The graduate talks about what they got from disclosing their offence, what they got from treatment, what they found difficult about it. There was one prisoner who denied any sexual motivation for his offence for over 20 years and wouldn’t cooperate with staff and he made a turn around, so we used him as the first person to act as the graduate. I have felt that if you set up a group well, it doesn’t matter, there can be arguments and tensions and people can fall out – you can always recover from it.

Pagé: I questioned the validity of open-ended groups at first but moving forward, the graduates would stay in the program and became role models for the other offenders coming into the group. The guys that are in the program really benefit from having mentors and knowing that everybody goes through the same process. For us, open-ended groups worked really well.

Carter: We do have open-ended groups or rolling programs. There are some advantages to that. Modelling of people actually disclosing and acting as role models for other group members can be beneficial. As far as I know that’s not being run with groups of sexual killers. But it would certainly be an interesting prospect to explore further.

Schweighofer: As someone who supervises staff, including female therapists, I was struck by your comment that male co-therapists were often unaware of some of the subtly sexually coercive behaviours that female staff were being subjected to. Were there any particular strategies or techniques that you were able to employ that heightened the male co-therapists’ or supervisor’s awareness?

Carter: As we started to develop there was more of an emphasis on staff support and welfare and these kinds of supervision issues. We are starting to look at some of these issues. For example, if a prisoner started to speak over a female facilitator and so forth, then the male facilitator would jump in and stop it. Rather than thinking about how they could support the female facilitator to deal with it themselves. Knowing that this happens and making sure that you turn to these issues and they are discussed would be a necessary thing to do. In addition, all people working in these groups have to go and see a counsellor during the time they’re working in treatment.

Perkins: It is quite difficult at times and it is important that there be space for looking at how, not just male-female but junior-senior facilitators operate together and what messages that conveys or doesn’t convey to men who may be attuned to potentially exploiting a situation sadistically.

Abbott: Mr. Carter mentioned the need for appropriate training and specialized training in this area for staff, especially staff that are providing treatment. What would you recommend in terms of components of training, or what did you have in place when you were working in that facility in training?

Carter: What we have is very different to what I’d do now if I was going back. Everyone on the unit was trained in the sex offender treatment program so that they could work together in a supportive way regardless of whether they were going to deliver treatment or not. We ran staff awareness throughout the whole prison. Some training would have been helpful regarding the integrated theories of sexual offending, awareness about working with people with psychopathy, and case formulation. The training would divide into different roles depending whether people were specifically working to support or deliver treatment. Additionally, training in defending reports at parole hearings should be included because you can do a very good job at case formulation, but if you can’t go out and defend your report then a panel may not take any interest in it. If I had the opportunity again, I would look at specific training packages and keep people up to date with the literature and research into sexual murder.