Mapping french people ’ s views on chemical castration of child and adolescent sex offenders

Sex offenses to children or young adolescents have become a growing public concern. Chemical castration is currently considered as the best available societal response to child sex abuse. It abolishes testosterone secretion, its effects are reversible and side effects are minor. It has been argued that offering convicted sex offenders with the possibility to be treated may be in contradiction with the bioethics principle of autonomy because the person has really no other choice -the alternative is usually a lengthy confinement sentence. In view of this controversy, we explored lay people’s and physicians’ views regarding the acceptability of chemical castration. Fifty participants (among them five physicians) judged the acceptability of castration in each of 36 scenarios consisting of all combinations of four factors: aggressor’s age (21 vs. 41-year old); (b) victim(s)’ age (5, 8, or 14-year old); (c) aggressor’s psychiatric status (no psychiatric antecedents, suffers from sexual deviation, or recidivist); and (d) family’s attitude (hostile to any kind of castration vs. approve castration). Participants’ ratings of acceptability were, on the average, very high, and 68% of participants considered that chemical castration was fully justified in all the cases that were shown. A small minority (8%) considered that first time offenders of young adolescents, without psychiatric antecedents, should not be chemically castrated. Another minority position (24%) expressed doubts regarding chemical castration of first time offenders without antecedents but they never strongly opposed it. Implications for bioethics are discussed.

Para citar este artículo: Sedkaoui, H., & Mullet, E. (2016).Mapping french people's views on chemical castration of child and adolescent sex offenders.Universitas Psychologica, 15 (3).http://dx.doi.org/10.11144/Javeriana.upsy15-3.mfpvSince the 70's, sex offenses to children or young adolescents have -although their prevalence has tended to decline in recent years-become a growing public concern (Putnam, 2003).It has been estimated that about 17% of adult women in the US have been the victims of sex offenses before age 17.In France, the corresponding statistics is about 10% (Ciavaldini, 2016).Public concern has been fueled by the realization that sex offenses may have severe consequences: Victims are more likely than non-victims to develop depression, addictive behaviors, and other forms of psychological maladjustments (Putnam, 2003).
For most people, child sex offenders are considered or described as pedophiles; that is, as people who manifest a strong, irrepressible sexual interest in youngsters.This is not always the case: Anti-social tendencies, a sense of entitlement to sex (often incestuous), problems with the adult sexual partner (e.g., conflict or unavailability), substance abuse, and sexual urges may be alternative motives for child abuse.Physiological factors can also play a role, as when a cerebral tumor gradually blocks the functioning of impulse control centers (Burns & Swerdlow, 2003).Child sex offenders are often depicted "as aggressive dangerous fiends waiting to seize and ravish unaccompanied young children" (Stone, Winslade, & Klugman, 2000, p. 85).In fact, most of the time (90%, according to current statistics), the molested children knew their abusers: family members, family friends, or educators (Chenier, 2012;Stone et al., 2000).
Intergenerational sexual relationships involving very young children have not always been considered as criminal in character.They have been common practice thorough human evolution (Seto, 2008), and sometimes viewed as honorific for the child (Spellberg, 1994).Today, however, people's reactions towards sexual relationships with children are negative to the extreme."The phenomenon of sexual abuse is intertwined with a strong emotionalism that exacts an almost visceral response in nearly everyone, and this emotionalism has confounded our lawmakers' collective abilities to separate legislative proposals that are functionally efficacious from those that are certainly well intentioned but nonetheless unsuccessful" (Edwards & Hensley, 2001, p. 84).
Societal responses to child sex offense have been of two kinds: penal and medicalpsychological (Stone et al., 2000).Penal responses have involved capital punishment, life confinement, and, more recently, confinement in association with chemical treatment.Medical-psychological treatments have included behavioral techniques (e.g., olfactory aversion therapy), cognitive-behavioral therapy, and surgical or chemical castration.Lösel and Schmucker (2005) reported that castration considerably reduced recidivism, and Hughes (2007) suggested that for behavior therapy to be effective it must be associated with chemical castration.Chemical castration therefore appears as the best available societal response to child sex abuse.In present days, chemical castration most frequently involves the parenteral administration of long-acting analogues of Gonadotropin-releasing hormone (GnRH) (these substances are more potent anti-androgens than Cyproterone acetate and medroxy-progesterone acetate, which were used in the past).They abolish testosterone secretion, their effect is reversible and side effects are minor.
In Western Europe (and in France in particular), when a person has, for the first time, been convicted for child sex offense, this person is usually confined during some months and then offered the opportunity to be treated by chemical castration with analogues of GnRH in association with psychological therapy.If the person rejects this possibility, this person can be sentenced to five to twenty years of prison, depending on the severity and number of previous assaults (Articles 227-25, 227-26 and 227-27 of the French Penal Code).
Some authors have argued that offering convicted sex offenders with the possibility to be treated is in contradiction with the bioethics principle of autonomy according to which any patient has the right to refuse or choose a treatment (Green, 1986): Even if the person agrees to be chemically castrated, this person has really no other choice.The alternative to treatment is either a lengthy confinement sentence or, in some countries, capital punishment.Given these circumstances, it would be an exaggeration to state that a convicted sex offender who "willingly" undergoes treatment has given voluntary consent to an offer of probation assorted with chemical treatment (see also, European Committee for the Prevention of Torture, 2009).
In view of the controversy regarding this issue, the present study aimed at exploring lay people's views regarding the acceptability of chemical castration, and the factors that impact on acceptability.The factors that have been considered were the victims' age and gender, the aggressor's age, psychiatric status and whether he was a relative of the victim, and the aggressor's family attitude to castration.As stated in the French Penal Code, young, non-recidivist aggressor must be offered the opportunity not to be castrated if they agree to benefit from psychological therapy.The aggressor's gender was held constant (male).
Our first hypothesis was that chemical castration would be considered as quite acceptable by most people.It was based on the consideration that child sex abuse is usually perceived as not simply bad but as totally disgusting (Jahnke, Imhoff, & Hoyer, 2015).According to Moral Foundation Theory (Haidt, 2013), disgust, when activated, tends to trigger manifestations of indignation and strong punitive impulses, as shown in previous studies on rape of women (Kamble & Mullet, 2016) and corruption in the judiciary.Our second hypothesis was that, despite possible unanimity among participants to condemn such practices, qualitatively different positions should be found, among them one that considers that circumstances can matter and that first time offenders would be treated less harshly than recidivists.

Participants
The 50 participants were unpaid volunteers from the region of Toulouse, France, who were informed about the goals of the study and gave their consent.Five of them were physicians.Their demographic characteristics are shown in Table 1.The lay people were approached by two trained research assistants while they were walking along the main sidewalks of Toulouse.The professionals were contacted at the hospital.

Material
The material consisted of 42 cards containing a scenario of a few lines, a question, and a response scale.Thirty-six of the scenarios were composed according to a four within-subject factor design, presented in the following order: (a) the aggressor's age (21 vs. 41-year old), (b) the victim(s)' age (5, 8, or 14-year old), (c) the psychiatric status of the aggressor (no psychiatric antecedents, suffers from sexual deviation, or recidivist), and (d) the family's attitude (hostile to any kind of castration vs. approve castration), 2 x 3 x 3 x 2. The aggressor's gender was held constant (male) and the victim's gender too (female).Six additional scenarios were taken from the set of thirty six and altered in two different ways.In three scenarios the victim was a young boy and in three other scenarios it was explicitly reported that the victim was a family member.These six variables represent the set of variables found in documents that describe most cases of pedophilia in the literature (Seto, 2008).
An example of scenario is the following: "Mr.Laurent Verfeuil is 21-year old.He has engaged himself in acts of pedophilia towards a young girl aged 8.He was arrested, and examined by a psychiatrist.The expert's conclusion is that Mr. Verfeuil suffers from sexual deviation, which means that he could recidivate if not adequately treated.Mr. Verfeuil's family is fine with the idea of chemical castration.The psychiatrist recommends mandatory chemical castration for Mr. Verfeuil."Under each scenario a question and a response scale were found.The question was, "to what extent do you believe that the expert's decision to recommend chemical castration in this case is justified?"The response scale was an 11-point scale (0-10) with a left-hand anchor of "Not at all" and a righthand anchor of "Completely."The cards were arranged by chance and in a different order for each participant.

Procedure
The site was a vacant classroom in the university or a vacant office in the hospital or in the participant's home.The procedure followed Anderson's (2008Anderson's ( , 2013) ) recommendations for this kind of study.Before responding all participants were informed about the nature and implications of chemical castration.The participants took 30-35 minutes to complete the ratings.For practical reasons, informed consent was given orally.It was given once the participants had fully realized the nature of the study.

Results
A cluster analysis was performed on the raw data (Hofmans & Mullet, 2013).Three clusters were identified; they are shown in Figure 1.An overall ANOVA with a design of Cluster x Aggressor's age x Victim's age x Psychiatric status x Family's attitude, 2 x 3 x 3 x 2, was first performed in order to examine whether these clusters significantly differed.The Cluster effect was significant, p < 0.001, and post-hoc analyses showed that the three means were significantly different the ones from the others, p < 0.001.Two interactions involving the Cluster factor were also significant.As the overall ANOVA showed significant effects, separate ANOVAs were performed on the data from each cluster.Their design was Aggressor's age x Victim's age x Psychiatric status x Family's attitude, 2 x 3 x 3 x 2. The composition of these clusters as a function of age, gender, and other demographic characteristics is shown in Table 1.
The first cluster (N = 4) was termed Depends on Circumstances since the responses varied as a function of the aggressor's age and psychiatric status, and the mean rating was close to the center of the scale (M = 5.54).Ratings were lower when the aggressor was 21 (M = 4.62) than when he was 41 (M = 6.46),F (1, 3) = 5.67, p < 0.1, Eta² p = 0.65, and when there were no psychiatric antecedents (M = 4.37) than in the two other cases (M = 5.44 and 6.82), F (2, 6) = 3.67, p < 0.1, Eta² p = .55.In 10 cases the ratings were on the "not justified" side of the response scale, namely when the aggressor was 21 and had either no psychiatric disorder or when he had, for the first time in his life, aggressed adolescents, not young children.There were no differences in regards to whether the victim was a member of the family or not, or whether the victim was a boy or a girl.
Source: own work The second cluster (N = 12) was called Mainly Justified since the mean rating was much higher than in the first cluster (M = 8.22), and only one case was rated lower than 5 (young aggressor of an adolescent girl, without antecedents, and from a family hostile to the procedure, M = 4.25).Ratings were, however, lower when there were no psychiatric antecedents (M = 6.69) than in the two other cases (M = 8.58 and 9.38), F (2, 22) = 41.08,p < 0.1, Eta² p = 0.78.There was a small difference in ratings whether the victim was a boy (M = 6.39) or a girl (M = 5.83), F(1, 11) = 5.19, p < 0.05, Eta² p = 0.32.
The third cluster, the majority cluster (N = 34), was called Always Justified .All ratings were higher than 8.60 and the mean rating was 9.67.As shown in Table 1, the three clusters significantly differ in terms of educational level: All members of the "depends on circumstances" cluster had a university degree.

Discussion
As hypothesized, participants' ratings of acceptability were, on the average, very high, and a high percentage of participants considered that chemical castration was fully justified in all the cases that were shown.This finding was consistent with previous findings from studies involving situations that are perceived as disgusting (Kamble & Mullet, 2016;López López et al., in press).It was also consistent with findings from studies on involuntary hospitalization of potentially dangerous patients (e.g., Guedj, Sorum, & Mullet, 2012).In this study, 95% of participants (lay people and professional health caregivers) agreed that involuntary hospitalization is acceptable when the patient presents a risk to others.As also hypothesized, qualitatively different positions were found, among them one minority position that considered that first time offenders of young adolescents, without psychiatric antecedents should not be chemically castrated.All the participants holding this position had university degrees but no physician was a member of this group.Another minority position expressed doubts relatively to the chemical castration of first time offenders without antecedents but they never strongly opposed it.Educational level was the only demographic characteristic that impacted on participant views.
These findings, in association with findings from other studies on patients' involuntary treatment (e.g., Lhermite, Muñoz Sastre, Sorum, & Mullet, 2015) led to think that, in some cases, the patient's right to autonomy principle of bioethics may be superseded by the nonmalevolence principle ("first, do no harm"), which applies in these cases to the patient's social environment (protecting the other young family members).Some authors have, however, argued that involuntary treatment does not always amount to denying patients' rights to autonomy (Douglas, Bonte, Focquaert, Devolder, & Sterckx, 2013).Chemical castration through parenteral injection of long-acting analogues of GnRH has few side effects.It allows patients to live a normal life, to take part of family celebrations, and lead a productive life, outside correctional facilities.In other words, chemical castration enhances the person's autonomy.
Some patients truly agree with their chemical castration because they feel uneasy with their sexual tendencies; in these cases there is no conflict of ethical values.Other patients acquiesce to chemical castration because they feel they have no choice.In these cases, there may be conflict in ethical values but this conflict is quite minor.The patient's right to autonomy is of course temporarily superseded by the principle of benevolence (treating the patient for his own good) but the final benefit for the patient is long-term right to autonomy.

Figure 1
Figure 1Acceptability judgments as a function of the victim's age and the aggressor's characteristics