Sexual Homicide in France

Criminological Theories of Modus Operandi

It has only been recently that criminological theories have been applied to the study of sexual aggression in general (see Lussier & Beauregard, 2018), and sexual homicide in particular (e.g., DeLisi & Wright, 2014; James et al., 2019). These theories suggest that modus operandi is directly and indirectly associated with the context and location of sexual assaults (Nagin & Paternoster, 1993; Piquero & Tibbetts, 1996): Each type of crime is shaped by a script, which in turn is influenced by the constraints on the crime, that is, its choice-structuring properties (Cornish, 1994). These constraints may be external (e.g., victim resistance), situational (e.g., crime scene characteristics), or internal (e.g., aggressor’s beliefs and emotions; Beauregard & Leclerc, 2007).

The sexual aggressor’s lifestyle and life events in the hours preceding his sexual assault may also influence the way he commits his crime. Blanchette, St-Yves, and Proulx (2009) have reported an association between sexual aggressors’ routine behaviour (e.g., social isolation, compliance with social norms, partying) and both their preparation of their crimes and their behaviour during their crimes (e.g., sexual acts, use of a weapon). Furthermore, Beauregard, Lussier, and Proulx (2005) have reported an association between situational factors, such as alcohol consumption, and the degree of victim coercion and injury. However, although these variables may influence the aggressor’s decision making, they cannot completely explain his behaviour during the crime. Further insight into this behaviour is provided by the rational-choice perspective.

The focus of the rational-choice perspective (e.g., Cornish & Clarke, 1986) is the decision-making process of criminals during their crimes. Chéné and Cusson (2007) analyzed the intentional factors (e.g., anger, deviant sexual fantasies) and situational factors (e.g., victim resistance) of the decision-making process of 141 sexual aggressors against women (including 40 sexual murders). They found that the circumstances of the crime (e.g., aggressor’s precrime emotions, presence of a weapon, aggressor–victim relationship) are key determinants of modus operandi. For example, sexual assaults committed by individuals who are angry and armed are more likely to culminate in the death of the victim. Although the application of criminological theories to research on sexual assaults has revealed the influence of contextual and situational factors on the commission of these crimes, these theories say little about the influence of affective and cognitive processes on modus operandi.

Psychological Theories of Modus Operandi

To take into account the cognitive, behavioural, motivational, and contextual factors that favour sexual aggression, Ward and his collaborators developed the self-regulation model (SRM; for example, Ward & Hudson, 1998). This model comprises four offending pathways: approach explicit, approach automatic, avoidant active, and passive avoidant. Sexual aggressors following the approach explicit pathway commit planned crimes, which are shaped by sexual deviance (e.g., sadistic sexual fantasies) and both positive (e.g., sexual excitation) and negative (e.g., anger) affects. Their feelings of satisfaction after the crime and their desire to more perfectly act out their fantasies lead these offenders to plan further sexual crimes. Prior to their crime, aggressors following the approach automatic pathway experience feelings of hostility towards women and entitlement to satisfy their sexual desires in whatever way they wish. Their modus operandi is impulsive and characterized by instrumental violence. Following their crime, positive affects (e.g., sexual satisfaction) favour the maintenance of a cognitive-affective state of preparedness until the next opportunity to assault presents itself. Finally, aggressors following the avoidant active or avoidant passive pathways attempt to avoid coercive sexual behaviours. However, their avoidance strategies are counterproductive (e.g., alcohol consumption) or nonexistent. Their sexual crimes are motivated by the desire for sexual gratification and are extensions of their lifestyles, which are dominated by negative affects (e.g., anxiety). While aggressors following the avoidant active pathway are opportunists, those following the avoidant passive pathway engage in implicit planning. Although Ward and Hudson’s study demonstrated the influence of internal constraints on sexual aggressors’ modus operandi, it did not take into account the interaction between internal and external constraints.

Proulx and Beauregard (2009, 2014) identified three offending pathways in sexual aggressors against women: sadistic, angry, and opportunistic. This typology was based an analysis of the personality traits, general and sexual lifestyles, life context, immediate precursors to the crime, and modus operandi of 180 sexual aggressors against women (including 59 sexual murderers). Sexual aggressors following the sadistic pathway had low self-esteem in the hours preceding the crime. Their modus operandi was shaped by sadistic sexual fantasies, and their crimes tend to be planned. Aggressors following the angry pathway led an antisocial lifestyle, were impulsive and extremely violent, and were not emotionally invested in their sexual partners. Their modus operandi was shaped by a desire for vengeance and by intense anger. Aggressors following the opportunistic pathway had a lifestyle which they found generally satisfactory but which did not allow them to satisfy their sexual needs. Their modus operandi was shaped by the desire to obtain sexual gratification and by the conviction that they were superior to others. Their crime generally was the result of opportunity and was characterized by instrumental violence. While Proulx and Beauregard’s (2009, 2014) studies demonstrate the heterogeneity of offending pathways in sexual aggressors, it should be noted that although their sample included many sexual murderers (32.8%), they did not specifically study these offenders’ modus operandi.

The Modus Operandi of SSMs and NSMs

James and Proulx’s (2016) systematic review of the modus operandi of SSMs and NSMs revealed several distinctive features of these two groups of offenders. In the year preceding their sexual homicide, SSMs feel like failures, and experience both contextual anxiety (e.g., job, legal) and relational difficulties (e.g., family disputes). In an attempt to cope with these problems, they resort to alcohol and drugs. However, while excessive alcohol consumption tends to disinhibit their behaviours, it may also lead to sexual dysfunctions (e.g., erectile, ejaculatory) and deterioration of an already problematic sex life (e.g., sexual inhibition). To cope with their accumulated relational failures, SSMs invest—and sometimes overinvest—in a fantasy world. As a result of habituation, their deviant sexual fantasies become increasingly extreme, invasive (in terms of duration, frequency, and intensity), and insistent (in terms of the need to act out their fantasies in real life). SSMs spend a good part of their days in their fantasy world, in which they plan their crimes, and, not surprisingly, their crimes are carefully premeditated. In the hours preceding their homicide, SSMs feel anger towards women, assemble an attack kit (e.g., knife, rope, gloves, hood, lighter, drugs, recording equipment), and seek out a victim with specific characteristics whom they can attack at a preselected location. They tend to use a ruse strategy to approach their victims, and then confine and restrain them to neutralize them and better satisfy their deviant sexual needs. Because their crimes never completely reproduce their fantasy scenarios, they obtain only partial sexual satisfaction, which motivates them to plan another attack. James and Proulx (2016) report that SSMs’ modus operandi is similar to those of sexual aggressors following the approach explicit (Ward & Hudson, 1998) and sadistic (Proulx & Beauregard, 2009, 2014) pathways.

On the contrary, NSMs, in the year preceding their sexual homicide, are generally dissatisfied with their lives: They experience great socioeconomic difficulty, are marginalized, and lead an unstable life. Because of their cognitions and social isolation, they develop negative affects (e.g., anxiety, frustration) and coping strategies (e.g., consumption of alcohol, frequenting of prostitutes). However, when these strategies prove ineffective, they accentuate NSMs’ aggressivity and sexual arousal. When a woman refuses to satisfy their sexual needs, these pent-up emotions ultimately erupt in an explosion of anger. NSMs’ impulsivity and anger are more than merely the consequences of their antisocial personality and their hostility towards women—they are also internal constraints that shape NSMs’ modus operandi, and drive them to commit crimes that are impulsive, brutal, and disorganized. Sexual homicide allows NSMs to express their anger at real or perceived injustices, with no consideration for the consequences to others. They do not appear to be concerned about leaving trace evidence at the crime scene—presumably because their mental state during the crime is influenced by anger and alcohol—and tend to be rapidly identified by the police. James and Proulx (2016) conclude that NSMs’ modus operandi is consistent with that of sexual aggressors following the approach automatic (Ward & Hudson, 1998) and angry (Proulx & Beauregard, 2009, 2014) pathways.

Although the results of James and Proulx’s systematic review may help investigators make decisions, there are three important limitations to that study that merit discussion. First, as the studies included in their review were, for the most part, conducted in North America and England, it is possible that their results do not apply to sexual murderers in other jurisdictions. Second, because of their group-formation criteria (SSMs: at least three victims; NSMs: one or two victims), their comparisons in fact included multiple-victim murderers in both groups (SSMs on one hand and NSMs with two victims on the other). Finally, their analysis was limited to univariate descriptive statistics, and it would be useful to verify the validity of their conclusions using more robust statistical techniques.

Aim of the Study

Criminological and psychological studies have demonstrated that sexual aggressors against women have a variety of modus operandi, and that these are shaped by specific contextual and situational factors; this also appears to be the case for North American SSMs and NSMs. However, no study has determined whether these observations are generalizable to sexual murderers elsewhere. Consequently, the objective of this study was to describe and compare the modus operandi, and associated contextual and situational factors, of French SSMs and NSMs; determine whether the results of James and Proulx’s (2016) systematic review can be generalized; and provide investigators with information that can help them develop investigative techniques specific to these offenders.

Participants

To be included in this study, an individual had to have committed a homicide (index offence) that satisfied at least one of the six sexual homicide criteria proposed by the FBI (Ressler et al., 1988): (a) victim attire or lack of attire, (b) exposure of the sexual parts of the victim’s body, (c) sexual positioning of the victim’s body, (d) insertion of foreign objects into the victim’s body cavities, (e) evidence of sexual intercourse (oral, anal, vaginal), and (f) evidence of substitute sexual activity, interest, or sadistic fantasy (e.g., genital mutilation).

The sample consisted of 120 men having committed 212 sexual homicides in France between 1975 and 2012. Of the 120 participants, 87 were NSMs (one victim, or multiple victims in a single event) and 33 were SSMs (two or more victims in separate events; FBI, 2008). To limit the potential bias generally associated with comparisons involving a group composed of serial offenders, only the characteristics of the SSMs’ first sexual homicide were analyzed. Moreover, seven of the homicides involved accomplices (for more details on the homicides involving accomplices, see Higgs, James, & Proulx, 2019). The final analysis was therefore based on 113 homicides committed by 120 sexual murderers. The majority of the participants were White (70.8%), and the mean age at the time of their first sexual murder was 30.3 years (SD = 10.9). Most were single (62.5%) and unemployed (61.7%), and had already been convicted of another crime (70.6%).

Instruments

Data were coded with the Sexual Murderers Multidimensional Inventory (SMMI; James & Proulx, 2015). This instrument collects information from case files on developmental, psychopathological, lifestyle, criminal career, predisposing, and precipitating factors, as well as on modus operandi, crime scene, and victim characteristics. All case files were evaluated by a clinical psychologist specialized in violent behaviours and crime scene analysis. To measure interrater reliability, 20 randomly selected files were coded independently by the first author, a behavioural analyst, and by an investigator of the Gendarmerie Nationale’s Behavioural Science Unit. Interrater reliability was measured in terms of an intraclass correlation coefficient (ICC) based on a two-way random effects model, with absolute agreement included in the ICC denominator. There was an excellent (Portney & Watkins, 2000) interrater reliability (ICC2 = .82).

Procedure

All participants had received a sentence for homicide. Confidential consultation of their criminal files was authorized by the Department of Justice of France. Data were collected between 2013 and 2015. To ensure the representativeness of the sample and avoid selection bias, the cases of sexual homicide were identified by investigators and behavioural analysts of the Gendarmerie Nationale’s Behavioural Sciences Unit and by prosecutors and chief registrars from different parts of France. Case files were located in 46 courthouses throughout France, and their mean length was 4,105 pages (range = 901-17,789 pages). These criminal files contained expert (psychiatric, psychological, neurological, medical), and social and institutional (social environment, school, work, military, criminal, prison) reports. Further information was obtained from investigative reports (crime scene photographs, and autopsy, toxicology, anatomy-pathology, genetics, and ballistic reports), psychosocial profiles of the victim, transcripts of witness interviews, interrogation of the murderer, and the arraignment orders of the Cour d’Assises. Data on the criminal record of the murderers were gathered from the National Criminal Database of France. In case of divergent information from offenders (i.e., information self-reported during interrogation or expert investigation) and official sources (e.g., autopsy), the latter was considered more reliable, and retained.

Variables

This study compares the modus operandi of SSMs and NSMs under four broad headings: precrime (n = 12), modus operandi (n = 30), postcrime (n = 27), and victimology (n = 10; see the appendix for variable description). Information was gathered from two sources: (a) criminal and judicial investigations, and (b) participant statements. The use of multiple sources allowed for the information to be cross-checked and for the creation of a more robust data set. Participant statements were collected on multiple occasions and at different points in the judicial process—witness examination (which seeks evidence which could lead to the laying of charges), instruction (which summarizes the prosecution’s evidence justifying the laying of charges), and judgement by the Cour d’Assise. Finally, a variety of professionals (e.g., police, judges, social investigators, psychologists, psychiatrists) collected statements from the sexual murderers and their entourage, which limited the potential for bias (retention or transformation of information) related to the interview context and the personal characteristics of the professionals.

Analytical Strategy

Differences in the occurrence of categorical variables in the two groups of participants (NSMs, SSMs) were evaluated through chi-square analyses. Fisher’s exact significance was used where the cell count requirement was violated. To evaluate the strength of the significant results, measure of association was used (φ). In addition, as the distributions of the continuous variables were asymmetric, the assumption of normality of the distribution of least squares was violated, and Mann–Whitney U tests were performed. All the statistical analyses were conducted using Statistical Package for the Social Sciences 25.0.

Modus Operandi of SSMs

SSMs do not have a lifestyle or psychosocial characteristics that raise the suspicions of those around them. They appear to be functional, and know how to show themselves in a favourable light and how to exploit their interpersonal relationships, even to the detriment of others. However, those close to them, particularly their spouses, are exposed to a problematic lifestyle (e.g., criminality, conjugal violence, coercive, and impersonal sexuality; James et al., 2019). The psychopathological profile of SSMs is characterized by sexual deviance (e.g., compulsive masturbation, sadistic sexual fantasies, sexual sadism), social isolation, low self-esteem, and a perception of women as the source of all their unhappiness (e.g., they perceive themselves to be rejected and humiliated by women). In addition, they perceive women to be objects in a constant state of sexual receptivity, and their own libido to be fundamentally uncontrollable. As a result of these cognitions, their emotional life is dominated by unverbalized negative emotions (e.g., anger), and they resort to sexually sadistic fantasies and behaviour (e.g., paraphilias, bondage, rape) to compensate for the poverty of their relational, emotional, and sexual worlds (James & Proulx, 2014). Their cognitions may be the source of relational problems (e.g., conjugal disputes) and anxiety (e.g., job-related stress, legal stress) in the year preceding their homicide.

SSMs’ homicides are triggered by tensions, especially sexual tensions (e.g., hypersexuality, deviant sexual fantasies), and they commit their crimes to act out extremely violent—and increasingly invasive and uncontrollable—deviant sexual fantasies (rape, sexual sadism, homicide; James et al., 2019). These fantasies not only provide a refuge that allows them to cope with their negative emotions and feel omnipotent, but are also powerful internal constraints that shape their modus operandi. In the 48 hr preceding their homicide, SSMs are sexually aroused and angry. Motivated by the need to act out their violent fantasies, they carefully plan their modus operandi and, notably, assemble an attack kit (e.g., weapon, rope). They contact their victim, whom they do not know, in the street, confine her, use restraints to neutralize her, and perform sexual acts on her that lead to the incidental tearing of her clothes (external constraint). It should be noted that confinement and the use of restraints has been reported to be associated with sexual sadism (Nitschke, Osterheider, & Mokros, 2009), and to allow the murderer to neutralize his victim (an external constraint; Sewall, Krupp, & Lalumière, 2013) and satisfy his deviant sexual fantasies (e.g., domination, possession, exertion of power, infliction of suffering, all of which are internal constraints) through a ritual involving torture (e.g., nonlethal strangulation; Hazelwood, Dietz, & Warren, 1992; Ressler et al., 1988). The victim’s physical suffering and psychological distress accentuate the SSM’s sexual arousal. During the assault, they are threatening and aggressive, and demand that the victim obey them, especially with regard to sexual acts (e.g., penile and digital penetration). Typically, they kill their victim by strangulation at an isolated location (e.g., wooded area).

Following their homicide, SSMs took precautions to avoid being identified by the police (e.g., destruction of objects, cleaning of the crime scene, disposal of objects and the victim’s identity card). There was thus little incriminating trace evidence at the crime scene, and their weapon was rarely found. Nevertheless, approximately 75% of them were apprehended within 24 hr. Apprehension of the remaining SSMs typically took much longer—in some cases, years. In the latter cases, the victim’s body was typically found in a skeletonized state (which renders identification of the murderer on the basis of trace evidence virtually impossible) near a body of water or wooded area. As the murderer’s sexual homicide is rarely a prefect incarnation of their violent fantasies, they obtain only fleeting and partial satisfaction of their deviant sexual needs, and tend to commit further homicides (James & Proulx, 2016).

Modus Operandi of NSMs

NSMs have a lifestyle and psychosocial characteristics that cause concern to, and raise the suspicions of, those around them. They are dysfunctional, have chronic alcohol problems, and may exhibit behaviours classically associated with this psychological problem (e.g., antisociality, criminality, violence; James et al., 2019). Their psychopathological profile is characterized by antisocial traits (e.g., hostility, impulsivity), and they perceive the world as a hostile and unstable place in which the exercise of power is necessary to avoid victimization. Consistent with these perceptions, they are emotionally detached and their behaviours are aggressive, impulsive, and unstable. They are hypersensitive, have poor self-control, and are unable to tolerate frustration and humiliation. Accordingly, when someone thwarts the immediate satisfaction of their needs, they feel diminished and attack the culprit to compensate for the perceived prejudice (James & Proulx, 2014). In the year preceding their homicide, they are dissatisfied, feel alone or rejected, and are failures in all spheres of their lives (social, occupational, sexual, criminal). In reaction to this life context, they develop negative emotions (e.g., humiliation) and consume alcohol to overcome their distress (Barbaree, Marshall, Yates, & Lightfoot, 1983; Marshall & Barbaree, 1990). However, when this coping strategy proves ineffective, their aggressivity and sexual arousal is exacerbated, and they turn to sexual coercion if a woman refuses to satisfy their sexual needs. Thus, while their homicides are triggered by a desire for sexual gratification, this desire is not accompanied by preexisting deviant sexual fantasies (James et al., 2019).

NSMs’ aggressive and impulsive personality traits, as well as their predisposition to sexual coercion, are not only the reflection of their antisocial personality and their hostility towards women (Beech et al., 2005; James & Proulx, 2014; Knight, 1999), but also powerful internal constraints that shape their modus operandi. In the 48 hr preceding their homicide, NSMs consume alcohol and are in a state of sexual arousal. Their lack of premeditation and their blitz approach strategy suggests an impulsive, brutal, and disorganized modus operandi. Their victim is often a student, less commonly someone who is intoxicated or possesses a physical disability. Depending on their relationship to her, their contact occurs in the street or at her home. They confine her, are threatening and aggressive, and force her to perform or submit to various sexual acts (e.g., penile and digital penetration). It should be noted that some NSMs are incapable of achieving erection, perhaps as a result of their alcohol intoxication (James & Proulx, 2014). They kill their victim by strangulation, at an isolated location (e.g., wooded area).

NSMs, in contrast to SSMs, do not appear to be concerned about leaving behind trace evidence (e.g., blood, sperm) at the crime scene, perhaps due to their state of intoxication (Beauregard & Martineau, 2014), and, unsurprisingly, are rapidly apprehended by the police (Balemba, Beauregard, & Martineau, 2014).

Models of Sexual Offending

The two modus operandi profiles identified in this study have many similarities with the offending pathways of sexual aggressors against women developed by Ward and Hudson (1998) and Proulx and Beauregard (2014), and with the modus operandi profiles identified by James and Proulx (2016). In fact, SSMs are similar to sexual aggressors following Ward and Hudson’s (1998) approach explicit and Proulx and Beauregard’s (2014) sadistic pathways, and to James and Proulx’s (2016) North American SSMs. Their lifestyle is dominated by deviance (e.g., compulsive masturbation, sexual sadism). To satisfy their emotional needs, they resort to violent sexual fantasies in which they commit crimes. However, as this sexual coping strategy is only partially effective, they become motivated to act out their fantasies in real life. Their crimes are carefully planned to allow them to exert a high level of control and domination over their victims, and their modus operandi is characterized by a variety of sexual acts (penile, digital penetration) and by a high level of coercion (e.g., torture). Following their sexual homicide, SSMs think about attacking another victim, and refine their modus operandi to reduce their risk of apprehension by the police and improve the congruence of their crimes and their fantasies. It is important to note that the modus operandi of French SSMs is very similar to that of North American ones.

French NSMs are similar to sexual aggressors following Ward and Hudson’s (1998) approach automatic and Proulx and Beauregard’s (2014) angry pathways, and to James and Proulx’s (2016) North American NSMs. They have an antisocial lifestyle characterized by behaviours (e.g., alcohol consumption) that may increase their risk of resorting to sexual violence. They are hostile and consider themselves entitled to demand satisfaction of their immediate needs. When a woman refuses to comply with their sexual demands, they resort to coercion. Their modus operandi is an extension of their lifestyle: impulsive, disorganized, and violent. This profile of NSMs is consistent with Beech et al.’s (2005) conclusion that these individuals’ impulsivity and violence may be the consequence of their cognitive deficits related to their perception of the world as a hostile and threatening place (i.e., the “world is a dangerous place” implicit theory), and that they justify their homicide as “as having resulted from something the victim said or did in a sexual context triggering a substantial sense of grievance held for some time against an intimate partner” (p. 1385).

It is noteworthy that although the modus operandi of the French and North American NSMs present more similarities than differences, their internal constraints differ: in French NSMs, it is the immediate need for sexual gratification; and in North American NSMs, intense anger. This difference may be partially due to the differences in sources of information: The French database collects information from interrogations and criminal records, while the North American database collects information primarily from institutional records and from post-sentencing semi-structured interviews, which may better detect murderers’ crime-phase anger. Cultural factors may also play a role. In their comparative study of French and Canadian sexual murderers, James, Proulx and Lussier (2018) demonstrated that the adult lifestyle of French and Canadian murderers differed, with sexual deviance predominating in the former group and antisociality in the latter. Consequently, the difference between the results of our study and James and Proulx’s systematic review—that is, need for immediate sexual gratification vs. the need to express intense anger—may reflect a greater predominance of deviant sexuality in the sample of French sexual murderers.