Abstract
This study examined all group sexual offending cases in the Netherlands between 1995 and 2009 (n = 26) in which at least one juvenile female offender (n = 35) had been adjudicated. Information from court files showed that the majority of juvenile female group sexual offenders have (inter)personal problems and (sexual) abuse experiences. The aims of the offender groups in committing the offense could be categorized in three themes: harassing the victim, sexual gratification, and taking revenge. The reasons why juvenile female offenders participated in a group could be categorized into group dynamics versus instrumental reasons. The findings are contrasted with findings on juvenile male group sexual offenders. Implications of the findings for research and treatment are discussed.
Introduction
Juvenile sexual offending is mostly committed by males. In the Netherlands, the country where the present study took place, about 850 juvenile sexual offenders were arrested by the police in 2009; 20 (2.4%) of these juvenile sexual offenders were females (de Heer-de Lang & Kalidien, 2010). Partly because so few juvenile females are prosecuted or convicted for sexual offending, little is known about juvenile female sexual offending.
The few studies that are available offer information about juvenile female sexual offenders in general but do not elaborate on important distinctions, such as between solo and group offenders. This is an important gap as we know from previous research that there may be important differences between (adult) sexual offenders who act alone and sexual offenders who operate with co-offenders (Bijleveld & Hendriks, 2003; Hauffe & Porter, 2009; Wright & West, 1981)
In fact, adult female sexual offenders very often co-offend. The co-offender is usually a man, who is often their intimate partner. In the literature, co-offending rates are reported that vary between one and two thirds of female sexual offenders (Bunting, 2007; Lewis & Stanley, 2000; Vandiver & Kercher, 2004), which is much higher when compared with adult and juvenile males for whom figures ranged between 10% and 20% in the United Kingdom and Australia and between 20% and 30% in the United States and South Africa (da Silva, Harkins, & Woodhams, 2013; Woodhams, 2009).
To our knowledge, hardly any research has focused on juvenile females who committed group sexual offenses. Almost all studies on juvenile female sexual offenders reported that the young women in their samples generally acted alone or did not specify co-offending (Fehrenbach & Monastersky, 1988; Hunter, Becker, & Lexier, 2006; Kubik, Hecker, & Righthand, 2003). A few researchers studied sexual co-offending of juvenile female offenders. Vandiver (2010) reported that about 50% of the juvenile females in her sample acted with a co-offender, while Hendriks and Bijleveld (2006) reported 70% co-offending. Wijkman, Bijleveld, and Hendriks (2014) reported that 58% of the juvenile female sexual offenders were group offenders. McCartan et al. (2010) reported only 14% co-offending. For all studies, it was unclear whether this co-offender was a romantic partner of the female offender or a friend or peer. Kubik and Hecker (2005) mentioned the presence of co-offenders but gave no further information about the extent of co-offending. Some researchers reported the presence of female sexual offenders in their sample, mostly only to help to procure the victim for the male members of the group and paid no further attention to these females (Horvath & Kelly, 2009; Porter & Alison, 2006).
In this study, we will focus on group sexual offending by juvenile females in the Netherlands. We analyze our findings with respect to the characteristics of the juvenile group sexual offenders, the characteristics of the group sexual offense, the background and manner in which the offenses evolved, and how offenders interacted with each other and with the victim (before, during, and after the actual offense). We also investigate the primary aims for committing a sexual offense among the investigated juvenile females, and we investigate different perspectives on why these offenders committed the sexual offense in a group. We contrast our findings with previous findings on juvenile male group sexual offenders. We analyze information from court files which contain validated data on personal characteristics as well as cross-validated offender and victim statements. Our sample is small; however, it is also complete in the sense that it comprises all juvenile female sexual offenders adjudicated in the Netherlands from 1994 to 2008.
Previous Research on Female Juvenile Sexual Offenders
Frey (2010) conducted a literature review on juvenile female sexual offenders. She found that their families are characterized by moderate to severe dysfunctioning (Mathews, Hunter, & Vuz, 1997; Tardif, Auclair, Jacob, & Carpentier, 2005) with many (>25%) shifts of caregivers (Roe-Sepowitz & Krysik, 2008) and inconsistent parenting (82%; Hickey, McCrory, Farmer, & Vizard, 2008). Reported sexual victimization varies from 26% (Roe-Sepowitz & Krysik, 2008), to 50% (Fehrenbach & Monastersky, 1988), to 100% (Bumby & Bumby, 1997; Hunter et al., 1993). Physical abuse and neglect as reported in these studies range from 12% (Roe-Sepowitz & Krysik, 2008) to 91% (Hickey et al., 2008). Hickey et al. (2008) reported that 77% of the juvenile females in their sample had experienced inadequate sexual boundaries within their family and that 49% had undergone treatment for mental health problems. Roe-Sepowitz et al. (2008) reported in their extensive study that almost 30% of the females faced problems with the use of alcohol and/or drugs, 49% experienced problems at school, and 53% reported delinquency before their sexual offense.
Because the majority of the studies are clinical (consisting of females who are treated for psychological problems after which in the clinical setting their roles as abuser become evident) and consist of case studies or descriptions of a limited number of young women, offense characteristics are often lacking. Some studies reported predominantly male victims (Hunter et al., 1993; Mathews et al., 1997), while others reported mainly female victims (Hendriks & Bijleveld, 2006; Vandiver & Teske, 2006). Most studies reported victims aged on average 11 to 12 years, and victims and offender were generally known to each other (Frey, 2010).
All in all, we see that previous research on juvenile female sexual offenders has reported multiple trauma such as sexual and physical abuse and neglect. Problems in domains such as school, antisocial behavior, and delinquency have relatively often been found.
Previous Research on Male Juvenile Group Sexual Offenders
Many studies have reported group sizes of about two to four offenders (De Wree, 2004; Horvath & Kelly, 2009; Porter & Alison, 2006; Woodhams, Gillett, & Grant, 2007). Some researchers reported offenders to have below average IQ scores (Bijleveld, Weerman, Looije, & Hendriks, 2007; ‘t Hart-Kerkhoffs, 2010), while others reported the opposite (Porter & Alison, 2006). Furthermore, offenders have often a problematic family background, with divorced parents. Offenders’ school performances are relatively weak, and truancy is common (Bijleveld et al., 2007; De Wree, 2004). Not much is known about the personality characteristics of juvenile male group sexual assault offenders, and some researchers reported that their personality profiles appear average and not different from other, non-sexual offenders (De Wree, 2004; Woods, 1969; Wright & West, 1981). Sexual abuse victimization of offenders themselves is reported to be low (between 6% and 8%; Bijleveld & Hendriks, 2003; Hauffe & Porter, 2009). Psychiatric problems were seldom reported: 4% (Porter & Alison, 2006) to 7% (Bijleveld & Hendriks, 2003).
Like group offending in general, group sexual offending must also be understood through the dynamics and processes of the group in which it is committed (Bijleveld et al., 2007; Hauffe & Porter, 2009; Wright & West, 1981). For example, male juvenile group sexual offenders more often use physical and verbal violence than male juvenile sexual solo offenders, and this can be explained by group processes (Woodhams, 2009). Hauffe and Porter (2009) suggested that juvenile male group offenders may more likely view victims as targets for behavior that benefits their status in the group, while lone rapists may see the victim more in terms of satisfying their sexual or relationship needs. Other studies found that offenders confirm and strengthen their status within the group by beating the victim (Bijleveld et al., 2007; Harkins & Dixon, 2009). Group bonding and elevation of masculinity have been reported as motivations for group sexual offenders by several researchers, as well as excitement and adventure (Bijleveld et al., 2007; De Wree, 2004; Franklin, 2004; Horvath & Kelly, 2009). However, sexual gratification was still reportedly an important goal of the offense (Bijleveld et al., 2007; De Wree, 2004; Franklin, 2004; Horvath & Kelly, 2009).
Perspectives on Co-Offending
Previous studies on juvenile female sexual offenders reported high co-offending rates, varying between 50% and 60% (Vandiver, 2010; Wijkman et al., 2014). However, because not much is known about the characteristics of group sexual assault committed by juvenile females, we also do not know how to explain why and when juvenile females commit sexual offenses with others. To understand the reasons of the juvenile females for participating in a group sexual offense, we build on theoretical literature about co-offending in general.
In the literature, three perspectives can be distinguished in explaining co-offending, namely, group influence, social selection, and the instrumental perspective (Weerman, 2003). These perspectives are used to explain juvenile co-offending, as well as co-offending in adult offenders, and to explain differences between juvenile and adult rates of co-offending. These perspectives differ in the presumed mechanisms that lead to co-offending or the reasons why offenders co-offend.
According to the group influence perspective, co-offending is the result of group influence and group processes leading to criminal behavior. These processes can be social learning or the acquisition of criminal attitudes, or group pressure felt by the members of the group. This perspective includes the theoretical notions of Warr (2002) who described a number of mechanisms that occur within juvenile groups. First, juveniles commit their crime in a group because they are afraid to be ridiculed by the other group members if they do not participate in the offending. Second, juveniles co-offend because they want to be loyal to their group. This means that they will engage in illegal behavior to preserve or solidify a friendship and that group members will protect each other when they are confronted with the police. Finally, juveniles may co-offend because of status enhancement. They can earn prestige and respect within in the group by participating in an offense.
The social selection perspective is built on the assumption that delinquent groups are formed because offenders select each other based on having the same characteristics, such as low self-control or a preference for non-conventional behavior, or being in the same place and context. Co-offending happens automatically when offenders stick together by social selection and when they happen to be in the same place when an opportunity for an offense occurs.
The last perspective is the instrumental perspective in which co-offending is deliberately chosen because it leads to an easier, more profitable, or less risky execution of a crime. A co-offender is selected because he or she can help to ease the execution of the offense or simply make the offense possible at all. The instrumental perspective can be seen as a more rational choice view: Offenders make a decision, and in this decision, they will decide whether they will need co-offenders or not to complete their crime.
The mechanisms that are presumed in the social selection perspective and the instrumental perspective have been identified as prevalent among juvenile male group sexual offending (Bijleveld et al. (2007); group influence mechanisms appeared to be much less prevalent.
Method
Sample
The data collection for this study was conducted in several steps. First, we collected information on all 143 cases registered between 1993 and 2008 with the Netherlands central prosecution service entailing a juvenile female defendant of at least one sexual offense. Of these 143 registered cases, 129 cases could be linked to a unique defendant criminal record. Fourteen cases could not be linked: Some appeared to refer to male offenders or to offenders who were too young (aged below 12 years) or too old (above 18 years) to be classified as a juvenile. Of these 129 suspects, 13 were acquitted and 22 were dismissed by the prosecutor for “technical reasons,” which implies that the prosecutor drops the case as he or she believes that there is insufficient evidence and the case will end up in acquittal. This resulted in 94 unique juvenile female sexual offenders. Some (n = 28) of these were hands-off offenders only, prosecuted mainly for human trafficking, the possession and distribution of child pornography, and indecent exposure.
For some of the remaining offenders, their case was dismissed by the prosecutor for policy reasons, such as that the defendant had started therapy or that the relationship with the victim had improved. Cases under this uniquely Dutch system of prosecutorial expediency are counted as convicted cases in academic research in the Netherlands (Wartna, Blom, & Tollenaar, 2008); therefore, we also included these cases in the current study. Our final research group contained 66 juvenile females who had been prosecuted for at least one hands-on sexual offense. The majority of this group, 38 juveniles (58%), committed one or more offenses with other offenders, while 28 females committed the offense alone (and were excluded from the current analyses). For these 38 female group sexual offenders, we analyzed their court files, and as far as possible, we tried to collect information about their co-offenders from these court files. For three offenders, the court file had been destroyed because of archival laws, so we finally used data on 35 juvenile female sexual offenders. They had offended in 26 different offender groups (defined as two or more offenders, in line with previous research about sexual co-offending, see, for example, Horvath & Kelly, 2009). Apart from these 35 juvenile female offenders, the case files also included 32 juvenile male co-offenders, 12 adult male co-offenders, and 3 adult female co-offenders. These offenders are not included in the sample but were important to understand the context of the offense.
The vast majority of the juvenile female group sexual offenders (71%) were ethnic Dutch. The remaining 29% were of Surinamese descent. The most common final educational level was lower general secondary education (n = 14) or special education (n = 9). At the time of the group sexual offense, the juvenile females were on average 15 years (SD = 1.6, range = 5.3).
Variables and Measurement Instruments
Offender and offense variables were scored from the court files, using the scoring tool for sexual offenders previously developed and used extensively in various studies by Hendriks and Bijleveld (2006) and by Wijkman, Bijleveld, and Hendriks (2010) for studying juvenile male and adult female sexual offenders. This tool consists of approximately 130 variables that cover personality characteristics, family functioning, traumatic experiences, school functioning, previous delinquency, and information about the offense characteristics. This tool was supplemented with a number of variables that are particularly important for this study on juvenile female group sexual offenders, such as experiences of physical or sexual abuse, sexual risky behavior, and relationships with and characteristics of any co-offender(s). It was also supplemented with 45 variables covering the variables of the group process: what happened in the group before, during, and after the offense. Court files in the Netherlands always contain the charge as well as the judicial decision, in which it is specified as to what offense the defendant was charged with and was found guilty of, as well as a verbal description of what behavioral acts were declared proven by the court. Each court file contains a detailed charge by the prosecutor, which lists the acts purportedly committed by the defendant. It also entails an exact description of the physical acts committed, and information about victim(s) as well as their relation to the offender, where the offense took place, and on what date and time. Almost every court file contained transcripts of the police hearings of the offenders, reports by police officers in various phases of the case, victim statements, and sometimes witness statements. If the prosecutor, judge, or the lawyer had requested psychological or psychiatric screening, the court file also contained the psychiatric and/or psychological reports. Such screenings are always carried out by certified forensic psychologists and psychiatrists. Reports are based on clinical judgment as well as on standard validated tests used in Dutch clinical assessments by (forensic) psychologists and/or psychiatrists, such as the Raven, the WISC (Wechsler Intelligence Scale for Children), and the MMPI-A (Minnesota Multiphasic Personality Inventory-Adolescents); however, the tests that were used varied between files and changed during the years that were covered by the current study. Moral development was mostly measured by clinical judgment of the certified forensic psychologist and psychiatrist. They asked the offender questions with which they could estimate the level of victim empathy and whether the offender regrets what she had done. These questions were often summarized and labeled as moral development.
Offender characteristics coded are intelligence level, school performance, neuroticism, self-esteem, suggestibility and psychopathology, family characteristics (such as experienced separations, relatives offenders lived with, violence between parents), intimate relationships, and alcohol or drugs abuse. Offense characteristics include the number of co-offenders/accomplices, the nature of the sexual acts, use of violence during the crime, relationship between offenders and victim, number of victims, and age and sex of the victim.
Finally, we reconstructed each offense (how it evolved and how the offenders interacted with each other and with the victim) from the court files. In assembling the reconstructions, we noted as factually as possible all behavior and communication between the offenders shortly before, during, and shortly after the offense. Not all offense situations could be reconstructed to every detail: Sometimes the court files lacked clarity about the (leading) role of the offenders, there was not enough information available about all offenders, or offenders’ statements were inconsistent. Whenever statements were inconsistent or unclear, we let the statements by the victim and/or findings by criminal justice officials prevail.
The results as reported in this study are a description of the characteristics of the offenders at the time of the group sexual offending. All court files, which contained the case number under which the court file at the criminal courts was stored, were scored before criminal career information was retrieved from rap sheets held at the centralized criminal record files office (the Netherlands JustId Office in Almelo) to prevent bias in the scoring process. Permission for this study had been obtained from the Prosecutor General and the Ministry of Justice.
Method of Analysis
The first step in the data analysis is to describe the characteristics of the juvenile female offenders, the offender groups, their victims, and their offenses. For the description of the characteristics of the perpetrators, victims, and the offenses as noted in the criminal and judicial records, we present simple frequency counts. Some variables could not be scored for all females. This is because, in the Netherlands, the prosecutor or the defense may only request psychological or psychiatric screening if psychological issues are supposed to be relevant for judging culpability or the need for treatment. Therefore, psychological and/or psychiatric screening had been requested for a subset of all 35 juvenile females. Whenever we report characteristics that may be supposed to be elevated in the subset that did receive screening, such as disorders, we do not presume that the percentage measured in the screened group can be generalized to the entire sample and simply report the actual number of cases diagnosed with a disorder (e.g., “13 juvenile females were diagnosed with ADHD”).
After the description, we analyzed the offense situation in a more extensive way to discern the interaction between offenders, the aim of the sexual offense, and the reasons why the offenders committed the offense in a group. For these analyses, we build on the “reconstructions” of each offense, containing information on the initiation, continuation, and ending of the offense; information on all offenders’ motives as expressed at hearings and through acts and expressions during the offense; and victim characteristics. We combined offenders and their offenses into groups with similar offending purposes based on the context and sequence of the events during the offenses (the onset, course, and ending of the offense; interaction between victim and offender, and offenders themselves). Because it was in most cases unclear who initiated the offense (offenders denied their involvement, or they were accusing each other of starting the offense), we decided to take the most often mentioned aim as the purpose of the group and not the aim of each offender separately for committing the offense. There was one group for which it turned out impossible to reconstruct this, so we have the aims of 25 groups. These analyses resulted in a small number of relatively homogeneous offending “themes,” predominantly based on the expressed aims. The interrater reliability, measured as percentage agreement, of this analysis was .84.
After constructing the aim of the group for committing the offense, the second and third author (both criminologists) read the coding forms and allocated all offender groups to one of the themes. After coding, re-viewing, and recoding all allocations, there was agreement about the general themes. The interrater reliability, again measured as percentage agreement, of this analysis was .88.
Finally, we coded the main reasons (based on the general perspectives on co-offending) for the juvenile female offenders to commit the offense with (a) co-offender(s). We started with the three perspectives as mentioned in the “Introduction” section. This coding was sometimes problematic. Because court files are primarily designed to assist the prosecutor in the criminal proceedings and not to reconstruct group dynamics, it was difficult to distinguish the group influence perspective from the social selection perspective. There was often a lack of clarity about the timing of an eventual group influence, whether the group originated shortly before the offense or whether the group had its origins long before the offense, and offenders were also engaged in conventional activities. Thus, during the re-viewing and recoding, we decided to combine the group influence perspective and the social selection perspective and rephrased this as the group dynamics perspective (to a certain extent, both group influences and social selection are based on group dynamics). So, in the end, we coded two main reasons for co-offending: group dynamics and instrumental considerations. For two offenders, we could not classify their reasons for co-offending, because too little or unclear information on the offenses and the formation of the group was present in the court files.
Results
Characteristics of the Offender Groups
The 35 juvenile females had offended in 26 offender groups comprising in total 82 offenders. In all but 2 groups, both male and female offenders were involved. The average size of the offender groups was 3 offenders. The smallest groups consisted of 2 offenders, while the largest consisted of 7 offenders. The average age of all offenders was 16.9 years (median = 15, modus = 14, minimum = 10, maximum = 56). Fifteen groups (58%) consisted of only ethnic Dutch offenders, 5 groups (19%) had a completely non-ethnic Dutch background, and 6 groups (23%) had a mixed background. Ten groups constituted a stable group, which means that they were seeing each other on a regular basis: They were, for example, romantic partners, siblings living in the same house, or juvenile females who were each other’s most intimate friend. All offenders were acquaintances or relatives of each other.
Five groups offended against more than one victim, with a maximum of four victims. The offenders were usually acquaintances or relatives of the victim (n = 24); only two groups victimized a person unknown to them.
Characteristics of the Juvenile Female Group Sexual Offenders
About one quarter of the juvenile females (23%) lived in a two-parent family, while 49% had divorced parents. For 10 juvenile females, no information was found about their family background. An ambivalent or bad relationship with parents was reported by 34%. Six juvenile females (17%) mentioned that one or both parents were alcohol dependent, and 14% said they had a delinquent father or brother. One third of the juvenile females reported they had a boyfriend, and 20% had a romantic partner who was their co-offender.
A little below one in three (29%) of the juvenile females functioned at borderline intellectual level (IQ = 71-84) or had mild mental retardation (IQ = 50-70). The moral development of one third was judged as below average, and about one third (33%) were described as (very) susceptible. Almost one in three (31%) was rated as having low self-esteem. Antisocial behavior prior to the sexual offense was reported for 37%. Based on the psychological and psychiatric reports, 11 juvenile females were diagnosed with a psychiatric disorder (n = 7) and/or showed signs of a personality disorder (n = 8). The psychiatric disorders that were diagnosed were conduct disorder (n = 5), oppositional defiant disorder (n = 1), and attention deficit hyperactivity disorder (n = 1). Four juvenile females were reported to have traits of a borderline personality disorder, one juvenile female was reported to have traits of an antisocial personality disorder, and one was reported to have traits of a dependent personality disorder.
The majority of the juvenile females (54%) had reported abuse or neglect: emotional and/or pedagogical neglect (26%) and physical abuse (29%). Sexual abuse was reported by 31% of all juvenile females; the majority of these juvenile females had been victimized by a non–family member, for instance, an acquaintance. One third of the juvenile females had exhibited risky sexual behavior such as sexual intercourse before age 14 years (21%) or sexual soliciting (13%). About one quarter had been bullied at school; 34% reported truancy. More than one third had had behavioral problems at school such as fighting, threatening other students, lying, stealing, or had conflicts with fellow students and/or teachers. About a quarter reported they had committed one or more offenses prior to the sexual offense, mainly theft, violent offenses, and vandalism.
Overall, for 63% of the juvenile females at least one problem was reported in the aforementioned domains (e.g., presence of disorders, experiences of abuse/neglect, sexual risk behavior, and problems with authorities or anger management).
Victim and Offense Characteristics
There were 33 victims in total. The average age of the victim was 14 years (minimum = 7, maximum = 23), and the majority (88%) of the victims were female. Five groups committed multiple offenses against the same victim(s) or against other victims.
Mostly, differences between the age of the victim and the offenders were small. However, the three youngest victims (7, 9, and 9 years) were all abused by romantic couples consisting of a juvenile female and her adult male partner.
The offenses were mostly committed in the house of the offender (46%) or in public places like a wood, a park, or a playground (46%). One group assaulted the victim in her own house (for one group, the crime location was unknown).
The role of the female offenders in the offense varied. Some were active and participated in the sexual acts or battered or threatened the victim (58%), while others provoked other group members or created opportunities for other group members to commit the offense, for instance, by introducing the victim to the offenders or by making no effort to stop the abuse. This shows that juvenile females charged for sexual offense are not a homogeneous group: Almost half of them did not actually committed a hands-on sexual offense. Nevertheless, they are legally considered as hands-on sexual offenders, because their actions were seen as having the criminal intent of a hands-on sexual offense.
Nineteen groups (73%) used some kind of violence during their offense. Fourteen groups used physical violence (beating, kicking), 3 groups used verbal violence (threatening the victim if (s)he would not cooperate), 4 groups humiliated the victim (spitting, yelling, and insulting), and 1 group destroyed personal items belonging to the victim. Four groups used a weapon during the offense to threaten the victim (gun) or to remove the clothes of the victim (knife).
Interaction Between Offenders and Between Offenders and Victims
Before the offense
In total, 16 groups made preparations for the offense, in various manners. In 8 groups, the offenders spoke with each other (explicitly) about having sex. In 1 group, the female offender asked her co-offender whether he would like her to “arrange” something for him with the victim, while in another group one of the male offenders told his sibling he wanted to have sex with the girl who was walking on the street. In other groups, offenders discussed with each other what kind of acts they were going to perform on the victim. Offenders also discussed where the offense could take place.
In 11 groups, the offenders talked with the victim before the offense about sexual acts. Offenders said to the victim they wanted to have sex, asked what kind of sex the victim would like to have, and in one case, offenders told the victim to participate in the sex or otherwise the offender would sell her to a friend. One couple first asked the victim whether she wanted to join them in having sex.
Three groups prepared themselves in a more practical way: One group bought rope so they could tie the victim, one couple doped the victim and her boyfriend so she could not resist and he could not disturb their threesome, and two groups let the female co-offender arrange for the female victim to be present. One group discussed in advance the way they were going to hit the victim and who was going to cut off the victim’s hair.
All in all, it appears that in a majority of offender groups, some kind of preparation for the offense had been made. Mostly, plans or expressions of intent had been voiced. Instrumental preparations were much rarer, probably because the presence of co-offenders rendered additional support unnecessary. However, even though it was clear in most cases what was going to happen just before the offense was initiated, the offenses do not appear to have been planned long beforehand.
During the offense
In the majority of the groups (62%), the juvenile female offender(s) participated in the sexual acts or in the violence. They abused the victim themselves, they held the victim so he or she could be abused by co-offenders, or they physically abused or verbally threatened the victim. In 10 groups, the juvenile females had no hands-on contact with the victim. For instance, they provoked the more active offenders to perform certain (sexual) acts, they introduced the victim to the other offenders while they knew that the victim was going to be abused, or they did not intervene while victims were abused in their house.
In 8 groups (30%), the victim was fondled, for example, by touching the breasts of the victim or touching her genitals. In 14 groups (54%), the victim was sexually penetrated; in 2 groups (8%), the victim was forced to perform oral sex; in 2 groups (8%), the victim was forced to undress or to show his penis to the offenders.
Only in five groups (19%), a distinct leader of the offense was present. Mostly, offenders collaborated without directives. Some offenders communicated with each other during the offense about practical things such as getting the victim in the right position. Evidence that the offenders tried to stop each other during the offense did not emerge in any of the group cases. When there was genital penetration, none of the male offenders used a condom.
In four groups, at least one of the offenders had used drugs or alcohol before committing the offense: In one group, all offenders had used alcohol; in another group, the offenders had committed multiple sexual offenses during which they had sometimes used alcohol and/or marijuana, while in one case the male and female offender doped the victim with speed and alcohol before they sexually abused her. In one case, only one juvenile male offender had probably used soft drugs before committing the sexual offense.
Verbal resistance (saying they did not want this or trying to talk the offenders around to stop the abuse) was shown by 12% of the victims, 31% resisted physically (by running away from the offenders, extreme movements so the offenders could not penetrate or tearing loose from the offenders).
In conclusion, in the majority of the groups there was hands-on contact between offenders and (the) victim(s). Penetration occurred in a majority of cases. A minority of victims resisted verbally or physically. The female offenders had an active role in a majority of the offenses, even though not all performed sexual acts.
After the offense
In 50% of cases, all offenders confessed to the offense. In 12% of the cases, none of offenders confessed, and in the remainder, some of the offenders confessed. In about one third of the groups (38%), offenders said they realized during the offense that the victim did not like the sexual acts. None of the offenders reported they had threatened the victim to prevent she/he would go to the police, and neither did offenders construct a scenario that could have been told to the police if they were questioned. Some of the offender groups (12%) brought the victim home after the offense, while the majority of the offenders resumed their daily activities like going back to school or by having dinner at home.
Seven offenders reported they regretted what had happened; three of them were feeling sorry for the victim, four were feeling sorry for themselves and regretted the trouble they had to deal with now. Four offenders accused each other of initiating the offense.
Thus, it appears that in quite a number of cases, offenders realized during the offense that they had gone too far. However, statements about the offenders’ perception of the offense were primarily obtained from the offenders after they had been charged with the sexual offense. Thus, defense strategies or neutralization techniques may have influenced the statements. Few preparations were made to prevent prosecution. The offenses often seemed to have an almost casual nature: After the offense, most offenders simply resumed what they were, there was little reliving of the offense.
Aims of the Offender Groups for Initiating the Offense
The primary aim for initiating the offense varied. Three qualitatively different categories or “themes” emerged, which are illustrated below with examples.
Ten groups committed the offense because of what we label “harassing the victim.” This was expressed by the offenders, for example, by stating that what happened was just to make fun. This aim can be illustrated with the following two examples:
A and B were hanging around at a snack bar. When they saw the victim (an intellectually disabled boy) approaching, the juvenile females started hissing sexually oriented remarks at him, told him to masturbate, and A took the victim’s hand and helped him. When he ran away, they chased him and forced him again to masturbate. At one point, the victim pulled up his trousers and A started hitting him. B extinguished a burning cigarette on his upper leg. The victim started yelling and cursed at them, after which A kicked him in his genitals. A police report mentioned that A had said that the victim sometimes masturbated in public and that it was stupid of her to beat someone up who was intellectually disabled. She was surprised herself with the sexual content of her aggressive behavior. She was known with criminal justice authorities for aggression regulation problems.
The victim, a woman of 21, was waiting for the subway when she was surrounded by a group of three juvenile females and a juvenile male. The juvenile females pushed her against an advertising column and touched the woman’s face, breasts, and genitals (over her clothes). The victim was asked if she was scared, and one girl was performing some kind of lap-dance on the victim. During this, the juvenile females were continuously laughing. The juvenile male was not participating in the touching, and the victim heard him saying to the juvenile females that they should stop. After a couple of minutes, the victim managed to push the juvenile females away and to get into the metro that was just arriving. A police report mentioned that the offenders said they committed the offense “because it was fun.”
Second, 10 groups committed the offense with a predominantly sexual aim. Nine of these groups consisted of couples, and 6 of these groups had female victims who were considerably younger than the male co-offender.
H, a 17-year-old juvenile female, wanted to experience a threesome. Her boyfriend, K, knew this and together they picked up another couple in a pub, after doping them with alcohol and speed. When they were all at the couple’s home, the female victim refused to have sex with H. K threatened to shoot the victim’s boyfriend if she didn’t have sex with H. At gun-point, the female victim had oral sex with H, while K was watching. H told the police it all was a little game, and the victim should not be so fussy about it.
C and D, two juvenile males of 13 years old, were hanging around at a playground when they saw three juvenile females approaching. C was curious how far he could go with a girl, and suggested to D they should try to finger one of the juvenile females, and he asked one of them. She stated that she hesitated but ended on the lap of C. She had unbuttoned her trousers herself and the juvenile male embraced her. The second of the other juvenile females, F, told him to hurry and she grabbed his hand and pushed his hand into the panties of the victim. When one of their parents was walking into the direction of the playground, the victim managed to free herself. The victim told the police that C and F had threatened to beat her up if she would say something about what happened to her parents. C told the police that he just wanted to know how far he could go with a girl and that he didn’t know the victim was only 10 years old. F reported that she had grabbed the hand of C not to prevent what had happened, but to be more supportive to the victim.
A third and last theme was found in five groups in which the initiation of the offense started with revenge. Three groups initiated the offense because they regarded the victim as sexually too obtrusive or because she had (had) an affair with a(n) (ex-)boyfriend of one of the offenders. One group assaulted the victim because she was gossiping about two of the offenders, and one group assaulted the victim because they were annoyed by her theatrical and exaggerating behavior.
P, a 15-year-old juvenile female, was together with one of her female peers, D, walking in the city. They spotted Z, an acquaintance, who was now having an affair with P’s former boyfriend. P and D decided to teach Z a lesson and invited the girl for a drink. When they were walking through a park, they hit the girl and forced her to undress herself. P forced the girl, by yelling and shouting, to put sticks into her own vagina. As this was happening, three male peers were crossing the park and they saw the three juvenile females. P told the juvenile males it was okay to rape Z because she was a slut any way. One of the juvenile males raped Z, and after this, they all left Z behind, bleeding heavily.
Reasons and Explanations for Participation in a Group Offense
For the majority (63%) of the juvenile females, group dynamics played a dominant role. These adolescent females had expressed statements in which it was said that she wanted to belong to the group, could not resist joining in, felt pressurized, did not think about what would happen as everybody had joined in, and thought it to be normal. Some of the juvenile females who committed the offense together with their male intimate partner reported they were forced (physically and/or emotionally) to join in the abuse. These are some illustrative statements derived from police reports:
“I couldn’t resist the pressure of the group and was afraid they would make fun of me if I would not join them.”
“I did it because I was afraid of my co-offender, he could be very threatening if I did not agree with him.”
“I wanted to belong to the group and therefore I joined them.”
“I was dragged into everything by my co-offenders; I could not assess the consequences.”
The remaining juvenile females (31%) co-offended mainly for instrumental reasons. In the court files, these adolescent females had made statements that made clear that having a co-offender had made the offense possible, easier, or more profitable. The following are a number of illustrations of this co-offending perspective:
A knew that the victim had sex with X, the boyfriend of A. Together with B, she decided to take revenge on the victim. B was the brother of X, and B felt rejected by the victim because she had refused to have sex with him. A and B lured the victim into a wood, and while A was hitting the victim, she was raped by B.
D thought that the victim was behaving sexually too obtrusive toward her boyfriend. She arranged a couple of friends to teach the victim a lesson. The victim was sexually harassed by the group.
Discussion
Not much is known about juvenile female group sexual offenders. This study analyzed data on all juvenile females adjudicated for a group sexual offense in the Netherlands in the period 1993-2008; as such, in a statistical sense, our sample constitutes a population. While the sample size may not be large in an absolute sense, we have presented unique material about a subject that, as far as we know, has never been studied this extensively. We were able to collect rich data on offender background characteristics. Also, using offender and victim statements and information collected by the criminal justice authorities as laid down in the court files, we could reconstruct the offenses and identify categories of aims for the offenses and motives for co-offending. Although our sample was derived from the prosecutorial registration system which is not developed for scientific research, it is a (virtually) complete registration of all cases registered with the prosecution service.
Overall, three findings stand out: Juvenile females who commit a group sexual offense are characterized by (sexual) victimization experiences and other (interpersonal) problems; there is heterogeneity in the aims of the sexual offense, with three main themes emerging from the data; and there is heterogeneity in the reasons why they committed the offense in a group.
With regard to the first main finding, juvenile female group sexual offenders appear marked in terms of psychological disturbance and victimization. For 63% of the juvenile females, disorders, victimization experiences, risky sexual behavior, and/or anger management/authority problems were reported in which they partly seem to differ from juvenile male group sexual offenders. For juvenile male group sexual offenders, many problems have been reported in the literature concerning family background and problems at school, but not many striking characteristics were reported or present concerning their personality and (sexual) victimization experiences. The prevalence of victimization of sexual abuse in this group of juvenile females stands out, though it is commonly assumed that females have an increased risk of sexual victimization when compared with males (Zahn et al., 2010) and that in particular juvenile female offenders have often been abused sexually (Wong, Slotboom, & Bijleveld, 2010). We were unable to find Dutch norm scores for psychological and psychiatric disorders, so we cannot gauge these findings.
The committed sexual acts were quite different when compared with group sexual offenses committed by juvenile males. Bijleveld et al. (2007) reported that 95% of the all-male juvenile group sexual offender groups penetrated the victim. This was strikingly lower in this female group, at 53%. In that sense, the group sexual offenses committed by these females may appear less serious than those committed by males, and one could regard female group sexual offenders, therefore, as a group of offenders with a different profile. However, the juvenile males reported on by Bijleveld et al. had undergone personality screening for their share in at least one group sexual offense, and in the Netherlands, personality screening is often requested when an offender had committed a serious offense or when the public prosecutor supposes that the offender committed the offense under the influence of disorders. The difference in severity of the sexual crimes may thus be due to the different manner in which the samples were construed.
Another difference that emerges is the fact that none of the female offenders reported they had threatened the victim to prevent she/he would go to the police, nor that they had constructed a scenario that could have been told to the police if they were questioned. This could imply that the offenders maybe had not expected that the victim would report the offense at the police or that the offenders had not thought about the possible consequences of their behavior. This assumption could be supported by the fact that when there was genital penetration, none of the male offenders had used a condom, which is quite low when compared with the 50% of male juvenile group sexual offenders who used a condom (Bijleveld et al., 2007). However, the latter could also be due to offenders being confident no police reporting would ensue. It could also mean that offenders did not have any experience at all with planning and committing sexual offenses. Last, it could mean that offenders were not afraid of pregnancies because their victims were young.
We found three main aims for initiating the sexual offense, namely, harassing the victim, sexual gratification, and taking revenge. The harassment aim was more prevalent here than among male group sexual offenders. However, this aim bears some similarity with that of juvenile males who tried to impress their male co-offenders to enhance their status in the group. Sexual gratification as an aim was present in this study but less often than in juvenile male sexual offenders (Porter & Alison, 2006; Bijleveld et al., 2007). All groups with a sexual aim victimized younger female (child) victims, which is not common in juvenile male sexual offending groups. The third initiation aim, taking revenge, was also found in Kubik and Hecker (2005), who examined juvenile female sexual solo offenders. Apart from particularities found in sexual sadism (Marshall & Kennedy, 2003), taking revenge is not a commonly found aim in juvenile male group sexual offenders. In that respect, therefore, female group sexual offending appears to differ. Gannon and Rose (2008) reported a similar offending theme in adult female sexual offenders, and possibly, this specific theme of taking revenge (by humiliation) is unique for female group sexual offenders. In that case, this raises the question whether these juvenile females should be seen as sexual offenders or as offenders who commit a violent offense with the sexual element added for additional humiliation.
We distinguished two reasons (based on different perspectives on co-offending) for committing the sexual offense in a group: because of group dynamics and because of instrumental reasons. When examining these perspectives, we see that some juvenile females reported they were dragged into the abuse by their peers. Those who committed the offense together with their male intimate partner in fact often reported they were forced (physically and/or emotionally) to join the abuse. Especially, the latter, being forced by a male co-offender to participate in the abuse, is often reported by adult female sexual offenders (Harris, 2010). The instrumental perspective that we identified may in fact go two ways: The female offender uses a male offender to humiliate the victim or to aid in carrying out the offense or the male offender uses a female to get access to other females or to younger children. We believe that the first instrumental reason might even be subdivided, namely, into juvenile females who are physically or practically not able to commit the offense and juvenile females who need their co-offenders for mental support to perform their act.
All in all, we see there are some striking differences between juvenile females and juvenile males who are involved in group sexual offending. This is remarkable because in almost all offender groups that we studied, at least one male offender was involved, so one would expect that the differences between group sexual offenses committed by solely males and those committed by mixed groups (with in most cases one or two females present) would not be that large. The differences may however be explained by general characteristics of male and female friendships. As mentioned by Weerman and Hoeve (2012), male friendships are generally more characterized by hierarchy and competition, while female friendships tend to be more strongly characterized by intimacy, emotional involvement, and confidentiality. Bijleveld et al. (2007) found that male bonding and showing male competence is an important feature of male group sexual offending. Exposing male competence, such as showing a condom containing sperm to the other offenders or showing sexual competence, is arguably not found in this sample because there might be no need or reward to show these male competencies to a (partly) female audience.
Some juvenile females seemed to have engaged in the offense not for their own sexual gratification (which was a common aim within juvenile male group sexual offenders) but for sexual humiliation to punish the victim or to take revenge. This particular type of offending seems to be unique to female sexual offenders in groups. The high prevalence of female victims of this kind of sexual offending is unlikely to reflect the sexual preferences of the female offenders. Instead, many of these juvenile females simply wanted to take revenge on someone who was a female. Being female themselves, they may have picked the group-wise and public sexual humiliation to achieve their goal best. The victim was chosen because the offenders wanted to take revenge on specifically this victim, and not because they wanted to get sexual gratification from having sex with any female. In juvenile male group sexual offending, we see that offenders more often victimize a female victim, regardless of who this may be, because their main goal is to achieve sexual gratification. Interestingly, male juvenile group sexual offending also often included the element of humiliation, but here, it is usually not applied to take revenge but to experience power and show off to male accomplished (see Bijleveld et al., 2007). However, we want to note that our comparisons with findings from studies on juvenile male group sexual offenders may be influenced by method differences, and further systematic comparisons would be needed to make better inferences about any differences.
One of the limitations of our study is the small sample size. This limitation is difficult to overcome as we studied all juvenile female sexual offenders known to the criminal justice authorities in the Netherlands since 1993. This small sample size implies that we could not do meaningful statistical analyses of differences between subgroups, for example, between the three categories with different aims for their sexual offense, or the female who actually committed the sexual acts themselves and the females who were merely accomplices or co-perpetrators. Another limitation is that we were unable to statistically test the differences between juvenile male sexual offenders and juvenile female sexual offenders involved in group sexual offending so that definitive statements about distinctive features for juvenile female sexual offenders are not warranted. Third, our sample consisted of juvenile female sexual offenders who were caught and prosecuted for their sexual offense, and it is likely that only the more severe offenses are reported to the judicial authorities. These convicted offenders are not necessarily representative of all young women who commit group sexual offenses as much offending is unreported. General reasons of victims for not reporting sexual victimization are blaming themselves for being raped/assaulted, fear of repeat victimization when the victim knows the offender, regarding the offense as minor, or a belief that reporting the crime would not make a difference (Fisher, Daigle, Cullen, & Turner, 2003). Furthermore, statements about victimization and the offenders’ perception of the offense were primarily obtained from the offenders after they had been charged with the sexual offense. Thus, defense strategies or neutralization techniques may have influenced the statements. It is also possible that the juvenile females were afraid or ashamed to report any sexual arousal while committing the offense. For that reason, it is likely we may have underestimated the extent to which juvenile females offended out of sexual motives. Similarly, the forensic psychologists who are assessing the offenders for psychological screening may not have registered these females’ sexual feelings and motives because they have never been educated in their clinical training how to assess sexual abusing behavior shown by females, because they do not think of women as sexual offenders or because they think that sexual arousal is not the cause of female sexual offending behavior and can therefore be neglected during the assessment (Denov, 2004). Finally, moral development was assessed by clinical judgment, which is probably not a very strong way of assessing.
Implications for intervention and treatment must be linked to the aims for committing a sexual offense. However, all juvenile females committed their offense in a group, which implies that group dynamics may be taken into account. For example, for some of these offenders, it could be risky to treat them in a group for their deviant behavior (Gifford-Smith, Dodge, Dishion, & McCord, 2005). Treating troubled, serious offenders like these juvenile females in groups may even increase adolescent problem behavior and negative life outcomes in adulthood, because group members with the same background may reinforce each other’s deviant behavior (Dishion, McCord, & Poulin, 1999). Clinicians should investigate whether juvenile female group sexual offenders are susceptible to peers in a usual, normal way (comparable to other adolescents), or whether they are highly susceptible as part of the offender’s general personality traits. Furthermore, it seems warranted to assess and treat offenders who acted as a leader during the offense different than offenders who have been mere followers.
Treatment should always be tailored to the individual needs of offenders. If we keep in mind that many of these females are marked in terms of trauma and that the prevalence of disorders appears elevated, interventions may need to address these issues first. It is to be doubted whether interventions should focus on sexual re-offending prevention similarly as in males, as the level of sexual recidivism in adult female sexual offenders is quite low—about 3% (6.5-year follow-up; Cortoni, Hanson, & Coache, 2010) and is low, 10%, in juvenile male group sexual offenders as well (10-year follow-up; Hendriks & Bijleveld, 2005). We expect it is likely low in this group of juvenile female sexual offenders too.
When we focus on the three different aims we identified from the data (harassing the victim, sexual gratification, and taking revenge), different intervention goals appear. First, juvenile females who committed the offense to harass a victim could benefit from interventions that focus on increasing cognitive and social skills with which self-control and (social) problem-solving skills can be improved and which encourages offenders to consider the consequences of their behavior. Examples of suitable interventions are the Reasoning and Rehabilitation Program (Robinson & Porporino, 2003) or the Enhanced Thinking Skills program (Sadlier, 2010). Second, because the offenses committed because of sexual gratification appear mostly instigated by the sexual motivation of the male co-offenders, juvenile females within this group will probably not benefit from interventions that focus on the sexual content of their offense. For these women, interventions should likely focus on increasing self-esteem and self-efficacy, so that they can better deal with future (high-risk) situations and stand up against their intimate partners (Ford, 2010). Juvenile females who took revenge on their victim, the third motive, gain probably the most of social-cognitive interventions that focus on increasing their behavioral and problem-solving skills (Hipwell & Loeber, 2006). An example of a suitable intervention is the Aggression Replacement Training (Goldstein, Glick, & Gibbs, 1998). While the focus of treatment may thus be primarily be such non-sexual issues, assessment and treatment should not overlook the sexual aspects of, and motivation for, the crime.
In summary, this research shows that juvenile female group sexual offenders are an interesting category of offenders, in which several subgroups can be distinguished. In only part of the cases, the main aim was to get sexual gratification, often not for the female co-offenders but for someone else. In the majority of the cases, the main aim of the sexual offense seemed to be to harass the victim or to take revenge and humiliate the victim. The characteristics and group dynamics differ largely from those that were found previously among male juvenile group sex offenders. Another important observation to be made is that although female juvenile group sex offenders appear to be a very heterogeneous group, the majority of them are characterized by having (inter)personal problems and (sexual) abuse experiences.
In future research, it would be interesting to study (sexual) re-offending of juvenile female group sexual offenders, because as far as we know, no studies have been conducted on this topic. Elaborating on the previous research suggestion, it would be useful to study the role of this particular group sexual offense in the criminal career of the juvenile females and to examine whether it was an once-only act, or the start—or maybe the end—of a criminal career. To conclude, there is some controversy in the literature about whether “duos” (in this study, the romantic couples) should be considered as a group and therefore included in the group dynamics research (da Silva et al., 2013). When it is possible to increase our sample, it would also be interesting to investigate what distinguishes such duos and larger groups.
Footnotes
Acknowledgements
The authors would like to thank the anonymous reviewers for their helpful comments on previous drafts of this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.