Foreword | This paper summarises the results of a small-scale study into the online interactions of suspected paedophiles with undercover Australian police officers posing as male children. The study provides insight to an under-researched area of how persons with a sexual interest in male children interact with potential victims and whether these interactions differ from online engagements with female children.
The findings provide an interesting and persuasive case to further examine online strategies used by persons wishing to exploit male children, as it highlights potential differences in approaches when compared with interactions between suspected offenders and female children. The differences include less aggressive online behaviour and language, with an emphasis on the male child’s sexuality and sexual experience. These strategies appear to assist the formation of an online friendship which precedes sexual topics, rather than the reverse as is largely reported in literature. These findings are likely to have implications for law enforcement and child protection initiatives.
There is limited research articulating the online interactions of persons who engage and solicit male children for sexual exploitation. As females constitute the majority of known victims, the majority of existing research has used data relating to female victimisation and generalised female-focused offender typologies to represent offending against young males. Research by Krone (2005) into online police ‘stings’ in Queensland identified a peculiarity in online offender interactions; when police assumed the online persona of a male child, no sexual solicitations were received and no offences were committed. This stood in contrast to the exchanges experienced when police posed online as a young girl, where interactions were typically characterised as highly sexualised and geared towards an immediate or short-term online sexual encounter.
This study describes the interactions of persons who have engaged with a male child persona used by undercover Australian police for the purposes of identifying persons who seek to sexually exploit children. Thematic analysis was conducted on chat room conversation logs for online cases conducted by the Queensland Crime and Misconduct Commission, Queensland Police Service and Australian Federal Police. These were compared with semi-structured interviews conducted with police officers working in taskforces established specifically to combat online child exploitation.
Greater knowledge of how Internet offending against children occurs is vital in combating the global problem of online child exploitation. The objective of this study was to enhance law enforcement’s knowledge about the behaviour of persons who use the Internet to exploit male children. As possible differences in suspected paedophile behaviour according to a child’s gender were identified, it presents a convincing case for further research into this area. It also suggests that employing the same policing methods to identify diverse offending behaviour may not be successful.
The Internet has provided anonymous and low-risk avenues for persons to gain access to children for sexual purposes, such as explicit conversations (chat), the taking and viewing of explicit images and as a vehicle for offenders to arrange to meet children for the purposes of engaging in sexual contact (O’Connor 2005; Stanley 2001). In response, police agencies worldwide have established taskforces commissioned to police the online environment. In Australia, police undertake sting operations in an attempt to identify child sex offenders by assuming the online identity or persona of a child. These operations investigate contraventions of ss 474.26 and 474.27 of the Criminal Code Act 1995 (Cth) which prohibit the use of a ‘carriage service’—that is, the Internet—to solicit children for sexual purposes. In Queensland, similar prohibitions are enacted under s 218A of the Criminal Code 1988 which prohibits the use of electronic communications with the intent to procure a child to engage in a sexual act. It also prohibits the exposure of a child to ‘indecent matter’ without legitimate reason. Indecent matter includes indecent sexual conversation or explicit sexual material such as pornography.
A comprehensive literature review revealed limited research describing the nature of interactions of those who engage male children on the Internet for sexual purposes. Instead, the focus has been on suspected offending that targets female children. In addition, police stings, for the most part, involve a police officer posing online as a female child as this has consistently achieved success in terms of identifying and prosecuting offenders. Consequently, a large portion of knowledge concerning online offending has been derived from behaviour exhibited by offenders attracted to female children and thus may fail to address the risk posed to male children.
When studying online police stings, Krone (2005) identified a peculiarity in offender interactions when police assumed the online persona of a young boy. In contrast to literature describing offender/child online engagements, these interactions did not contain the highly sexualised conversations experienced when police posed as a girl. The aim of this research was to further investigate the police experience when posing online as a male child and compare it with portrayals of Internet-based strategies used to exploit young girls.
For the purposes of this study, paedophilia is defined as a preferential sexual attraction to children that may or may not be acted upon (APA 1994). They are a subset of child sex offenders who have committed an offence against a child (Cunneen cited in James 2000).
The relevancy of gender
Paedophilia involving a preference for males is identified at approximately twice the rate of a preference for females (APA 1994). However, girls are more likely to be victims of sexual assault and online sexual solicitation (Choo 2009; Finkelhor, Mitchell & Wolak 2000; Internet Crime Forum 2001; Wolak, Mitchell & Finkelhor 2006). Geiser (1979) believes that the dynamics of male sexual abuse differ significantly from those involving females. The preferential and fixated offender typologies of Lanning (2001) and Groth, Hobson and Gary (1982) respectively show a primary sexual preference for boys, with Abel et al. (1987) and Abel (1989) suggesting that male-focused paedophiles are more likely to choose extra-familial victims and commit contact offences. As familial offences were more likely perpetrated against females and include a high instance of multiple offences against a single victim, they are more likely to be reported (Abel & Osborne 1992) and thus captured in official statistics. Incorporating these findings, it is possible that some offenders who prefer male victims are unlikely to be satisfied with online contact.
According to the Australian Broadcasting Authority and NetAlert (2005), boys and young children aged 8–11 years are more likely to access the Internet for games and music, while girls and older children aged 12–13 years predominantly use it for email and instant messaging. These findings are significant because the places children spend time online are likely to be targeted by paedophiles. For example, offenders seeking boys may target media such as gaming and sports rather than chatrooms. A preferential offender wanting a relationship with a young male may elude police stings that instead snare those driven by a desire for a sexual encounter online or short-term gratification—a dominant characteristic of online interactions with female children (Krone 2005). It may also be that those seeking boys are more inclined to seek out victims in real life rather than being content with online contact. According to Sullivan and Beech (2004), the majority of professional perpetrators (who sexually abused children with whom they worked) exclusively abused male children and had a predilection to use the Internet to access pornography or to engage children for sex.
In order to provide a qualitative comparative study to Krone’s work, semi-structured interviews with police experienced in investigating online child exploitation matters in Australia, together with documentary case studies of online police stings, were carried out. Ten sworn respondents provided information by interview (n=6) or by self-completed questionnaire (n=4) containing the same questions asked during the interviews. Those who completed the questionnaire were not available for interview due to operational reasons. Fifteen investigative case studies consisting of online chat logs were provided to the researcher, depicting the online engagements between suspects and police officers posing as male children between 13 and 14 years of age. Due to the low number of cases held by participating agencies that used the male persona, it was necessary to obtain cases dating back to 2003 and to include matters that did not result in an arrest. Of the 15 case studies, only three resulted in an arrest. All except one case involved a police officer posing as a male child (a sting), with the exception involving a referral to police after an adolescent’s family discovered their teenage boy may have been solicited online. Police then assumed the online identity of the victim. The low number of cases involving the boy persona and the arrest rate are in contrast to the successful pursuit of female-focused offenders as identified by Krone (2005). The low identification/arrest rate did not allow for offender demographic data to be interrogated.
Investigative accounts of undercover police who were experienced in using the male child persona revealed several types of problematic content during online exchanges between the ‘boy’ and the would-be offender:
- cautious introduction of sexual topics;
- enquiring about the ‘boy’s’ sexual experience and physical body; and
- relaying a sexual fantasy with the ‘boy’ playing a central role.
Police noted differences in the nature of interactions with the ‘boy’ compared with the ‘girl’ persona they used. Most notably, interactions with the ‘boy’ were less sexually explicit and less aggressive or forceful. Offending behaviour towards boys was noted by some police as being geared towards establishing mutual respect and trust, while with female children it was often structured toward domination. Police suggested that aggressive tactics such as blackmail and threatening behaviour may be used more against female victims. In contrast, the ‘boy’ invoked protracted conversations where the focus was on establishing a friendship rather than the short-term sexual gratification that typically characterised interactions with a female child. In terms of contact offences, it could be surmised that boys may have greater capacity to defend themselves than females and thus greater effort to groom a boy is required. This may translate to the online environment, particularly with those who ultimately want to commit contact offences against a male child.
According to police, it was not uncommon for suspects to transfer pornography during online engagements as a way of desensitising the ‘boy’. Some offenders—who appeared to be driven by short-term sexual gratification—sent images to the ‘boy’ during an initial chat. Conversely, others waited until a few conversations had transpired. Adult gay pornography and child pornography depicting boys were often used to normalise the behaviour depicted in the images. Police noted that the majority of images of a male-focused suspect would depict male children, particularly images exhibiting their ‘preferred’ child’s physical characteristics (ie blonde hair, blue eyes).
The nature of the case studies mirrored police descriptive accounts. In all but one of the case studies, five types of overt sexual activities were undertaken by suspects, with many suspects undertaking more than one activity during online engagements:
- webcam exposure (the suspect exposed himself naked to the child);
- indecent image transfer (child or adult pornography sent to the child);
- direct solicitation (requests to meet the child offline for sex or to make contact offline);
- describing a sex act or sexual activity with the child; and
- other sexualised conversation (ie conversations about sex that may be construed as indecent when a child is exposed to them and do not otherwise fit into the categories above).
Thematically, the case studies showed varied interactions with the suspects, which were categorised into four types of exchanges:
- short-term sexual gratification (n=8)—characterised by a desire for short-term sexual encounters and participation in overt sexual activities;
- longer-term procurement (n=2)—closely resembled grooming by displaying effort to befriend or establish a relationship with the ‘boy’;
- cautious, more restrained exchanges (n=3)—sex was introduced cautiously (or not at all) and no other overt sexual activity was undertaken; and
- educating the ‘boy’ in gay matters (n=2)—no sexual interest in the ‘boy’ was exhibited, rather there was a willingness to educate him in sexual matters.
How sex is introduced
Three patterns were identified in 10 cases in terms of how a sexual topic was introduced into conversations. All occurred within the initial stages of conversation:
- verification—general chat—child’s sexuality—sexual topic;
- verification—child’s sexuality—sexual topic; and
- verification—sexual topic.
Verification was usually carried out by asking ASL? (age, sex, location?). In the first and second patterns, the ‘boy’s’ sexuality was established prior to the introduction of a sexual topic and may have been a way of gauging the ‘boy’s’ receptiveness to speaking openly with the suspect. In most instances where the ‘boy’s’ sexuality was established within the initial stages, it was followed by a sexual topic such as enquiring about the ‘boy’s’ sexual experience and/or his physical body. Many suspects were also willing to entertain non-sexual topics after sex had been introduced. This may be a strategy to maintain rapport with the ‘boy’ in order to establish a friendship or relationship.
Short-term sexual gratification
This group was characterised by three main types of interaction:
- direct sexual exchange (in terms of wanting phone sex or cybersex);
- discussing or giving instruction regarding sexual activity; and
- wanting to meet offline for sex.
All interactions in this group used the three patterns outlined above showing the progression to a sexual topic. The majority of the short-term sexual gratification interactions displayed a general pattern, in that overt sexual activities occurred after the ‘boy’s’ sexuality had been established. In most instances where the suspect requested to meet, a sex act was then described to indicate what activity the suspect wished to engage in at the meeting. Describing a sex act was the last overt sexual activity undertaken in many cases within this group.
These cases showed quick progression to explicit conversation and overt sexual activities and resembled one type of interaction described in Krone’s (2005) study. Some of these suspects appeared to have been satisfied with online contact only, however, if contact was maintained with the ‘boy’, it may have progressed to a physical meeting. Due to the small numbers and suspects’ willingness to undertake a number of overt sexual activities, it was difficult to categorise this group’s behaviour, except to define them by their strong desire for immediate or short-term sexual gratification.
Longer-term procurement interactions also showed a pattern of progressing from the verification stage and establishing the ‘boy’s’ sexuality to discussing sexual experiences and transferring child pornography to the ‘boy’. The suggestion of meeting in person occurred as a last step, but unlike the requests that occurred within the short-term gratification cases, they were subtle suggestions. These suspects wanted a friendship and relationship with the ‘boy’ and showed characteristics of Lanning’s (2001) preferential offender. However, their behaviour did not reflect the distinct stages of O’Connell’s (2003) ‘cybersexploitation’ wherein general stages of grooming were described:
- friendship forming;
- relationship forming;
- risk assessment;
- exclusivity; and
O’Connell’s (2003) study is based on her participant observational method of posing online as a young girl. According to O’Connell (2003), the risk assessment stage involves establishing trust and assessing the likelihood of the suspect being detected; it occurs after a relationship is formed. This stage was not evident in the cases assessed in this study. For example, the suspects transferred child pornography seemingly without regard for getting caught or discovered by potential detectors of illegal activity, such as parents. In addition, the fourth stage of exclusivity (where the suspect asks whether the child trusts them and ensures the child does not disclose their contact) was not identified in the case studies.
Within the longer-term procurement cases, discussing sexuality and sexual experience established mutual understanding between the suspect and the ‘boy’ and formed the basis of the friendship/relationship, rather than the relationship forming first. In addition, the use of non-sexual conversation interspersed with sexual topics blurred the distinct progression from non-sexual to sexual interactions that is evident in O’Connell’s (2003) typology. While most grooming research suggests a gradual, slow introduction of sex, Sandfort (1984) found in his study of 25 boys aged 10 to 16 years involved in contact paedophile relationships, that sex was introduced quickly into the relationship. In a third of cases, sexual activity occurred during the first encounter. In another third, sex occurred after several meetings and for the remainder of cases, it occurred after a month or longer. This may suggest that the gradual introduction of sex may be more representative of the offending pathways for those who target female victims.
Cautious, more restrained exchanges
The cautious and more restrained exchanges showed early signs of longer-term procurement. These offenders were described by police as those who introduced sex in a way that is difficult to prosecute. No patterns were identified in these exchanges apart from two suspects enquiring about the child’s sexuality as part of the verification process.
The suspects in this group professed their sexual preference for boys. They did not make sexual advances and did not ask for personal contact. These cases, if allowed to continue for a longer term, may have culminated in a relationship or introduced sex more overtly, depending on the motivation of the suspects. The use of non-sexual conversation in this group (similar to the longer-term procurement) highlighted the suspects’ ‘friendly’ persona which helped maintain rapport, particularly when used during sexual topics.
Educating the boy in homosexual matters
These suspects did not show a sexual interest in the ‘boy’ but offered to answer questions about homosexuality. The suspects transferred pornography at some stage during the engagements, followed by limited conversation about the ‘boy’s’ sexuality. It appeared that the images were sent to educate the ‘boy’ rather than to entice him into sexual conversation or activity. This stands in contrast to the three groups discussed earlier.
The study did not support Krone’s (2005) findings that male children do not invoke sexualised responses from online suspects. In this small study, many of the suspects who engaged the ‘boy’ participated in overt sexual activities and their interactions could be characterised as direct sexual exchanges, discussing sexual activity and/or wanting to meet offline for sex. This group closely resembled the exchanges outlined by Krone (2005) in his description of offender interactions with young females.
Other types of offenders were identified, some of whom bore the characteristics of a preferential offender willing to forego immediate sexual gratification in an attempt to establish a friendship or relationship with the ‘boy. However, in the cases studied, befriending the ‘boy’ did not mean sex was gradually introduced into the relationship as noted in the grooming literature (Choo 2009; O’Connell 2003). Instead, sexual topics were used early in the relationship as a mechanism to ingratiate suspects with the ‘boy’, supporting Sandfort’s (1984) study into paedophilic relationships of boys and older males.
Many cases showed attempts to exploit the ‘boy’s’ vulnerabilities, which resonated with police experience, in that some offenders target boys who are still coming to terms with, or who are confused about, their sexuality. According to Cunneen (cited in James 2000), the sexual curiosity of boys, particularly at puberty, assists the offender to commit contact offences, particularly as sexual encounters can be physically enjoyable for the boy. The grooming of children capitalises on the child’s burgeoning sexuality and if the activity is pleasurable, it can reduce the risk of the child reporting the activity (Sanderson 2004). For boys, the fear of being labelled homosexual can also ensure the abuse is not reported (Buchanan 1997; Ferraro & Casey 2005). This may be why O’Connell’s (2003) exclusivity stage was not identified in this research. Exploiting a boy’s curiosity or confusion about his sexuality may be enough to reduce any likelihood of disclosure, even if contact with the offender did not progress to physical contact.
A key theme through the majority of interactions was the focus on the ‘boy’—whether it was enquiring about his sexuality, physical body, sexual experience or involving him as the central role in a sexual fantasy. Offenders seeking information about a boy’s sexuality and physical characteristics is not identified in the wider literature but may be an indication of the male-preferential offender described by Lanning (2001). The focus on the ‘boy’ was also highlighted in the describing of sex acts, all of which involved the ‘boy’. Further, in all instances involving child pornography, the suspects used images depicting boys. According to Seto, Cantor and Blanchard (2006), the use of child pornography can be a strong indicator of paedophilia.
The longer-term procurement and cautious exchanges may represent cases where suspects have used the Internet to befriend a boy with the purpose of gaining physical access to the child, which is congruent with the findings of Abel et al. (1987) and Abel (1989). Sullivan and Beech (2004) found that for some male-preferential offenders, the Internet serves as a means of increasing their opportunities to meet suitable boys for contact offences. In addition, Quayle and Taylor (2001) quoted a preferential offender as saying that his preferred contacts were those who would befriend him and want a relationship in addition to sex. It is possible that suspects, whose conduct displayed a willingness to forego short-term gratification, may constitute more serious offenders. Yet they may go largely undetected as their cautious behaviour is not conducive to the time-sensitive nature of police stings.
There were other notable distinctions identified when compared with research based on female victim offending. First, the cautious behaviour of those who engaged the ‘boy’ helped evade police identification. Second, the offenders were less aggressive and forceful than what is reported in literature. For example, neither police accounts nor chat logs contained engagements whereby forceful or aggressive language was used, as characterised in O’Connell’s (2003) ‘cyber-rapist’. The cyber-rapist uses swearing, name calling and emotional blackmail to coerce the child into participating in cybersex. It is possible that suspects use the sexual curiosity of the ‘boy’ or confusion about his sexuality as a subtle yet manipulative tool replacing the need for overt or aggressive tactics to keep the ‘boy’ silenced or to expose him to sexualised conversation and activity. It is also possible that with regards to male victims, the Internet provides a vehicle more conducive to non-violent behaviour.
The study also revealed that suspects who engaged the ‘boy’ did not masturbate by webcam and only one indecently exposed himself using a webcam. In contrast, these activities are widely reported in literature including Krone’s (2005) study. Instead, suspects were more likely to portray themselves as a friend and more willing to engage in non-sexual conversation. Further comparison with Krone’s (2005) findings also suggests that female children may be targeted more frequently in terms of requests to meet offline (concurring with reported online rates of sexual solicitations) and the transfer of indecent images.
Implications for law enforcement
This study shows that online police stings can engage persons who are satisfied with a random or short-term online sexual encounter with a child. Yet the identification of suspects pursuing male children stands in contrast to the very successful pursuit of female-focused offenders. It is possible that employing the same methodology to capture the diverse offending behaviours of paedophiles may not be an effective approach.
There is evidence to suggest that paedophiles, reportedly with a higher rate of sexual attraction to boys, are willing to procure a child over time with the ultimate goal of committing contact offences. A tendency to focus on securing a quick arrest, rather than engaging in protracted interactions may be excluding some offenders from arrest, although they may represent an equal or greater danger to children. Police may be more successful in conducting protracted investigations in cases involving a ‘boy’ and concentrating on developing a relationship with those who show a resemblance to the cautious and more restrained exchange offender group, as they may later progress to more overt grooming tactics.
An increase in arrest statistics derived from police stings may not necessarily translate to a safer online environment. By their nature, stings are short-term focused and largely attract individuals who present an immediate risk to children. These offenders may include those who, despite their feelings, do not act out and may represent a lower-risk offender compared with those who want to groom a child for contact offences. A more comprehensive examination of the stages of interactions with a male child is required to determine at what point an offender type can be established; that is, one who is short-term focused or a preferential paedophile. As this study has shown, the use of themed child pornography or enquiring about the ‘boy’s’ physical body or sexuality may provide some indication of a desire for sexual contact with a male child.
Due to the limited number of police who held online experience posing online as a male child, the dataset used in this study is small and the findings may not be generalisable. Further, the study describes the interactions between police posing as male children between 13–14 years of age or those who have engaged the male child persona in specific chat rooms in online environments. As such, these findings may not have application to younger male children or older adolescents, or address the risk to male children outside chat rooms and forums on the Internet. There is also no attempt made to address the risk to female children, nor the risk to young children who frequent chat rooms putatively for female use. Finally, the police experience may vary in jurisdictions with different legislation and investigative practices, particularly in countries that do not have anti-grooming legislation.
The assistance of the Queensland Police Service, the Queensland Crime and Misconduct Commission and the Australian Federal Police in facilitating this research is greatly appreciated.
All URLs correct at August 2010
- Abel GG 1989. Behavioral treatment of child molesters, in Stunkard AJ, Baum A (eds), Perspectives on behavioural medicine. New York: Lawrence Erlbaum
- Abel et al 1987. Self-reported sex crimes of nonincarcerated paraphiliacs. Journal of Interpersonal Violence. 2(1):3–25
- Abel GG & Osborn MA 1992. The paraphilias: The extent and nature of sexually deviant and criminal behaviour. Clinical Forensic Psychiatry 15(3): 675–687
- American Psychiatric Association (APA) 1994. Diagnostic and statistical manual of mental disorders DSM-IV, 4th ed. Washington: American Psychiatric Association
- Australian Broadcasting Authority & NetAlert Limited 2005. Kidsonline@home: Internet use in Australian homes. Sydney: ABA and NetAlert
- Buchanan D 1997. Homosexual ‘pedophilia’ and the NSW Police royal commission.
- Carr A 2004. Internet traders of child pornography and other censorship offenders in New Zealand. Wellington: Department of Internal Affairs Ti Tari Taiwhenua
- Choo K-KR 2009. Online child grooming: A literature review on the misuse of social networking sites for grooming children for sexual offences. Research and public policy series no. 103. Canberra: Australian Institute of Criminology.
- Ferraro MM & Casey E 2005. Investigating child exploitation and pornography: The internet, the law and forensic science. London: Elsevier Academic Press
- Finkelhor D, Mitchell KJ & Wolak J 2000. Online victimization: A report on the nation’s youth. Durham: University of New Hampshire
- Geiser RL 1979. Hidden victims—the sexual abuse of children. Boston: Beacon Press
- Groth AN, Hobson WF & Gary TS 1982. The child molester: Clinical observations, in Conte JR & Shore DA (eds), Social work and child sexual abuse. New York: Hawthorne Press
- Internet Crime Forum 2001. Chat wise, street wise—children and internet chat services. http://www.internetcrimeforum.org.uk/chatwise_streetwise.html
- James M (ed), Paedophilia: Policy and prevention. Research and public policy series no. 12. Canberra: Australian Institute of Criminology.
- Krone T 2005. Queensland police stings in online chat rooms. Trends & Issues in Crime and Criminal Justice no. 301. Canberra: Australian Institute of Criminology.
- Lanning KV 2001. Child molesters: A behavioral analysis, 4th ed. Washington: National Center for Missing and Exploited Children
- O’Connell R 2003. A typology of cybersexploitation and on-line grooming practices. http://image.guardian.co.uk/sys-files/Society/documents/2003/07/24/Netpaedoreport.pdf
- O’Connor C 2005. Child pornography and the internet—A statistical review. Australian Police Journal Dec 2005: 190–199
- Quayle E & Taylor M 2001. Child seduction and self-presentation on the Internet. Cyberpsychology and Behaviour 4(5): 597–607
- Sanderson C 2004. The seduction of children: Empowering parents and teachers to protect children from child sexual abuse. London: Jessica Kingsley Publishers
- Sandfort TGM 1984. Sex in pedophiliac relationships: An empirical investigation among a nonrepresentative group of boys. The Journal of Sex Research 20(2): 123–142
- Seto MC, Cantor JM & Blanchard R 2006. Child pornography offences are a valid diagnostic indicator of pedophilia. Journal of Abnormal Psychology 115(3): 610–615
- Stanley J 2001. Child abuse and the internet. Child abuse prevention issues paper no. 15. http://www.aifs.gov.au/nch/pubs/issues/issues15/issues15.pdf
- Sullivan J & Beech A 2004. A comparative study of demographic data relating to intra and extra familial child sexual abusers and professional perpetrators. Journal of Sexual Aggression 10(1): 39–50
- Taylor M, Quayle E & Holland G 2001. Child pornography, the internet and offending. ISUMA. Summer Ete 2001
- Wolak J, Mitchell KJ & Finkelhor D 2006. Online victimization of youth: Five years later. National Centre for Missing and Exploited Children. Durham: University of New Hampshire
About the Author
Angelique Grosskopf is a Criminal Intelligence Analyst for the Australian Federal Police