Volume 2 N° 2 • Summer 2001 • ISSN 1492-0611

Child Pornography, the Internet and offending

Max Taylor, Ethel Quayle and Gemma Holland [ * ]


Deux perspectives complémentaires—l’une juridique et l’autre psychologique—sur la pornographie infantile sont ici présentées accompagnées de la description d’une typologie émergente permettant de comprendre la nature des images véhiculées sur Internet. Des données tirées du projet européen COPINE (Combating Paedophile Information Networks in Europe) sont utilisées pour illustrer la nature du matériel mis à la disposition des individus démontrant un intérêt sexuel à l’endroit des enfants. Ces données révèlent également les moyens d’accéder à ce matériel, ce que font les délinquants avec Internet et les changements de comportements qui s’ensuivent chez ces mêmes individus. L’article se termine par un regard sur une série de préoccupations émises à l’endroit de la pornographie infantile, d’Internet et des comportements délinquants. (Traduction: www.isuma.net )


Two complementary perspectives on child pornography—legal and psychological—are presented and an emergent typology for understanding the nature of such pictures on the Internet is outlined. Data from the Combating Paedophile Information Networks in Europe (COPINE) project is used to illustrate the nature of the material available to people with a sexual interest in children, where it can be found and how offenders use and are changed by the Internet. It concludes with a consideration of issues that are of concern in relation to child pornography, the Internet and offending behaviour.

THE COPINE PROJECT AT the Department of Applied Psychology, University College Cork, has been in existence for the last four years and forms a part of the activities of the Child Studies Unit, the focus of which relates to the needs of vulnerable children. The COPINE project seeks to address children’s vulnerability in relation to the new technologies, and in particular issues related to child pornography and the Internet. An important feature of the project, which has been funded by grants both from the EU and private sources, has been its links with the law enforcement community in Ireland, the United Kingdom, Belgium, the Netherlands and the United States. Currently the Project is involved in three research areas: the maintenance of a reference data base of child pornography; an assessment of dangerousness in paedophiles through their collection of child pornography; the nature and incidence of child sex tourism and child trafficking in Europe. A new project (jointly with Radda Barnen and Childnet) concerned with victim identification will begin shortly. The data discussed in this paper arises out of the Project research and will be used to clarify issues relating to the nature of child pornography, how it is accessed on the Internet and used by offenders and the problems that potentially arise from this.

Most of us would agree that child pornography constitutes sexualised pictures involving children. However what precisely is meant by sexualised can vary depending on whether a legal or the more subjective perspective of the adult with a sexual interest in children is taken. Legal definitions tend to emphasize obscene or sexual content as an essential quality of the images, but of course such definitions may vary depending on the legislature within a given country. This is a particular problem in relation to the Internet where cultural, moral and legal variations make it difficult to define “pornographic” in such a global society.[ 1 ]

The International Criminal Police Organization (Interpol) standing working group on offences against minors defines Child pornography as: “...the consequence of the exploitation or sexual abuse perpetrated against a child. It can be defined as any means of depicting or promoting sexual abuse of a child, including print and/or audio, centred on sex acts or the genital organs of children.”

A similar emphasis on the sexual nature of the material can be found in the U.K.’s Criminal Justice Protection of Children Act of 1978 which was amended in 1994 to state that, “it is an offence for a person (a) to take, or permit to be taken or to make, any indecent photographs or pseudo-photographs of a child; (b) to distribute or show such indecent photographs or pseudo- photographs.”

However, while such definitions are clearly important from a legislative perspective, they are of limited value in helping us understand why such images are produced or collected, or the function that they might have in offending behaviour. Over emphasis on highly sexualised or obscene images deflects from the fact that any image can be sexualised and fantasized over, and what makes that image of the child important to the adult is the psychological role it plays in arousal and masturbation.[ 2 ] Legal definitions enable us to draw boundaries, but do not help in delineating the problem. Indeed, it may be argued that such definitions give rise to the notion of “good” and “bad” offenders, depending on the degree to which the viewer finds the material objectionable.

Collections of child pornography do not arise by accident, but by the deliberate choices an individual makes to acquire material. Nor are such pictures a homogeneous category. By emphasizing a psychological approach to such images, we can identify a range of discernibly different pictures that are collected by adults with a sexual interest in children, only some of which may be illegal. It is possible to construct a continuum of increased deliberate sexual victimization, ranging from everyday and perhaps accidental pictures involving either no overt erotic content, or minimal content (such as the depiction of a child in their underwear) to pictures showing actual rape and penetration of a child, or other gross acts of obscenity. Such a continuum may be helpful in characterizing the nature of a given collection, and may also elucidate factors that may enable and sustain offender behaviour. Central to this is an acknowledgement that child pornography is not a victimless crime. Regardless of the content of the picture, each time that an image of a child is accessed for a sexual purpose, it victimizes the individual concerned.

The value of generating a typology of pictures arose from the large reference database amassed by the COPINE Project. This currently exceeds 80,000 individual still pictures, as well as a large number of video sequences. The COPINE collection contains examples of most of the material publicly available on the Internet, and focuses particularly on newer material. A descriptive analysis identified 10 levels of severity, based on increasing sexual victimization, and quite deliberately included pictures that do not fall within any legal definition of child pornography. These 10 levels of severity can be briefly described as follows:

It is important to emphasize in relation to collections of pictures, that even Level 1 images can be sexualised and fantasized over, and may be used to both promote and sustain sexual fantasy.[ 3 ] It is also the case that boundaries between pictures can be blurred, but such a typology encompasses the wide array of material attractive to the adult with a sexual interest in children, and places the emphasis back on the child as victim, rather than the end-product of the obscene photograph. This is of particular importance in the context of Level 3 pictures (pictures which are secretly or surreptitiously taken). Lack of knowledge of victimization by the subject does not diminish its gravity.

Collections of pictures are rarely a random aggregation of individual images. Most pictures occur as part of a series and it is usual for such series to have a narrative or thematic link. The narrative may be an aid to fantasy and filling gaps in the series may be highly reinforcing to the collector. Locating a picture, or a series of pictures, at some level on the continuum needs to be considered in the context of other issues relating to a collection. These include the size of the collection and the obsessional qualities relating to its organization, and storage, the principle themes illustrated, the presence of new or private material, and the age of the children depicted in the pictures. Recent evidence[ 4 ] suggests that the age of children in new pornography is reducing with the implication that such victims may be less able to disclose the abuse than would older children.

The COPINE database is drawn from pictures from newsgroup postings. Its principal practical value (as distinct from its research role) has been to facilitate the identification of new children involved in child pornography, as distinct from existing material. The database takes two forms: first, an archive of old pictures which are known to be over 15 years old, and second, a searchable archive of new (less than 10 years old) and recent (10 to 15 years old) pictures, which are collected daily from in excess of 60 newsgroups which are known to carry child pornography. The archive focuses on levels 6-10 in the above typology and is a large, representative sample of new photographs in the public domain. Through the database, the Project has extensive records of the nicknames and Network News Transport Protocols (NNTP) posting host addresses under which newsgroup postings have been made.

In total, the database includes approximately 80,000 still pictures and 400+ video clips. Of the 80,000 images, more than half are of girls. Of those pictures of girls categorised as level 7 and above, about 7 percent are new. Approximately 26 percent of similar level boy photographs are new (categorised as level 7 or above). In these new pictures, forty one percent of the girls and 56 percent of the boys are between the ages of 9-12, the rest being younger. The vast majority of both sexes of children depicted in the new pictures of level 7 and above are white Caucasian, with Asiatic children more likely to appear in posed images (levels 5 and 6). There is a marked absence of black children in any of the age groups, and as yet there is little evidence as to why this should be the case. Anecdotal information from the COPINE interviews with offenders suggested that many show a preference for thin, fair children, where genitalia are clearly visible and where there are no secondary sexual characteristics.

Currently, photographs are appearing in the newsgroups monitored by the project at a rate of about two new children per month. This is highly variable, but there is evidence that the age of the children (particularly females) is getting younger, that there is an increase in “domestic” production (where the settings are family rooms), and that there has been a growth in the number of photographs of East European children. Over the last year, there has also been an increase in the number of pictures whose origins appear to be commercial web sites, based in South America or Eastern Europe; these pictures tend to be in levels 5 and 6. These, and the emergence of more explicit child pornography from Eastern Europe access to which involves payment, seem to represent a growth in commercial exploitation of the market for child pornography.

It can be estimated that between 300-350 of the children included in the new/recent material have been subjected to serious sexual assault (being present in pictures categorized as level 7 and above). Of the 1,600-1,800 children who have been photographed while posing naked (levels 5 and 6), it is reasonable to assume that a number of these will have been sexually assaulted, either outside of being photographed or without the images being distributed. It is also reasonable to assume that the figures given here are an underestimate of the numbers of pictures in circulation as the amount of material in private circulation is currently unknown.

It is important to stress that technological changes are evident in the emergence of material distributed through the Internet, but Video still seems to remain the principal primary production medium for child pornography. This is reflected in new recent Internet material in the amount of high quality video captures. The age, sex and ethnicity of the children in the COPINE database videos clips are very similar to the distribution found in the still pictures. Fournier de Saint Maur[ 5 ] has suggested that the Internet is fast becoming a significant factor in the sexual abuse of children and the principal means of exchange of child pornography.

The new technologies also mean that “users” of the Internet can quickly become “producers.”[ 6 ] Such photographs do not need to be commercially processed which allows for a much greater level of security in their production. Recent cases in the United Kingdom give some indication of the extent to which this occurs. Images can also be altered by the paedophile to suit personal preferences, or can even be made up of components from separate photographs. Such images are often called “pseudo photographs.”[ 7 ] Indeed, technology is now allowing the creation of lifelike child pornography without the use of children at all.[ 8 ] The Internet also allows for on-line abuse of children. For example, in 1996, a group of paedophiles that called itself the Orchid Club was arrested in the United States. Using a digital camera, one of the group members transmitted real-time images of a child being sexually assaulted and responded to requests from the club’s members in directing the abuse. Members of the Orchid club lived in the United States, Europe and Australia.

A major source for both text and images are the Usenet Newsgroups. There are some 30,000 such groups on the Internet covering a wide range of subjects (the vast majority of which are innocent and appropriate) but paedophiles use some of these newsgroups to communicate with each other, to ask advice about computer-related problems or the availability of new software. They also use them to request pictures from other paedophiles and to post them. The pictures in themselves act as a form of currency, legitimizing activity and creating social cohesion. Data from the COPINE Project suggests that over 1000 illegal photographs per week may be posted into what is effectively a publicly available service.

Bulletin Board Systems (BBS) are similar to Newsgroups but with more real time involvement. Web based BBSs have played an increasingly important role in the on-line paedophile community in recent years. Durkin and Bryant talk of them as high tech party lines, by which users can send and receive text, engage in conversations, and both upload and download files.[ 9 ] A more familiar means of communication is through e-mail and the mass distribution of e-mail files through List Servers (Listservs).

A further important source of both information and material is Internet Relay Chat (IRC). This is a vast multi-user discussion forum, which allows users to communicate through text but in real time. Paedophiles in IRC can exchange files directly with one another using a Direct Client-to-Client (DCC) feature and can distribute images through Fserves. This allows the user to access a certain part of another user’s hard drive to upload and download files. Paedophiles can therefore be invited to ‘visit’ somebody else’s collection and take what material they want.

The World Wide Web (www) is another source of child pornography. It is now relatively easy to create Web pages and features built into Web browsers enable users to capture content created by other people with a minimum of effort.[ 10 ] The proliferation of free web servers has meant that individuals can upload a web site containing child pornography anonymously, and with ease.

It is unknown how many people look at child pornography on the Internet, nor is it always clear what motivates them to do so. Out of the many thousands of people who simply look, there are a much smaller number who are involved in the distribution or production of material, or both. As part of the COPINE Project, we have been conducting a series of semi-structured interviews with people who have been identified, largely through the judicial system, as people expressing a sexual interest in children. This ongoing series of interviews includes downloaders (no assault); downloaders and assault (no production); downloaders and distribution (no assault), downloaders and producers (assaults); downloaders, producers and distributors (assaults); sexual assaults only. The ensuing transcripts have been analyzed within a qualitative framework. A thematic analysis was used to establish a broad framework for understanding the data,[ 11 ] followed by a more discursive analysis to look at how individuals negotiate their accounts. This “applied” discourse analysis[ 12 ] was also informed by Holloway and Jefferson’s guidelines for both interviewing and analyzing data with defended subjects.[ 13 ]

Emergent themes from the interviews[ 14 ] illustrated that people who use the Internet to download pornography move through a series of stages in their offending behaviour which is directly influenced by their level of engagement with the Internet. Setting events for downloading included histories of early sexualised behaviour, inadequate adult socialization, dissatisfaction with current persona and an acquisition of computer and Internet skills. Initial contact with the Internet was often accompanied by accessing adult pornographic sites before a search for child pornography. Accessing such material facilitates engagement with a virtual community, further normalizing the collection of material, as well as promoting further engagement with the Internet and its corresponding technology. For all people interviewed, this was followed by a steady increase in on-line behaviour and a reduction in other outside social engagement. Increasingly large quantities of material are rapidly collected, and what emerges are different forms of collecting behaviour, with time spent sorting and cataloguing images. Such categories might be quite rudimentary, for example organized around age or sex of the children photographed, or more sophisticated with a focus on content of activities. For those people who download and go on to engage in social contacts on the Internet, the process of sustaining that engagement requires credibility. Such credibility is often achieved through the exchange or trade of material such as pictures, text or fantasy stories.

The process of collecting appears to be an important psychological process in itself. The rapid acquisition of images largely goes hand-in-hand with the acquisition of technical skills. Collecting also leads to an increase in fantasy and sexual activity, particularly masturbation in relation to images or through engaging in mutual fantasies with others while on-line. With increasing mastery of the Internet comes also a sense of power and control, and this is also evidenced in some of the preferred fantasy material about “teaching” children about love and sex. Where downloaders also communicate with each other, this becomes a powerful justification for their activity, and there is constant reference to the importance of on-line relationships over material. For downloaders who trade images, the notion of photographs as currency is important. They are currency in terms of trading for new images, but they are also currency in maintaining existing on-line relationships and giving credibility. Text never exclusively replaces images, but for many downloaders it seems to play an important role in the process of engagement, and may enable the formation of close relationships, some of which move from being “virtual” to “real.”

This process can clearly be seen in the case of P.G., the creator of w0nderland (a world-wide private network of individuals who traded texts and images with each other on IRC), who described himself as a person with “mental health problems” that pre-dated accessing material on the Internet. He positioned the responsibility for his “engagement” with the Internet as lying with a friend: “Well he always joked about taking me over to the dark side” and went on to describe the Internet as “a doorway... to the dark side.” Through the Internet he was able to stake a claim to being good at something, feeling competent in his ability to use computers and his expertise in the area of security. P.G. used a model of addiction to explain his behaviour on the Internet, “A junkie par extraordinaire,” but made constant reference to setting limits in terms of the pictures he both accessed and distributed. With reference to finding images that completed a series he said, “It’s kinda like an art collector who finds a lost Picasso....” As the amount of time spent on-line increased, he talked about a sense of losing control and an inability to stop: “That’s like trying to stop Niagara Falls.” P.G. demonstrated a rapid escalation of collecting through increased technical expertise, showing an increased interest in the Internet as a community: “I’m like a virtuoso pianist ... but the instrument I play is a computer.” His interest was in his ability to control his virtual social world, and there were significant feelings of betrayal when at one point his own group left him and excluded him from their network. He desperately needed people to like him, and when on-line he would adopt a persona of a “likeable rogue.” “It was the most important thing to me... I had almost no friends in real life and what few friends I had... I kept at arms length.” Through the Internet he achieved status through the quality of his collection and his computer skills and experienced a real sense of loss when his computer and his collection were seized: “I lost my best friend when I lost my computer.”

The appeals of the Internet for those with a sexual interest in children are many. Social connections can vastly increase, and such virtual communities, while providing a safe haven, also allow for the control of social distance and intimacy.[ 15 ] The Internet allows the user to achieve a sense of mastery on-line, and status and prestige are achieved by those with good computer skills. Such skills, and the ability to master the virtual environment provide a sense of power. Users gain social confidence while on-line and report that they find it easier to make friends,[ 16 ] using the metaphor of the Internet as the Prozac of social communication. There is also the suggestion of altered states of consciousness through Internet use,[ 17 ] facilitating levels of engagement that would otherwise be seen as aversive or inappropriate.

The appeal of the Internet is problematic in relation to sexual interest in children. Paradoxically, Internet use among offenders increases socialization by providing a potentially enormous virtual community, while at the same time reducing the number of real social contacts. This functions in a way that allows for the normalization of sexual interest in children and enables engagement through the reduction in outside social contacts that might otherwise challenge the acceptability of the interest. It also functions in a way that allows the avoidance of personal responsibility by creating anonymity, facilitating identity choices and movement through identities.[ 18 ] With regard to sexual behaviour, the Internet lowers sexual inhibitions and the engagement in the sharing or trading of images acts as a form of social reinforcement. There is also evidence that it increases the level of sexual activity, both in relation to pictures and text: “It was that often... I did it till I couldn’t cause it hurt,” (D.H. w0nderland operator). The ability to manipulate the images to fit a sexual fantasy further makes an object of the child and enables control of the source of stimulation. Knowledge of the risks being taken in downloading material can also serve to heighten arousal.

Like all other offenders, people who download child pornography are able to legitimize their activity, and the role of the virtual Internet community in generating such a body of legitimizing stories appears to be important.[ 19 ] Offenders talk of a lack of any objective measure as to whether the child in the picture was actually being abused, or relate their own experience of abuse. A frequent comment refers to the smiling faces of the children in the pictures, as proof of their enjoyment, or personal fantasies of abuse that normalize the process of collecting. Whatever the level of participation in such a “community,” the process of obtaining photographs through the Internet validates and legitimizes such activity and provides a sense of support to those with a sexual interest in children. Such an engagement is flexible and dynamic and offenders can engage at multiple levels through different kinds of self-representation.

While those who access child pornography through the Internet evidence many similarities with other child sex offenders, they also demonstrate varying measures of “functional addiction,”[ 20 ] such as salience, mood modification, tolerance, withdrawal symptoms and conflict and relapse. This is an important consideration in the development of both appropriate assessment and treatment programs for offenders.

There are a number of concerns that arise out of this research. Despite the temptation of putting resources into the detection of the relatively easier offences of possession and trading, the emphasis must also be on child protection and the identification of children. Training for law enforcement and other professionals needs to address the evidential issues relating to the pictures, with emphasis on tracing the chain of postings and seeing the picture content as evidential forensic material. To understand the processes involved requires knowledge of the behaviour of downloaders and the language they use. These concerns are echoed in assumptions made about current treatment for such offenders and whether they are sufficiently similar to other sex offenders to be included in the same programs.

Granic and Lamey suggest that through Net experiences people have come to reinterpret society, relationships and the self.[ 21 ] The possibility that people are changed by and subsequently contribute to change through the Internet is highly relevant to the area of sexual offending. Through the Internet we see a potential change in the offender’s beliefs, values and cognitive styles. The fact that through the Internet users can in the main go anywhere and say anything without any official governing body restricting those actions means that for some people this will be their first experience of acting outside the confines of a conventional hierarchy. Granic and Lamey make the important observation that “conventional hierarchies are disrupted by a distributed, decentralised network in which power is spread among various people and groups and one voice does not dominate or pre-empt others.” Such experiences may empower some people such as sex offenders who have otherwise felt marginalized in conventional society. Those who have never been able to function at an optimal level in the real world may now feel that they have the chance to do so now that conventional structures are broken down: “So... I’m still wearing the mask in effect... I’m still having to hide who I am. I can’t be myself, and I miss being on-line for that more than anything else ‘cause there is nowhere I can be myself now,”(D.H. w0nderland operator).

It is also apparent that the ISP industry has a role to play in the regulation of child pornography on the Internet. Such control might take place at a National level, either through self-regulation or statutory control. This might be achieved through the mandatory reporting of child pornography, which will inevitably raise issues about censorship and the confidentiality of records. Another important issue here relates to the length of time ISPs retain records of user activity. It may also be the case that control could be exerted on an International level through the setting of common standards and values and by establishing protocols in relation to “rogue” countries who continue to facilitate the production and distribution of images portraying the sexual abuse of children.


*   Max Taylor, Ethel Quayle and Gemma Holland are with the Department of Applied Psychology, University College Cork, Ireland. A version of this paper was presented as part of the Research Seminar series, hosted byJustice Canada's Research and Statistics Division.

 1.  Y. Akdeniz, “The regulation of pornography and child pornography on the Internet,” The Journal of Information, Law and Technology (1997). Available on-line at http://elj.warwick.ac.uk/jult/internet/97_lakdz. Visited September 20, 2000.

 2.  M. Taylor, “The nature and dimensions of child pornography on the Internet.” Paper prepared for the International Conference “Combating Child Pornography on the Internet,” Vienna, Austria, September 29 to October 1, 1999. Available online at http://www.stop-childpornog.at/, visited October 2, 2000.

 3.  D. Howitt, “Pornography and the paedophile: is it criminogenic?” British Journal of Medical Psychiatry, Vol. 68, (1995), pp. 15–27.

 4.  Taylor, op cit.

 5.  A. Fournier de Saint Maur, Paper prepared for the International Conference “Combating Child Pornography on the Internet,” Vienna, Austria, September 29 to October 1, 1999. Available online at http://www.stop-childpornog.at/, visited October 2, 2000.

 6.  G. Thomas and S. Wyatt, “Shaping Cyberspace — interpreting and transforming the Internet, Research Policy, Vol. 28, (1999), pp. 681–698.

 7.  Taylor, op cit.

 8.  M. Healy, “Child pornography: An international perspective.” Prepared as a working document for the World Congress Against Commercial Sexual Exploitation of Children, 1997. Available online at http://www.usis.usemb.se/children/csec/215e.htm, visited 29 September, 2000.

 9.  K.F. Durkin and C.D. Bryant, “Log on to sex: some notes on the carnal computer and erotic cyberspace as an emerging research frontier,” Deviant Behaviour, Vol. 16, (1995), pp. 179–2000.

 10.  Thomas and Watt, op cit.

 11.  J.A. Smith, “Semi-structured interviewing and qualitative analysis,” in J.A.Smith, R. Harré and L. Van Langenhove, Rethinking Methods in Psychology (London: Sage, 1995).

 12.  For further discussion of this approach, see: C. Willig, Applied Discourse Analysis (Buckingham, Open University Press, 1999), and L.A. Wood and R.O. Kroger, Doing discourse analysis (Thousand Oaks: Sage, 2000).

 13.  W. Holloway and T. Jefferson, Doing qualitative research differently (London: Sage, 2000).

 14.  E. Quayle, C. Linehan, G. Holland and M. Taylor, “The Internet and Offending behaviour; a case study,” Journal of Sexual Aggression, in press.

 15.  K. Young and R. Rodgers, “The relationship between depression and Internet addiction,” Cyber-Psychology, Vol. 1, (1998), pp. 25–28.

 16.  J. Morahan-Martin and P. Schumacher, “Incidence and correlates of pathological Internet use among college students,” Computers in Human Behavior, Vol. 16, (2000), pp. 13–29.

 17.  H. Bromberg, “Are MUDS communities? Identity, belonging and consciousness in virtual worlds,” in R. Shields, Cultures of the Internet: virtual spaces, real histories, living bodies (London: Sage, 1996), pp. 143–152.

 18.  S. Turkle, Life on the screen: identity in the age of the Internet (New York: Simon and Schuster, 1995).

 19.  K. Durkin and C. Bryant, “Propagandising pederasty: a thematic analysis of the on-line exculpatory accounts of unrepentant paedophiles,” Deviant Behaviour: An Interdisciplinary Journal, Vol. 20, (1999), pp. 103–127.

 20.  M. Griffiths, “Sex on the Internet. Issues, concerns and implications,” in C. von Feilitzen and U. Carlsson (eds.), Children in the New Media Landscape (Sweden: UNESCO, 2000).

 21.  I. Granic and A.V. Lamey, “The self-organization of the Internet and changing modes of thought,” New Ideas in Psychology, Vol. 18, (1999), pp. 93–107.