Explanations given by child pornography
offenders for their crimes
Michael C. Seto,
* Lesley Reeves
& Sandy Jung
Royal Ottawa Health Care Group, Integrated Forensic Program, Brockville, Ontario, Canada;
Nova Scotia Department of Justice, Correctional Services, Forensic Sexual Behaviour Program,
Dartmouth, Nova Scotia, Canada &
Grant MacEwan University, Edmonton, Alberta, Canada
Abstract The explanations given by child pornography offenders for their crimes were explored in
two samples, one interviewed by the police as part of a criminal investigation and the second assessed
by clinicians following a child pornography conviction. There were many similarities across the
two samples with regard to demographic characteristics, criminal history and explanations. Many
offenders in both samples admitted possession of child pornography, a majority admitted they
deliberately accessed child pornography, and substantial minorities acknowledged their sexual interest
in child pornography and/or children. Similar proportions claimed curiosity or accidental access.
Relatively few offenders reported inter net addiction, child pornography as a substitute for contact
offending or indiscriminate sexual interests. There was evidence to suggest that the recently ar rested
offenders were more sexually deviant, as they were more likely to have images of boys, larger
collections, images depicting sexual violence or other paraphilic content and more involvement in
online trading and communication.
Keywords Child pornography; inter net offending; motives for crime; explanations; paedophilia;
There is increasing concern about child pornography offenders, paralleling large increases
in the number of arrests and convictions for child pornography offences in the past
five years (Mitchell, Wolak & Finkelhor, 2009; Motivans & Kyckelhahn, 2007; Wolak,
Finkelhor & Mitchell, 2005), and new legal and public policy initiatives such as the formation
of specialized law enforcement task forces and mandatory minimum sentences. Because
research in this area is new, there are many questions about the characteristics of child
The present study focuses upon the explanations given by child pornography offenders
to account for their crimes. Through examining what offenders say, we hope to shed light on
the motives for accessing child pornography. These explanations, in turn, would suggest
different offender management strategies. For example, someone who reports credibly child
pornography use as a result of sexual attraction to children requires different interventions to
*Corresponding author: E-mail: Michael.Seto@rohcg.on.ca
Journal of Sexual Aggression
(July 2010), Vol. 16, No. 2, pp. 169180
ISSN 1355-2600 print/1742-6545 online #2010 National Organisation for the Treatment of Abusers
someone who claims credibly that his child pornography use is a result of compulsive sexual
behaviour involving pornography in general. We are also interested in these explanations
because they may reflect denial or minimization of personal responsibility for the crimes (e.g.
claiming that their access was accidental when there is forensic computer analysis to show
downloading and repeated viewing), and denial or minimization of personal responsibility can
interfere with treatment or supervision compliance (Looman, Dickie & Abracen, 2005).
It has been suggested that there are meaningful differences between child pornography
offenders who use or produce child pornography as part of contact sexual offending;
those who seek child pornography in order to gratify their sexual interest in children through
masturbation and fantasy; those who collect child pornography because it is unusual or taboo
or because they are interested in many different kinds of pornography; and those who access
child pornography out of curiosity about the nature and availability of child pornography
(Lanning, 1992). One of the earliest studies of explanations given by child pornography
offenders was a qualitative study of interviews with 13 men convicted of downloading online
child pornography (Quayle & Taylor, 2002). Six categories of explanations for the use of child
pornography were identified by these authors:
1. As a means of achieving sexual arousal, where images were used as either a substitute
or a stimulus for contact sexual offending.
2. As a source of pleasure through collecting a complete series of images.
3. As a means of enabling online social relationships with like-minded others.
4. As a replacement for absent or unsatisfying relationships in the real world.
5. As therapy for exploring and dealing with one’s problems.
6. As a manifestation of the addictive properties of the internet.
A recently published review of the literature on internet child sexual offending suggests
that child pornography users fall into one of four major categories (Beech, Elliott, Birgden &
1. Those who fuel existing or developing sexual interests in children.
2. Contact sexual offenders who also use child pornography as part of a larger pattern
3. Impulsive and curious individuals.
4. Those who deal with child pornography for non-sexual reasons, such as financial
All these authors distinguish between sexual and non-sexual explanations for child
pornography offending; both typologies emphasize sexual interest in children as a prominent
explanation. There is evidence that being sexually interested in children is an important (but
not the sole) explanation for child pornography offending. Seto, Cantor and Blanchard
(2006) examined the sexual interests and behaviours of 685 male patients to investigate
whether child pornography offending was a valid diagnostic indicator of paedophilic sexual
preference, as assessed by phallometric testing. They found that child pornography offenders
were significantly more sexually aroused by stimuli depicting children than were contact
offenders with either child victims or adult victims. Both Bernard (1985) and Riegel (2004)
have also described the overlap between paedophilia and child pornography use. Three-
quarters of Bernard’s sample of 50 self-identified paedophiles reported that they collected
photographs of semi-nude or nude children, with a majority indicating that they took the
photographs themselves. Riegel (2004) obtained responses from 290 self-identified paedo-
philes (most of them attracted to boys) and found that 95% reported they had used child
pornography at some time in their lives.
170 M. C. Seto et al.
Although sexual interest in children is important, this does not rule out other
explanations for accessing child pornography and it certainly does not mean that someone
who seeks child pornography because of paedophilia might not give another explanation when
arrested and interviewed by the police. It also does not mean that explanations for child
pornography offending might not change over time or across different circumstances (e.g.
Bourke & Hernandez, 2009; Elliott, Beech, Mandeville-Norden & Hayes, 2009).
The present study
The present study was conducted to examine the explanations given by child pornography
offenders at two different stages of criminal justice involvement: post-arrest when questioned
by the police, and post-conviction when questioned by clinicians. We hypothesized that the
explanations given by offenders would be influenced by when they were questioned, reflecting
differences in the purpose of the interview (evidence for prosecution versus evidence for
clinical recommendations) and interview style (more probably adversarial versus more
probably therapeutic). We thought individuals questioned by the police at the pre-trial stage
would be more likely to explain that their child pornography offences were a result of internet
or pornography addiction, curiosity or accidental access, because these explanations would be
seen as less culpable than acknowledging a sexual interest in children or reporting that viewing
child pornography was a substitute for contact offending (implying the individual would
otherwise be at risk for contact offending). In contrast, we thought offenders questioned after
conviction would be more willing to acknowledge a sexual interest in children, indiscriminate
sexual interests or use of child pornography as a substitute for contact offending.
We did not have data from the same sample of child pornography offenders followed over
time, but we had access to data from two different samples, one questioned by the police and
the other interviewed by clinicians, to examine explanations in a preliminary way and set the
groundwork for future research following the same sample over time. We did not attempt to
test the validity of the explanations, as our focus was to describe the range of explanations
given and to examine the similarities and differences in explanations between the two groups
(representing two stages of criminal justice involvement).
Based on the research reviewed earlier, we predicted that the most common explanation
given would be that the child pornography material was sexually arousing. Based on anecdotal
and clinical reports, we also expected that other explanations implying non-paedophilic
motivations would be given (e.g. compulsive internet or pornography use, curiosity,
accidental access). Lastly, based on our assumption about the effects of stage of criminal
justice involvement, we predicted that the clinical sample of offenders would be more willing
to admit that they accessed child pornography on purpose and that it was sexually arousing to
them because their court cases had already been dealt with, whereas the police sample of
offenders would be more likely to claim that their crimes were a result of addiction, curiosity
or accidental access, because they were awaiting trial.
We obtained information from 50 closed case files made available by the Toronto Police Service.
All were individuals who were arrested, charged and eventually convicted of child pornography
offences: 17 offenders were convicted for accessing and possessing child pornography, and
33 offenders were convicted of possessing, distributing or making child pornography available
Child pornographers’ explanations 171
to others. Arrests were made between 1 March 2001 and 24 January 2007, with the majority
(60%) occurring between 2004 and 2006. The sample was comprised of adult men who ranged
in age at the time of their arrest from 19 to 65 years old [mean 37.6 years, standard deviation
(SD)12.0]. Other demographic information is presented in Table I. For the large majority
of the sample (92%), the index offence was the first time they had been questioned by
authorities about their child pornography use. The information on explanations was obtained
post-arrest but prior to criminal trial. Criminal history is also presented in Table I.
In contrast, the Edmonton clinical sample was assessed at the Northern Alberta Forensic
Psychiatry service following their conviction for child pornography offences. Some of these
offenders also participated in treatment at the service. A file search of adult sexual offenders
was conducted at an outpatient forensic psychiatric clinic that provides community-based
assessment, treatment, consultation and education services for forensic clients. The files of
34 male offenders who were convicted of possession and/or distribution of child pornography
were identified from the clinic database through therapists’ identifications of relevant cases
or treatment referral lists maintained by the clinic. Of the 34 offenders, half (50%) were
sentenced to terms of incarceration, 15 (44%) received community sentences in lieu of
incarceration and two (6%) received probation.
This sample of child pornography offenders was comprised of males who at the time of
their index offence ranged in age from 17 to 59 years (mean 33.8, SD11.2); offences
occurred between 1 November 1996 to 18 July 2006. Additional details about this sample
are also provided in Table I. The offenders’ ages at the time of their assessment ranged from
20 to 63 years (mean36.6, SD11.4), and referrals to the clinic spanned from September
2001 to May 2008.
Ta b l e I . Demographic characteristics, social functioning, and criminal history
Police sample (%)
Clinical sample (%)
Age (years) Mean37.6 Mean 33.8 t
SD12.0 SD 11.2 p0.15
Caucasian ethnicity 40/49 (82%) 31/34 (91%) 0.34
Married/common-law 9/41 (22%) 8/34 (24%) 1.00
Employed at time of arrest 39/50 (78%) 24/34 (71%) 0.45
Work or volunteer with children 12/36 (24%) 1/34 (3%) 0.001**
Social difficulties with adults 21/50 (42%) 10/34 (29%) 0.26
Previous child pornography offence 4/50 (8%) 2/34 (6%) 1.00
Previous contact offence (child) 7/50 (14%) 4/34 (12%) 1.00
Previous contact offence (adult) 2/50 (4%) 2/34 (6%) 1.00
Previous non-sexual violence 4/50 (8%) 1/34 (3%) 0.64
Previous non-violent offence 5/50 (10%) 10/34 (29%) 0.04*
No previous criminal history 32/50 (64%) 21/34 (62%) 1.00
*pB0.05; **pB0.01. Social difficulties with adults were coded as ‘‘yes’’ if the offender stated that they had
difficulties in their interpersonal relationships and/or had few if any rewarding relationships with adults. For
example, an offender might describe himself as a ‘‘loner’’ who did not have any friends. SD: standard
172 M. C. Seto et al.
The Toronto police sample was gathered from the police case files. These case files typically
had the police synopses of the offences, investigator notes and criminal records. Most files also
had hand-written or transcribed interview notes, a videotaped record of any statements made
by the suspect, forensic computer analysis focusing upon child pornography images and
samples of child pornography seized by the police. Some files also contained pre-sentence or
clinical assessment reports.
The Edmonton clinical sample was gathered from the files of sexual offenders who had
previously or were currently attending a forensic outpatient psychiatric clinic for services.
Most patient files included a pre-sentence or treatment report with the patient’s history that
was recorded by a psychologist, psychiatrist or social worker; collateral information on the
offence in the form of a police narrative or a prosecutor’s information sheet; and other
pertinent information (e.g. criminal record, third-party reports from probation or other
psychologists). The explanation data were taken from clinic reports, other file documents and
the police investigation records when available. Details regarding the content of the child
pornography were taken solely from confirmed collateral sources rather than self-report of the
offender. If such collateral information was not available, the child pornography content was
Twelve randomly selected files were coded by two independent raters to examine inter-
rater reliability. Inter-rater reliability was generally good, with kappas greater than 0.60 and
percentage agreements of 90% or higher. We retained two variables with poorer reliability*
claiming pornography addiction and claiming accidental access of child pornography*
because these explanations are mentioned frequently in the clinical literature. We attempted
to code minimizations of responsibility (blaming others, blaming personal circumstances or
blaming the availability of child pornography or internet access for the offences) but the inter-
rater reliability was too poor for these variables, so they were not analysed.
We coded demographic characteristics, criminal history and details about the child
pornography collections in terms of age and gender of children depicted, format (images,
videos, stories), explicitness of the contact and estimated volume. These study variables were
based on information available to us from the police or clinical files, respectively, and thus
were based on collateral sources of information such as forensic computer analyses conducted
by police officers or court documents. Age of children depicted was coded differently across
the two samples; in Toronto, any content in a particular age category was coded, whereas in
Edmonton the most prevalent age category was coded.
We also coded the explanations given by individuals during their police interviews
(Toronto) or during their clinical assessments (Edmonton). These study variables were based
upon offender self-report and thus are vulnerable to social desirability and other response
biases. These data were also vulnerable to variations across interviewers (police investigators
or clinicians) in the questions they chose to ask and in the information they recorded.
We coded as ‘‘yes’’ or ‘‘no’’ whether the offender admitted that they were in possession of
child pornography, whether they deliberately accessed child pornography and whether they
admitted that child pornography (or children) was (were) sexually interesting to them. We
also examined if offenders claimed that accessing child pornography was used a substitute for
sexual contacts with children.
Child pornographers’ explanations 173
In addition, we examined other sexual explanations that might be given, particularly
whether the offender claimed that he had indiscriminate sexual interests and thus child
pornography was just one manifestation of their online sexual activity, or whether they
claimed an addiction to pornography or the internet more generally. We also examined
explanations of collecting as a hobby or curiosity. Lastly, we examined whether the offender
claimed that the child pornography access was accidental or whether they claimed no recall of
why they accessed child pornography. In addition to the explanations that were given by child
pornography offenders, we also coded aspects of their child pornography-related activities.
These other aspects are summarized below.
Efforts to hide child pornography activities. Coded as ‘‘yes’’ if there was evidence the individual
had used encryption or other software in an attempt to hide their child pornography-related
activities (e.g. using a programme such as Evidence Eliminator to erase browsing history and
related records). This and subsequent variables were coded as ‘‘no’’ if there was no evidence,
or left missing if there was insufficient information to make a determination.
Online community participation. Coded as ‘‘yes’’ if there was evidence that the person had
participated in some form of online communication with others who were sexually interested in
children or child pornography (e.g. chatroom or newsgroup). In many cases child pornography
offenders used aliases or posed as someone else (e.g. a minor themselves).
Trading. Coded as ‘‘yes’’ if there was evidence of trading child pornography content with
others, including e-mailing, using peer-to-peer file-sharing programmes and setting up a
computer as a server for others.
Sexual content. Coded as ‘‘yes’’ for sexually explicit content if there were depictions of genital
touching, masturbation, oral sex or intercourse involving a child, either with another child or
with an adult. Coded as ‘‘yes’’ for sexually violent content if there were depictions of sexual
assault where a child appeared to be in distress or pain or it was clear there was coercion or
force involved (e.g. child was tied up). Coded as ‘‘yes’’ for other paraphilic content if there were
depictions of other atypical sexual themes, such as bestiality, sadism, masochism or fetishism.
Volume. The estimated number of child pornography images was based on forensic analysis for
the Toronto police sample. Only the explicitly stated number of child pornography files in the
police narrative, agreed statement of facts used in court or prosecutor’s information sheet
were coded for the Edmonton clinical sample.
The results from the two samples are compared in Tables I III. Table I describes the
characteristics of the child pornography users, Table II describes the explanations given for
the use of child pornography and Table III describes child pornography-related activity and
content. The numbers of offenders varied across comparisons because of missing data. We
compared the two groups using either a t-test (for age) or Fisher’s exact test for the
categorically coded variables (because of the small cell sizes).
The samples were similar in terms of average age, Caucasian ethnicity, marital status
and employment status. The average age and predominantly Caucasian ethnicity were
consistent with previous studies of child pornography offenders (e.g. Seto & Eke, 2005;
174 M. C. Seto et al.
Wolak et al., 2004). The police sample was significantly more likely to work or volunteer in a
role that brought them into regular contact with children than the clinical sample.
The two samples were also mostly similar in their criminal histories, although the
Edmonton sample was more likely to have a previous non-violent offence than the Toronto
Table II. Explanations for access of child por nography
Police sample (%)
Clinical sample (%)
Admitted child pornography possession 43/50 (86%) 31/34 (91%) 0.73
Admitted deliberate access 40/50 (80%) 22/34 (65%) 0.14
Admitted sexual interest in child
23/50 (46%) 13/34 (38%) 0.51
Claimed indiscriminate sexual interests 3/50 (6%) 1/34 (3%) 0.64
Claimed non-paedophilic sexual motivation 11/50 (22%) 3/34 (9%) 0.14
Claimed pornography addiction 5/50 (10%) 10/34 (29%) 0.04*
Claimed internet addiction 4/50 (8%) 4/34 (12%) 0.71
Claimed substitute for contact offending 3/50 (6%) 2/34 (6%) 1.00
Claimed collecting hobby 3/50 (6%) 2/34 (6%) 1.00
Claimed curiosity 20/50 (40%) 9/34 (27%) 0.25
Claimed accidental access 20/50 (40%) 11/34 (32%) 0.50
Claimed lack of recall 8/50 (16%) 1/34 (3%) 0.08
No explanation provided 11/50 (22%) 1/34 (3%) /2
One explanation 13/50 (26%) 2/34 (6%)
Two explanations 8/50 (16%) 8/34 (24%)
Three or more explanations 18/50 (36%) 23/34 (68%)
Changed explanations 26/50 (52%) 9/26 (35%) 0.22
Table III. Child pornography behaviour and content
Police sample (%)
Clinical sample (%)
Tried to hide child pornography 18/20 (80%) 8/34 (8%) 0.0001**
Participated in online child pornography
27/34 (79%) 13/34 (38%) 0.001**
Traded child pornography with others 22/32 (69%) 6/34 (18%) 0.0001**
Child pornography images 50/50 (100%) 33/34 (97%) 1.00
Child pornography video 32/40 (80%) 15/34 (44%) 0.002**
Child pornography stories 11/42 (26%) 4/34 (12%) 0.15
Any sexually explicit child pornography 40/46 (87%) 25/34 (74%) 0.15
Any sexually violent child pornography 17/32 (53%) 8/34 (24%) 0.02*
Any other paraphilic content 16/29 (55%) 3/34 (9%) 0.0001**
Mostly boy child pornography 15/41 (37%) 1/31 (3%) /2
Mostly girl child pornography 19/41 (46%) 12/31 (39%)
Both boy and girl child pornography 7/41 (17%) 18/31 (58%)
1100 items 6/37 (16%) 7/26 (27%) /2
1011,000 items 10/37 (27%) 13/26 (50%)
1,00110,000 items 9/37 (24%) 6/26 (23%)
10,000items 12/37 (32%) 0/26 (0%)
Child pornographers’ explanations 175
sample. The majority of offenders in both samples had no previous criminal history before
their arrest and conviction for child pornography offences. The two offender samples were
similar for many of the explanations they gave for their child pornography crimes. The clinical
sample of offenders, however, were statistically significantly more likely to claim an addiction
to pornography. The police sample of offenders were more likely to give no or only one
explanation, while the clinical sample of offenders were more likely to give multiple
(sometimes contradictory) explanations for their use of child pornography.
The two groups differed on several aspects of their child pornography activity and
content. There was a statistically significant difference between the two groups in terms of
online activity, with the police sample being much more likely to hide their child pornography,
participate in an online community or trade child pornography with others than the clinical
sample. There were significant missing data for the police sample, however. This may reflect a
sampling bias, in that the police sample of offenders were more likely to be detected by the
police as a result of their online activities, and/or a reporting bias, in that the clinical sample
was less willing to disclose their online activities than the police sample. The latter explanation
seems unlikely, however, as the clinical sample of offenders had already been convicted and
thus had less reason to hide online activities than Toronto offenders still undergoing police
The police sample of offenders were almost twice as likely to have child pornography
videos in their possession, more likely to have child pornography depicting sexual violence and
more likely to have other paraphilic content in the pornography examined by police than the
clinical sample. In addition, the police sample was more likely than the clinical sample to have
child pornography depicting mainly boys, rather than both boys and girls, and were more
likely to have 10,000 or more child pornography images in their possession.
A number of similarities were noted across the two samples despite coming from different
stages in criminal justice involvement (pre-trial versus post-conviction). They were similar in
terms of the demographic characteristics that were examined, previous criminal history and
many of the kinds of explanations that they gave. Many offenders in both samples admitted
possession of child pornography, a majority admitted that they deliberately accessed child
pornography and substantial minorities acknowledged their sexual interest in child porno-
graphy and/or children. Similar proportions also claimed curiosity or accidental access.
Contrary to our predictions, the police and clinical samples did not differ in the likelihood that
they explained their child pornography offending as a result of sexual interest in children or as
a result of curiosity or accidental access. The police sample were less, rather than more, likely
to report pornography addiction than the clinical sample. Relatively few offenders in either
group reported other explanations we coded, namely internet addiction than the clinical
sample, child pornography as a substitute for contact offending or indiscriminate sexual
interests. These low base rates make statistical detection of any true group differences on
these variables unlikely without much larger samples.
We were not able to examine all the explanations suggested by Lanning (1992), Quayle
and Taylor (2002) or Beech et al. (2008). In particular, we did not have data about collecting
child pornography because it is taboo (Lanning); as a rewarding behaviour in and of itself, as a
substitute for offline relationships or for self-help purposes (Quayle & Taylor, 2002); or as part
of a larger pattern of sexual offending (Lanning; Beech et al.). We did find support for the idea
that a sexual interest in children and/or in child pornography is relevant in many cases, as this
176 M. C. Seto et al.
explanation was given by a substantial proportion of child pornography offenders, whether
interviewed pre-trial by the police or post-conviction by clinicians. We also found support for
the idea that some child pornography offenders would claim curiosity or accidental access.
Consistent with Quayle and Taylor’s idea that involvement with online activities, including
child pornography, might reflect dissatisfaction or problems with real world relationships,
almost half the offenders were deemed to have social difficulties with adults. We did not access
online social relationships directly in this study, but a substantial number of child
pornography offenders, particularly in the police sample, participated in online communities
and traded child pornography with others. As we have noted already, few offenders in either
sample explained their crimes as a result of indiscriminate sexual interests, internet addiction,
collecting child pornography as a hobby or collecting child pornography as a substitute for
contact sexual offending.
Some unanticipated differences were observed between the two groups. The police
sample of offenders gave fewer explanations, possibly because they were interviewed post-
arrest but pre-conviction, while the clinical sample of offenders were more likely to claim
pornography addiction as an explanation of their crimes. There was also evidence to suggest
that compared to the clinical sample, the police sample were more likely to be sexually
deviant, because they were more likely to have content focusing upon boys, more likely to have
large collections of 10,000images, more likely to have child pornography depicting sexual
violence and more likely to have pornography depicting other paraphilic content (assuming
that, as is true for contact offenders, an interest in boys is more indicative of paedophilia than
an interest in girls; larger collections of child pornography indicate more involvement in terms
of time and effort, and involvement suggests a persistent motivation; and that possessing
depictions of sexual violence or other paraphilic content is suggestive of these sexual
interests). This finding is disturbing, because the police sample of offenders were also more
likely to work or volunteer in positions that brought them into contact with children. This may
reflect a sample selection bias, as police investigators have indicated that they prioritized cases
where they were aware that the suspect worked with children; at the same time, this could also
have been true for the investigators who arrested the offenders in the clinical sample.
Although their collections suggested otherwise, the two groups did not differ in the likelihood
that they admitted to having a sexual interest in child pornography or children, suggesting that
the police sample of offenders were more likely to deny this motive in light of their
Future directions and implications
We are particularly interested in examining the validity of the explanations given by child
pornography offenders. For example, one could test to see if offenders who acknowledge
sexual interest in child pornography or children respond differently on measures of sexual
interests*phallometric testing of penile responses, viewing time for sexual images*than
those who deny any paedophilic sexual interests, consistent with phallometric research on
contact offenders (e.g. Freund, Chan & Coulthard, 1979). Similarly, less intrusive measures,
such as the implicit association test, may differentiate those who admit versus those who deny
the commission of child pornography offences (see Nunes, 2009).
There may indeed be a typology of child pornography offending, with different motives
underlying different patterns of offending, but this needs to be tested further using objective
information. Our results provide support for some aspects of the typologies suggested by
Lanning (1992), Quayle and Taylor (2002) and Beech et al. (2008), but not all. Future
research could also test potential explanations that we did not address in this study,
Child pornographers’ explanations 177
for example, Quayle and Taylor’s (2002) idea that child pornography collecting may be a
rewarding behaviour in and of itself for some individuals (e.g. from the satisfaction of
completing a sought-after series of images).
Many offenders in both samples gave multiple and sometimes contradictory explanations
(e.g. claiming initially that child pornography was accessed accidentally but acknowledging
later in the interview that the content was sexually arousing). One could also assess change in
explanations over time in the same sample of child pornography offenders, and determine if
the explanations given are related to treatment or supervision compliance and behaviour. In
this study, we found that a third to half of the child pornography offenders changed their
explanation(s) at some point. Determining explanations for offending may be useful in
guiding clinicians to make decisions about what the treatment focus should be. For instance,
if an offender claims he uses child pornography to avoid contact offending, it may be
beneficial for the treatment provider to address the concern that he has these interests and
identify new ways to refrain from contact offending. Also, through questioning the offender
about his motives, supervisory professionals, such as probation officers, could enable
offenders to recognize the circumstances or states they would be in when approaching an
increasingly risky situation. For example, in the case of an offender who claims accessing child
pornography is a collecting hobby, probation could inquire about the offender’s use of their
leisure time (e.g. how much time, money and energy spent on the activity) and what other
goals he has that this leisure activity may be affecting, thereby focusing upon what their life
may look like with a significant reduction in this hobby.
It may also be the case that offenders with different motives also differ in the risk they
pose of future offending, with a particular concern about sexual contacts with children.
Research on paedophilia suggests that child pornography offenders who are motivated by a
sexual interest in children or by indiscriminate sexual interests may pose a greater concern
with regard to risk to offend against children than those who are motivated as a result of
compulsive behaviour involving the internet or pornography, curiosity or other non-
paedophilic reasons (see Seto, 2008).
There were a number of limitations with this study. First, it was a retrospective file review so
there were sometimes substantial amounts of missing information, depending upon the police
investigator or clinician who was involved, quality and detail of notes and foci of interviews.
This might account for why fewer clinic-assessed offenders were coded as trying to hide child
pornography, participate in online communities or trade child pornography with others; not
because they were less likely to engage in these activities, but because the police were
particularly interested in these aspects and were therefore more likely to ask relevant questions
in their interviews.
Another issue is that some of the data relied on self-report, with obvious reasons for
offenders in both samples to minimize their perceived culpability, especially through the
explanations they provided (Nugent & Kroner, 1996; Pollock & Hashmall, 1991). We expect
that some individuals who denied a sexual interest in child pornography or in children were
lying; in contrast, we do not expect individuals who did not have any such sexual interests to
report them, given the potential sanctions for this admission. The veracity of the offenders’
explanations cannot be determined, because we did not usually have access to collateral
information (e.g. individuals who claimed they had child pornography because of a general
pornography addiction may have sought treatment for pornography addiction previously or
had problems as a result of their adult pornography use in previous relationships or work).
178 M. C. Seto et al.
The potential for self-reporting biases was particularly true of the clinical sample, where more
of the data relied on self-report, whereas we had access to police investigator notes and
forensic analysis for the Toronto sample. It would be useful to replicate this study using a
standardized and prospective data collection procedure, to ensure that adequate information
is obtained for all the study variables.
A third limitation of this study is the small sample size, with a total of only 84 child
pornography offenders. This limited our statistical power to detect differences, especially for
variables with relatively low endorsement rates (e.g. claiming indiscriminate sexual interests).
Replication of this research with larger samples would be helpful in understanding the
explanations and motives of child pornography offenders.
Finally, our interpretation of the group comparisons is cautious, because there are many
differences between the Toronto and Edmonton samples that might explain the differences in
explanations that were observed. Ideally, we would have followed a larger sample of child
pornography offenders over time, to see if the explanations they gave changed from the time
they were investigated and interviewed by police to the time they were referred for a clinical
evaluation. It is interesting to note, however, the similarity we observed on many of the
We believe the results of this study add to our knowledge of the motivations of child
pornography offenders and to the development of appropriate treatment or supervision
strategies. For example, a substantial minority of child pornography offenders in both samples
acknowledged being sexually interested in children or child pornography, consistent with the
findings reported by Seto et al. (2006), using objective testing of sexual arousal patterns. This
suggests that the effective management of child pornography offenders needs to address these
sexual interests. In contrast, indiscriminate sexual interests or pornography addiction were
reported infrequently and are likely to be rare considerations in the treatment or supervision
of child pornography offenders.
Beech, R. B., Elliott, I. A., Birgden, A., & Findlater, D. (2008). The internet and child sexual offending: A
criminological review. Aggression and Violent Behavior,13,216228.
Bernard, F. (1985). Paedophilia: A Factual Repor t. Rotterdam: Enclave Press [in German].
Bourke, M. L., & Hernandez, A. E. (2009). The ‘‘Butner Study’’ redux: A report of the incidence of hands-on child
victimization by child pornography offenders. Journal of Family Violence,24,183191.
Elliott, I. A., Beech, A. R., Mandeville-Norden, R., & Hayes, E. (2009). Psychological proﬁles of internet sexual
offenders: Comparisons with contact sexual offenders. Sexual Abuse: A Journal of Research and Treatment,21,7692.
Freund, K., Chan, S., & Coulthard, R. (1979). Phallometric diagnoses with ‘non-admitters’. Behaviour Research and
Lanning, K. V . (1992). Child Molesters: A Behavioral Analysis. Report no. NC-70. Washington, DC: National Center
for Missing and Exploited Children. Available at: http://www.missingkids.com/en_US/publications/NC70.pdf
(accessed 9 October 2009).
Looman, J., Dickie, I., & Abracen, J. (2005). Responsivity issues in the treatment of sexual offenders. Trauma, Violence,
& Abuse,6, 330353. DOI: 10.1177/1524838005280857.
Mitchell, K., Wolak, J., & Finkelhor, D. (2009). The National Juvenile Online Victim Study: Methodology Report. Crimes
against Children Research Center. Available at the Crimes Against Children website: http://www.unh.edu/ccrc/pdf/
N-JOV2_methodology_report.pdf (accessed 31 July 2009).
Child pornographers’ explanations 179
Motivans, M., & Kyckelhahn, T . (2007). Bureau of Justice Statistics bulletin: Federal prosecution of child sex exploitation
offenders, 2006. Available at the US Department of Justice, Ofﬁce of Justice Programs website: http://
www.ojp.usdoj.gov/bjs/pub/pdf/fpcseo06.pdf (accessed 3 July 2008).
Nugent, P. M., & Kroner, D. G. (1996). Denial, response styles, and admittance of offenses among child molesters and
rapists. Journal of Interpersonal Violence,11,475486. DOI: 10.1177/088626096011004002.
Nunes, K. L. (2009). Measuring child molesters’ implicit cognitions about self and children. In D. Thornton &
D. R. Laws (Eds), Cognitive Approaches to the Assessment of Sexual Interest in Sexual Offenders (pp. 125143).
West Sussex, UK: John Wiley.
Pollock, N. L., & Hashmall, J. M. (1991). The excuses of child molesters. Behavioral Sciences and the Law,9,5359.
Quayle, E., & Taylor, M. (2002). Child pornography and the internet: Perpetuating a cycle of abuse. Deviant Behavior,
Riegel, D. L. (2004). Effects on boy-attracted pedosexual males of viewing boy erotica [Letter to the Editor]. Archives
of Sexual Behavior,33,321323.
Seto, M. C. (2008). Pedophilia and Sexual Offending Against Children: Theory, Assessment, and Intervention. Washington,
DC: American Psychological Association.
Seto, M. C., Cantor, J. M., & Blanchard, R. (2006). Child pornography offenses are a valid diagnostic indicator of
pedophilia. Journal of Abnormal Psycholog y,115,610615. DOI: 10.1037/0021-843X.115.3.610.
Seto, M. C., & Eke, A. W. (2005). The criminal histories and later offending of child pornography offenders. Sexual
Abuse: A Journal of Research and Treatment,17, 201210. DOI: 10.1007/s11194-005-4605-y.
Wolak, J., Finkelhor, D., & Mitchell, K. J. (2004). Internet-initiated sex crimes against minors: Implications for
prevention based on ﬁndings from a national study. Journal of Adolescent Health,35, 424.e11424.e20.
180 M. C. Seto et al.