“It was in Control of Me”: Notions of Addiction and Online Child Sexual Exploitation Material Offending

Abstract

This paper focuses on notions of “addiction” among users of online child sexual exploitation material (CSEM) through a comparative analysis of two qualitative studies. The first is a 17-month anthropological ethnography (participant observation, interviews, and focus groups) in UK group programs for CSEM users, and the second is based on interviews with individuals in sexual offense treatment units of a US prison. We thematically analyze the narratives of 103 CSEM users at different timepoints and settings from pre-trial to incarceration. Those citing “addiction” focused on three areas. First was pornography progression. Second were perceived indicators of “addiction” and alignment/analogy to other addictions, including ideas about losing control; ignoring detrimental consequences and continuing; physiological signs similar to withdrawal; likening to substances; and progression from “softer” to “harder” material. Third, less common was rejecting the “addiction” label, citing choice and responsibility. We situate these results within debates and literature regarding pornography and Internet “addiction”; implications of the label; societal conceptions of sexual offending; harms of CSEM; and treatment/prevention considerations. Lastly, highlighting the merit of interdisciplinary comparative qualitative analysis, we demonstrate similarities in narratives despite differences in location, timeframe, setting, conviction status, intervention programming, and research methods.

Keywords

child sexual exploitation material pornography internet addiction sexual offending

The proliferation of child sexual exploitation material (CSEM), and the reduction in barriers for CSEM consumption, are concerning developments in the digital age. Recently the National Center for Missing and Exploited Children has documented reviewing over 322 million images and videos of CSEM (National Center for Missing and Exploited Children, 2022), while studies of the largest peer-to-peer network have reported the most prevalent query and 1% of all queries being about CSEM (Steel, 2009), with 775,941 computers in over 100 countries sharing it (Wolak et al., 2014).

Alongside uptake of digital technologies, there has been debate about whether to classify new “addictions” related to both the Internet and pornography. More generally, there is also debate about how to conceptualize problematic pornography use (Kraus & Rosenberg, 2014). The present paper contributes to such debates by reporting on notions of “addiction” among CSEM users. The article is rooted in a comparative analysis of two qualitative studies from the UK and USA, encompassing 103 CSEM users at different timepoints and settings from pre-charge/pre-trial to sentence. The focus is not on how current psychological practice conceptualizes these issues; rather, we seek to emphasize how participants conceptualize their behavior and the implications of this.

We begin by detailing the various ways in which problematic pornography and Internet use has been understood in clinical and academic spheres, followed by outlining our methodologies for the studies and this paper. We then discuss how CSEM offending was sometimes cast by participants as an attempt to exercise power and control. What follows is the substantive focus of the article: participant narratives of pornography progression and “addiction”; their indicators of “addiction” and the alignment to characteristics of other addictions; and the less common rejection of an “addiction” label. We explore possible functions that participant narratives serve, and how the perceived process of “addiction” and subsequent framing/telling of life events may culminate in potential turning points for change and desistance or not (Presser & Sandberg, 2015). We end with a discussion about the results in a broader context and literature, including debates about new “addictions,” societal conceptions of sexual offending, harms of CSEM, and treatment and prevention considerations. Most striking throughout are similarities across participants from different countries, research settings, intervention programs, timeframes, and engaged using diverse methodologies. This article thus provides insight into CSEM offending that moves beyond individual research projects and associated limitations, while highlighting the merit of interdisciplinary comparative qualitative analysis for studying this criminal phenomenon.

“Addiction,” Pornography, and the Internet

For clinicians and researchers, a divisive question is whether problematic pornography and Internet use can be considered “addictions.” Problematic pornography use has largely been constructed using traditional conceptions of addiction, which center on a loss of self-control (Taylor, 2019). Furthermore, by 1981, the World Health Organization (WHO) established a definition of addiction that could fit “virtually any behaviour that is substituted for a prior behaviour – even behaviours that entail no use of psychoactive substances” (Reinarman, 2005, p. 312). While the DSM generally uses this term to describe addictions to substances, the behavior of gambling was recently included in the DSM-5 (American Psychiatric Association, 2013), supporting the notion of behaviors as addiction.

Early work surrounding pornography-as-addiction described repetitive misuse as self-medication to regulate negative affective states, and a loss of control or inability to control one’s behavior (Carnes, 1983, 1989). At present, the concept of pornography and sexual addiction continues to be contentious (Walton et al., 2017), with multiple other terms and classifications developed to describe clusters of sexual behaviors that include problematic pornography use. Researchers have suggested that excessive pornography use may be indicative or a symptom of a paraphilic disorder (Kafka, 2001). In some cases, it can be classified as a symptom of other psychological issues such as obsession and compulsion (Cooper et al., 2004) or an impulse control disorder (Shapira et al., 2003). The WHO (2020) recently added “compulsive sexual behavior disorder” (inability to control urges and impulses resulting in repetitive sexual behavior) as a type of impulse control disorder, under which pornography use could be one among multiple behaviors causing distress. The concept of “hypersexuality” has also been used to classify individuals who have difficulty regulating sexual behavior (Reid, Li, et al., 2011). Symptoms include repetitive fantasies, increases in engagement and preoccupation with sexual fantasies and behavior, and inability to control sexual fantasies or behavior despite attempts, where these cause distress and potential harm to oneself or others (Kafka, 2010). Some have argued for the inclusion of hypersexuality in the DSM as its own disorder and not as a symptom of other disorders (Reid, Garos, et al., 2011), but this is currently not the case. Despite these developments, the addiction framework continues to permeate discourses of frequent/compulsive or deviant pornography use (Taylor, 2019). Specifically regarding CSEM users, Quayle and Taylor (2002) note that in interviews with 13 participants, all “made reference to the Internet and addiction when talking about the compulsive elements of downloading” (p. 352).

Similar to problematic pornography use, debates surrounding “Internet addiction” and “Internet sexual addiction” are vast (Duffy et al., 2016; Pies, 2009). Those who agree Internet addiction (e.g., Young, 2009) and Internet sexual addiction (e.g., Griffiths, 2012; Young, 2008) are disorders point to “alarming prevalence rates” (Cash et al., 2012, p. 292). These are said to be accompanied by similar symptoms associated with other addictions: mood changes; preoccupation with the Internet; inability to control length of time online; increased Internet usage to create a desired mood; withdrawal; and continued habitual use despite negative consequences for familial, work, or academic commitments (Cash et al., 2012; Kirmayer et al., 2013).

Those who criticize these concepts suggest that “addiction” is a metaphor, as the Internet and pornography are not substances, and that there is a lack of sufficient evidence to support an addiction framework (Kirmayer et al., 2013; Taylor, 2019). This metaphor is controversial because, “Just as you can substitute the words internet pornography for the word gambling to generate a new disorder, so you can for a huge number of other words representing behaviours that are known to be taken to extremes” (Dunn et al., 2012, p. 275), including potentially nearly all activities of daily life (Billieux et al., 2015). Critics further argue that the Internet is a medium used to feed other compulsions as opposed to an addiction itself; for example, individuals preoccupied with sex can reinforce this online (Blaszczynski, 2006). Therefore, some argue that there is a need to distinguish between addictions to, versus on, the Internet (Widyanto & Griffiths, 2007).

Problems with agreement arise because consistent definitions do not exist, sample sizes are often from small subpopulations (e.g., students), and diagnoses are regularly based on subjective judgments as opposed to robust clinical criteria, making data comparison difficult (Blaszczynski, 2006; Dunn et al., 2012; Schell, 2020; Widyanto & Griffiths, 2007). Because of such inconsistencies, studies cite “Internet addiction” prevalence rates ranging from 0.3% to 38% depending on definitions used (Cash et al., 2012). Furthermore, while there is an abundance of self-help resources targeted at individuals with apparent online pornography “addiction” (e.g., Carnes et al., 2007; Wilson, 2017), both Internet addiction (Schell, 2020) and compulsive sexual behavior (Kraus et al., 2016) are not in the most recent DSM-5. Measures to assess the former are said “to be lacking a rigorous analysis on their reliability and validity that justifies their widespread usage” (Schell, 2020, p. 687), while the ability to diagnose sexual addiction is impeded by “significant gaps in understanding” (Kraus et al., 2016, p. 2097). It is within this larger context that the present research was undertaken.

Method

The methodology for this paper combines a sub-set of results from two projects. Both authors have published previous articles from their respective studies, using distinct aspects of the data about topics not covered here (Holt et al., 2021; Rimer, 2017, 2019, 2021). For this article, through a targeted comparative analysis from Studies 1 and 2, results were combined toward the theme of addiction. We report the methods for each study and the present paper below, including how we determined sample size, data exclusions, and all analyses.

Methods and Participants: Study 1

Study 1 was anthropological and ethnographic, investigating three research questions:

(1) How does looking at CSEM fit into participants’ lives, and how does the Internet facilitate this?

(2) How do participants’ constructions, perceptions, and actions compare to dominant cultural understandings of childhood and sexuality?

(3) What are the efforts to normalize participants, their actions, and their perceptions about the Internet, sexuality, and childhood/children?

The project included 17 months of participant observation in community-based UK group programs for 81 individuals arrested for CSEM (10 distinct groups over their full programs, totaling close to 100 sessions); 31 semi-structured interviews with group members; 15 semi-structured interviews with staff from the administering agency; and two staff focus groups. The groups were psychoeducational (but not labeled “treatment”), in two locations, and had one three-hour session per week with between six and nine members and two facilitators. The program was voluntary: it was pre-trial for most, and not mandated by courts. Participants were at varying stages in the justice process (ranging from recent arrest to a year-and-a-half awaiting trial), and on occasion, they received sentences while in the program. Seventy-seven participants (95%) completed the program, while two left early for prison sentences, and two left by choice (5% total non-completers). Knowing participants’ motivations for taking part is difficult. It is feasible that some recognized their behavior as problematic and were striving for change, that some wanted to convey to others an attempt to change, and/or that some wanted to project a positive image for court. By request, and after concluding, participants could be given a letter that summarized the program and their individual attendance record. Beyond this, the groups were not connecting to pending cases; in the context of the program, staff did not evaluate/assess participants for court cases, write reports for court, or testify.

During participant observation for Study 1, the researcher took notes, observed, and spoke openly, however did not make changes to program content or structure. Information from groups was documented in fieldnotes, while interviews and focus groups were recorded and transcribed. Interviews were one to two hours in length, semi-structured, and used a 25-question guide so that data could be compared. This had five sections: background; Internet and pornography; children and childhood; insights into offending; and current circumstances (for example questions see Rimer, 2017, 2019, 2021).

Involvement in the research was voluntary, and participants’ ability to take part in the program was not affected if they chose not to consent. Participants received an information sheet in the mail prior to the start of groups as part of their welcome package and met the researcher at the beginning and break of session one to ask questions. They then decided about consent from home away from the agency, researcher, and fellow groups members. All did consent; however, the process was setup such that the researcher would only take part if every member consented independently. Every person gave written informed consent to have the researcher in groups, and again if interviewed. All were assured the researcher would not contact them outside of groups. Participants were also guaranteed confidentiality in line with agency policy (information was confidential unless it indicated undisclosed abuse of a child or that a child was in danger), and as much anonymity as possible (e.g., first names only in interactions and full anonymity in research material/outputs). The project received university ethics approval prior to fieldwork.

Selection of participants was dependent on the agency and was not separate from the program’s admissions process. Corresponding to literature (Babchishin et al., 2011; Henshaw et al., 2017; Seto, 2013), all CSEM user participants were men, all except two were white (98%), and their ages ranged widely (20s–70s with a cluster in middle age). Thirty-six had children (44%), while 39 did not (48%), and six were unknown (7%). Fifty-eight had current or ex-partners or spouses (72%), while 19 did not (23%), and four were unknown (5%). Eight spoke about their abuse as children (10%). Participants’ professions were heterogeneous and in industries including civil service, healthcare, education, IT, defense, trades, transportation, and unemployment. In terms of former justice system contact, 71 (88%) were arrested for CSEM only and sexual offenses for the first time. The remaining 10 had prior CSEM convictions (n = 5; 6%), previous contact offenses against children (n = 1; 1%), a voyeurism conviction (n = 1; 1%), an arrest for sexual assault against an adult (n = 1; 1%), and concurrent online grooming charges (n = 2; 2%). All were arrested for viewing or possessing CSEM, with fewer for distributing. None were known to have produced CSEM.

The analytical process for Study 1 was exploratory and inductive: there was no pre-determined framework for coding, nor hypotheses to test. Transcripts and fieldnotes were coded openly and clustered for themes based on frequency, consistency, and outliers rooted in participants’ statements and actions (18 themes in total by conclusion). To ensure findings were accurate, analyses compared themes across field sites, groups, sessions, and individuals. Fieldnotes were most helpful in analyzing group processes and developing themes across the sample. Information from fieldnotes could then be expanded in interview analysis, along with an assessment of similarity in results across both interviews and groups. In short, the method was informed chiefly by thematic analysis (Braun & Clarke, 2006; Nowell et al., 2017). However, as is convention in anthropology, themes were adapted in an ongoing manner.

Groups were safe spaces in which participants were encouraged to speak freely without judgment. Being pre-trial, the setting is also unique in the CSEM literature. However, there were limitations. First, the researcher did not have access to case files, so what was said could not be compared to forensic evidence. Second, in groups, the researcher was limited to content from the curriculum. Third, the research took place post-arrest, meaning participants had time to reflect and perhaps construct narratives, thus not necessarily representing what they thought while offending. Fourth, the sample was self-selecting and cannot be generalized (though this is not the goal). Finally, the timing of groups might have meant some participants altered their statements to appear favorable or align with facilitator expectations; however, the purpose again was not to test truthfulness.

Methods and Participants: Study 2

Study 2 involved semi-structured interviews with 101 individuals convicted of a sexual offense who were incarcerated in specialized treatment units in a minimum-security US correctional institution. The broader project focused on participants’ stories, the perceived role of pornography, and the desire for access to pornography post-release. The research questions were:

(1) Are there distinct differences among individuals and the perceived role of pornography in their sexual offenses when compared using risk levels as measured by the STATIC, STABLE, and the overall combined score?

(2) What are the common themes among individuals’ experiences regarding pornography and how do they construct the role of pornography in their offending and in their lives post-release?

After university ethical review and approval, data collection occurred over 4 months. Participation in the study was voluntary. To recruit, mental health staff described the research to all those housed within the treatment units. Interested individuals were asked to indicate this via signup sheets posted in common areas of these units, and participants signed up using their prison ID numbers. These persons were scheduled for interviews by mental health staff. All interviews occurred in the office of mental health staff, where auditory privacy could be maintained. The researcher completed the informed consent process with each participant, which emphasized several key elements. First, the researcher explained that, although she was granted access to participants and physical space in the institution, she was not affiliated with the Department of Corrections. Second, responses would only be manually documented, and although mental health staff would know which individuals participated, they would have no access to responses, and no identifying information would be collected or maintained. Lastly, participation had no effect on parole or release decisions. Additionally, due to the sensitive topic, the researcher attempted to create a space where participants were comfortable talking candidly, free of judgment. This was done through open dialogues, empathetic and active listening, and when appropriate, humor.

Interviews were documented by hand as the institution did not allow recording. To increase accuracy, a research assistant was present for all interviews. Both the researcher and assistant recorded responses and notes, which were compared prior to analysis and combined to increase reliability. Each interview lasted 30–45 minutes. An interview guide that consisted of 30 open-ended questions was used to allow for comparison of responses. Questions focused on experiences of pornography such as age of onset, genres watched, duration of sessions, frequency of viewing, and effects on self, sexual relationships, fantasies, and behavior. Basic demographic information was also collected (age, race/ethnicity, prior occupation, relationship/marital status, and convictions). Pornography was defined as any visual content or literature used to stimulate sexual feelings, become aroused, or achieve orgasm, which is consistent with research and practice definitions (Diamond, 2009).

All participants were male and aged 20–67 (mean of 40). The majority were white (n = 73; 72%). Black males represented nearly a quarter of the sample (n = 24; 24%), while three others were Hispanic (3%), and one was Native American (1%). In terms of criminal sexual conduct (CSC) convictions, individuals were serving sentences for CSC third degree (n = 28; 28%), CSC second degree (n = 20; 20%), CSC first degree (n = 18; 18%), CSEM (n = 15; 15%), contact offenses and CSEM (n = 7; 7%), indecent exposure (n = 5; 5%), lesser sex crimes (n = 5; 5%), and crimes with a sexual element (n = 3; 3%).

The analytic process for this paper with the Study 2 data, along with how it relates to the inductive analysis outlined in the previous section from Study 1, is detailed in the next section. Study 2 also had limitations similar to Study 1. First, as the researcher did not have access to investigation or treatment files, the veracity of information could not be confirmed. Second, while the researcher clarified that she was not affiliated with the Department of Corrections, interviews took place in spaces used by prison employees. Thus, building trust and rapport was likely constrained by the environment. Related, while US mandatory disclosure laws do exist and can potentially affect research of this nature, these did not impact interviews (the researcher conducted the interviews, was not part of the Department of Corrections, was not a mental health provider, and did not ask questions that could result in the need for mandatory reporting). Third, the research took place post-conviction, meaning participants had ample time to reflect and perhaps construct narratives surrounding their behaviors, and so the responses may not have mirrored their pre-conviction thoughts or actions. Fourth, as participation was voluntary, the sample was self-selecting and cannot be generalized. Finally, it could be possible that some participants altered their statements due to social desirability or the belief that participation would benefit the parole process.

Methods and Sample: The Present Paper

This specific paper stems from a targeted thematic analysis (Braun & Clarke, 2006; Nowell et al., 2017). Through open coding and then clustering of codes into themes, the completed analysis of Study 1 established “addiction” as a theme in that study. Initial exploration of data in Study 2 also found “addiction” to be a noteworthy recurring topic. To systematically examine similarities and differences between studies/samples, findings from Study 1 were first generated following inductive thematic analysis of the pre-conviction UK sample (all CSEM users) as outlined above. The theme of “addiction,” along with related sub-themes, then provided a coding framework for Study 2 to use deductive focused coding and explore if the same themes applied to the CSEM users in the post-conviction US sample. This coding framework included the following:

⁃ labeling offending an “addiction”

⁃ progression

⁃ desensitization

⁃ increasing tolerance

⁃ increasing depravity

⁃ decreasing emotional response

⁃ loss of control

⁃ unable to resist

⁃ likening CSEM to drugs

⁃ likening CSEM to alcohol

⁃ ignoring potential consequences

⁃ physiological symptoms

⁃ feeling a buzz

⁃ contrast of “real” you to “addict”

⁃ addictive personality

⁃ abdicating responsibility

⁃ removing blame

⁃ lack of empathy

⁃ collecting

While some of the above may not seem straightforwardly tied to “addiction” (e.g., lack of empathy), these were identified as relevant because/when they were mentioned in the context of discussing “addiction” or tied to an “addiction” narrative. For example, in the Results, we quote participants who focus on lack of empathy when drawing comparisons and making analogies between their CSEM use and characteristics of drug addiction.

With the deductive coding, analysis was not rigid: new information from Study 2 was also documented, added, and then re-compared to Study 1. These additional codes included:

⁃ attempting to conceal

⁃ feeling shame and guilt

⁃ lacking in relationships

⁃ escaping

⁃ CSEM as coping

⁃ setting boundaries

⁃ learning to control addiction

⁃ improving resources/preventing

⁃ rejecting addiction construct

The rationale for the above coding process was threefold. First, the analysis focused on one theme, and so the method was best suited to pinpoint this theme within two large datasets; analysis was around a specific theme, as opposed to open coding of both datasets followed by a much larger comparison of every possible theme (Braun & Clarke, 2006). Second, Study 1 was used as a basis for the “addiction” coding framework because this project was open-ended, longer than Study 2, and had a larger sample of CSEM users (81 participants over 17 months of fieldwork vs. 22 participants over 4 months of fieldwork). This provided a large amount of data initially examined in an open, inductive manner from which to guide the analysis, as opposed to using Study 2 as a basis, which was more limited in scope and size and therefore likely to produce a less exhaustive set of codes. However, as noted, Study 1 provided a framework, but the process was iterative as opposed to exclusionary. Third, Study 1 focused solely on CSEM offending and CSEM users, and the notion of “addiction” was a common theme. Study 2 dealt more broadly with pornography and both online and offline sexual offenses, and so by specifically using a purposive sample of CSEM users from Study 2 to compare to Study 1, deductive focused coding was the most appropriate next step in order to precisely and rigorously explore the theme in Study 2.

As each individual study was conducted separately and had its own ethical approval with data access restrictions and associated informed consent procedures, the raw data could not be and was not shared between authors. This means that the primary researcher of each study coded their respective dataset (one coder per study). O’Connor and Joffe (2020) note the importance of intercoder reliability, and while intercoder reliability could not be measured here, the researchers underwent a rigorous and systematic process through a reflexive and team-centered approach to limit inconsistencies. Specifically related to thematic analysis, Nowell et al. (2017, p. 4) state that to ensure trustworthiness, researchers can partake in practices including “prolonged engagement with data,” “use of a coding framework,” “persistent observation,” “researcher triangulation,” “peer debriefing,” “documentation of team meetings,” “team consensus on themes,” and “themes and subthemes vetted by team members.” We engaged in each of these. The researchers worked long-term in fieldwork and with their data. A coding framework was used throughout and was updated as the analysis evolved. The researchers came together frequently to review the coding structure, which was derived from the data. Each meeting, the researchers discussed coding decisions, shared findings from their individual datasets, debriefed, and then went back to their respective datasets, ensuring both researcher agreement/consensus and inclusion of outliers. In each meeting, clear outcomes were defined, and goals for the next meeting were determined. This procedure was repeated throughout the coding process to strengthen communication, collaboration, and the analysis (Campbell et al., 2013; Nowell et al., 2017). These meetings represented intercoder agreement, where coding decisions are clarified and any potential discrepancies are identified (Garrison et al., 2006). This reflexive and iterative process then culminated in the complete “addiction” results for this paper. In addition, it is worth noting that according to Braun and Clarke (2019), intercoder reliability measures are often not appropriate for thematic analysis, as reliability measures stem from a positivist framework which are then attempting to be applied to a different methodological, analytical, and epistemological approach.

The sample for this article includes all 81 participants in Study 1 (as all were CSEM users), and the 22 individuals in Study 2 who had a CSEM conviction (CSEM-only and dual offenses). Therefore, the total sample size for this paper is 103. We argue that including those with dual offenses (both CSEM and contact offenses) is warranted. First, all those included were involved with the justice system for CSEM. Second, one participant in Study 1 had previous dual offending convictions, so both samples included these individuals. Third, CSEM users may have contact offenses unknown to authorities (Neutze et al., 2012). Thus, excluding those with dual offenses from Study 2, and at the same time assuming those in Study 1 had no contact offense history due to lack of charges, is not sound reasoning. Instead, including all participants with known CSEM offenses provides a unique opportunity to compare pre- and post-conviction samples derived from different countries, settings, programs, timeframes, and research methods. This can demonstrate commonalities among diverse groups, providing information about CSEM offending that transcends individual studies and their associated limitations.

Results

In presenting results, we now follow a prominent participant narrative, which begins with descriptions of CSEM offending by some as an attempt to exercise control or power. Following on from this and continuing participant narratives, the paper moves to notions of pornography progression, finally culminating with apparent “addiction,” which represents the extreme end of progression and a loss of control. Please note that throughout, ellipses in quotes signify that a word or words have been removed but the meaning remains unchanged.

Offending as an Attempt at Control and Power

For some participants, engaging with CSEM was described as an attempt to exercise control or create power, often in a life claimed to be otherwise scattered and meaningless. Here, offending was sometimes a short-term coping mechanism or “escape” that provided a feeling of control when facing depression, sadness, low self-esteem, or stress:

An escape mechanism, a coping mechanism. An avoidance of what I should be really sorting out about myself. (Study 1, P72)

I would look at the pictures for stress relief. I would sit there for hours just looking…I had really bad anxiety and was working and taking classes. I used it as an escape not to think about anything else. (Study 2, P45)

For others, offending was an attempt to feel powerful:

…it gave me some, um, some, some control. Um, that perhaps I felt was lacking in my real, relationships and sexual relationships. Um, some power perhaps. Um, that I felt was lacking. (Study 1, P63)

Having those pictures. I had like three or four on my phone. They were like the popular girls. So it gave me a sense of power. (Study 2, P23)

Sometimes control and power was also felt because participants had a secret that was theirs to own: “Acting in secret was the thing for me…I never had to tell anyone, it was mine” (Study 1, P48). In Study 1, P15 discussed how he “enjoyed having a secret no one knows about” while others remarked, “It’s like finding yourself in a special world, not like your neighbor’s world” and “It’s like I could see someone else’s secrets…every time I saw something new and something forbidden, I would get a thrill” (P28, P10). In Study 2, a participant similarly noted that “the dark web leads to deeper and darker secrets” (P10). Control manifests here in having the power to decide if/when to disclose secrets, or knowing taboo facts about others.

Another expression of control was evident in those citing collecting as a motivation. Like using pornography or CSEM to compensate for emotional, social, and relationship difficulties, collecting was seen by some as a way to create meaning, and thus was a short-term solution to unresolved insecurities. Taylor and Quayle (2003) note that collecting sometimes does provide motivation beyond only sexual aspects of CSEM offending. Speaking to this, some participants amassed large collections of all kinds of pornography:

I was collecting adult pornography, like there’s no tomorrow…one was a one-terabyte hard drive. That was the latest one. The others were 80-gig and 250-gig, and I’d filled them all. (Study 1, P66)

I had over 100,000 pictures. Every age. I was lazy, I didn’t get rid of anything. (Study 2, P14)

For some, collecting also involved constantly looking for new pictures, or attempting to find the “perfect” image or video, while simultaneously never really being able to describe it:

…you’re looking for the ultimate picture, or, or film, even though you don’t know what it is you’re looking for. You know, there has to be that one out there that, you’d look at it and say, “I’ve found it.” (Study 1, P46)

The compulsion is looking for that perfect image. You don’t know where it is, what it is. It’s like looking for the end of a rainbow, it’s never there. (Study 1, P65)

I always looked for new pictures. I would get bored and look for better ones. (Study 2, P14)

Building a collection was thus “controlling something that is yours that you give order to” or, put even more simply, “something you can control” (Study 1, P9, P72).

At an extreme, sometimes collecting became engrossing, including bulk downloading while away from the computer. For example, a participant in Study 1 filed and organized CSEM with “thousands of folders.” On workdays, he would spend two to three hours doing this, while on weekends it would take up more time. He described this collecting as “satisfying” and a “mastery experience.” He told his group it was “the only satisfactory thing in my life.” In Study 2, P54 similarly noted that he “made different folders for all of the porn,” that “sometimes I spent the whole day looking at porn, I had like half a million pictures saved on my computer,” and that “it was all consuming.”

Pornography Progression

How did participants reach the point of excessive collecting or being consumed by CSEM usage? A possible answer lies in a common story for both those who cited collecting as a motivation and those who did not: narratives of offending trajectories often started with looking at “normal” pornography, then sliding down a slippery slope in depravity (e.g., pictures of supermodels/celebrities, to hard-core pornography, to sadomasochism, to CSEM):

…looking at celebrities and stuff and I always liked the sort of the thin, or waif type women…I don’t know you just start looking at things like, somebody give me a website name which was all legal still, but then [inaudible] teen in such and such and you'd look at it, and after a while you’re thinking “that’s not a proper teen.” (Study 1, P30)

I was watching just regular straight porn. It developed over time. I started watching the teen stuff. I started looking, going into chat groups. And then I started with pre-pubescent girls. (Study 2, P45)

An essential aspect of this narrative was the accessibility of pornography online. Ease of access provided opportunity to view all types of pornography, which one participant referred to as “the holy grail, it’s your birthday” (Study 1, P12). In this sense, the Internet provided a boundless array of both legal pornography and CSEM, which some participants described in terms relating to progression of time spent engaging with the material:

…instead of just collecting a magazine and using it for the odd night or something then throwing it away and not doing anything for three months, you, you know you get a dozen videos a day for a whole year or something. (Study 1, P62)

In my late 20s and 30s I started watching it online and then it was nine to 10 hours a day, seven days a week. It was a full-time thing. I watched literally everything. There is no porn I haven’t seen. (Study 2, P56)

How did participants describe progression starting and then continuing? For some, after discovering CSEM, they felt excitement in doing something taboo, forbidden, or illegal:

…it was a buzz of, breaking a taboo…then, the next time you do it maybe the thing of, “I’ve done it, the police haven’t been to my door. I got away with that.” And that adds to the, the buzz of doing it. (Study 1, P41)

Porn got boring so I had to look for more taboo stuff. Normal stuff was whatever. I needed over the edge. I had seen most categories. So I sought out child porn. (Study 2, P30)

For others, progression started when they used pornography to alleviate boredom:

I was bored, I was lazy. It was an easy way to stimulate my mind. Um, easy way to stimulate me sexually as well. (Study 1, P24)

I was lonely and bored. And then the viewing porn led to other things. (Study 2, P56)

Still with others, progression was more entrenched in their pasts. Some had a long-standing relationship with pornography, which then grew online. Sometimes this also meant they had wide-ranging pornographic interests, corresponding to the heterogeneous nature of some CSEM user collections (Elliott & Beech, 2009; Krone, 2004):

I guess I’ve always had a, an interest in pornography, I suppose. And, or masturbation. Um, and certainly having the Internet facilitated that. So I think I, I went a bit, raaaaa, crazy. (Study 1, P63)

Porn desensitized me early in life. Early on it was a learning tool. A how-to. Later I felt maybe my brain is becoming passive and porn is normal. (Study 2, P63)

A sexual element is inextricably tied to these kinds of narratives, along with the notion of progression more generally, through masturbation and arousal (Seto, 2013). Progression was often accompanied by a perceived process of desensitization, which has been similarly described by scholars in empirical research (Quayle & Taylor, 2003) and reviews (Seto, 2013; Wortley, 2012), with participants claiming there was both a reduction in emotional response and a “need” to see increased depravity to create arousal:

I got to the point where I was seeing so much porn, it became completely meaningless. (Study 1, P33)

Because I’d been into pornography in the first place, that made it, easier to go against something that I knew was wrong. (Study 1, P41)

Anything not in moderation desensitizes you. Empathy was something I didn’t have. (Study 2, P82)

I started out with vanilla and then got into worse stuff. (Study 2, P30)

Thus, as discussed so far, some participants used CSEM to establish a sense of control or power, and many described a narrative of progression. However, progression reached a peak, and control was lost, when participants claimed to become “addicted,” a finding also echoed by Quayle and Taylor (2002) in their smaller-scale research. Participants’ appeal to this idea and term is problematic for both agency and risk: “addiction” insinuates a condition out of one’s control therefore impacting notions of responsibility and choice, yet at the same time, being “addicted” suggests such a high level of involvement that risk could be high (e.g., to offend offline or reoffend online). These ideas are explored further below.

From Progression to “Addiction”

Framing excessive Internet and pornography use as “addiction” is a contested scholarly topic, and concepts such as “compulsive sexual behavior” and “hypersexuality” are now also employed by some clinicians and researchers to describe these behaviors. However, focusing on participant framings, many attributed their continued offending at least partly to an “addiction,” with 22 interviewees in Study 1 and 15 in Study 2 alluding to or citing it:

I don't care what anybody says, it is an addiction. Once you’ve had the fix, you say, “oooh,” you know, you’re shocked and it’s like, you know, once you have your first cigarette you’ve coughed and you’re a bit sick…But then, that doesn't say you’re never gonna try it again. So you have your second cigarette, and you cough and you feel a bit sick, but not as bad as last time. And, it’s the same with the Internet… (Study 1, P46)

I mean since I was 14, I’ve been addicted to looking at pornography in general. Um, and then, say as time’s gone on it’s gone to more depraved. (Study 1, P16)

Porn led to all of my problems. It’s definitely an addiction. (Study 2, P56)

I was addicted to porn…It is a battle I will have to continue to fight. (Study 2, P79)

Moreover, beyond simply referring to “addiction,” participants also aligned their behavior to multiple additional features common to other addictions.

Participants’ Indicators of “Addiction”

Addictions have certain characteristics, as noted in the Introduction, and participants described these in recounting their offending. Some discussed lacking/losing control, and how they felt that they could not stop, apparently losing their agency to the “addiction”:

I wanted to be in control of it. And I wasn’t. It was in control of me. (Study 1, P21)

I didn’t want to be what I did, and I didn’t want to have the images I saw…So I deleted everything. But, at a certain stage, could be a few weeks could be a few months, I couldn’t resist, and went online again. (Study 1, P37)

It was just in my 20s seeing those pictures. It opened up the floodgates and I couldn’t shut it off. (Study 2, P14)

Coupled with lack of control, like with other addictions, was recognizing and ignoring potential negative and detrimental consequences while still engaging in the actions:

I would talk to myself, “you could go to jail. Look at all the consequences, for your family, for yourself, for your life.” At the end of the day I was just, “well I don’t care. I’ll take the consequences, I’ll risk the consequences,” cause that fix, was what I wanted. (Study 1, P51)

I knew I’d lose my job, I knew I’d lose my home, I knew I’d lose my car, I knew I’d lose my friends. You know, I knew I’d probably go to prison…So why would I, who want a peaceful quiet life, doesn’t want to get into trouble with anybody, wants to carry, wants to enjoy their job, wants to carry on living a sort of peaceful happy life, why would I do this…I can’t see why I would logically do that unless I felt that kind of drive. And, and I wanted to stop, I wanted to stop [hits table], so many times!…I felt like the scum of the Earth…I think that’s addiction. (Study 1, P43)

It was really bad. I would stay up all night watching it. And then I would skip work. (Study 2, P82)

I was online at least 12 hours a day. I wasn’t eating or sleeping. My activity deteriorated. My social and home life deteriorated. I isolated myself. (Study 2, P10)

On a biological level, some participants also interpreted physiological symptoms of their “addiction” to be similar to other dependencies, describing a form of withdrawal:

…it would be like having to have a fix of some kind of drug, that’s the only way I can explain it. This need. My heart’d be thumping, I’d have this dryness in my mouth, and I would just, I would just get on the computer and I’d, and I’d have to do it. (Study 1, P43)

I would literally get sick at work because I missed porn. I missed watching it. (Study 2, P38)

Important here is likening to substances. This analogy has become common with clinicians, with some referring to Internet pornography “as the ‘crack cocaine’ of sex addiction” (Griffin-Shelley, 2014, p. 323). For participants, likening to substances could have been an attempt to address criticisms of metaphor described earlier by either explicitly equating CSEM to drugs/alcohol, or implicitly comparing it by using words such as “fix” and “high”:

It was like, how many times did you manage to avoid clicking that link and resist that temptation until, you know, one day out of after maybe after 100 you suddenly went bang!.…it’s like getting your first hit of heroin, or crack cocaine or something, you know, off you go. (Study 1, P36)

I think it’s like a recovering alcoholic, you have to wake up every day and say it’s one more day. (Study 1, P45)

I was happy when I got my fix. I was excited to watch it. It was addiction. I didn’t need to be sexually gratified. Looking at it. It was ritualistic, and part of the whole high. (Study 2, P38)

Aligning with narratives of progression, as well as likening to substances, those citing “addiction” often described movement from “softer” to “harder” material. For example, over his cycle of offending, P17 in Study 1 found it to be like “needing more and more, harder and harder drugs.” Such an analogy to increasing tolerance was common:

[Pornography magazine] was a gateway drug for me because it, it started the appetite. But then it wouldn’t satisfy anymore, because it was too soft…I had to travel much, deeper and darker to get the same rush. (Study 1, P43)

I described it, when you first drink you have one beer and you’re pissed. And then, a couple of years down the line you might need 10 beers…So, it just becomes an even more, sort of risqué, to get that same, call it arousal, shock, shame, whatever you know, all these things that, um, I couldn’t get from, from normal pornography. (Study 1, P75)

The stuff I watched got nastier and nastier. And my time online increased. (Study 2, P54)

This process was said to change some from who they “really were” to “addicts”:

…yes I suppose it [empathy] must have been there but, but you didn’t care, cause you wanted that fix…you’re, that drug addict that needs the money to get that drug you need and you go out and, rob an old lady in the street. And if you weren’t on drugs you wouldn’t do it. (Study 1, P65)

This point is important, as appeals to addiction have consequences: an addiction model has potential to insinuate control is lost because of the Internet as a medium, which either exacerbates existent pathologies or facilitates new ones. This could present a CSEM user as having limited personal agency, choice, and responsibility (Quayle & Taylor, 2002).

Rejecting or Resisting the Addiction Construct

The above points about agency, choice, and responsibility were sometimes discussed by participants. While less common, some noted control lies with the individual, and one does not want to abdicate responsibility by calling offending an “addiction”:

It’s attractive to maybe say to yourself that it’s uncontrollable. (Study 1, P81)

The only thing that worries me by labeling it as an addiction is that it takes away part of the blame, which makes me uncomfortable. (Study 1, P63)

This has a whole empathy bit that doesn’t exist with drugs, which are just substances. (Study 1, P8)

I don’t like the word addiction, because that implies a chemical dependency. I would say it’s a compulsive behavior. I have been reading in here about neural pathways and how they develop. And you can develop new neural pathways and retrain your brain. So that is what I am trying to do. (Study 2, P30)

These participants suggested that “addiction” could rather be “an ego-syntonic choice to engage in behaviours that provides stimulation and excitement in the absence of any resistance or attempt to cease” (Blaszczynski, 2006, p. 9). As they point out, “addiction” might be a way for these individuals to minimize self-blame for breaking norms and laws, as “addictions” may be more forgivable than deviant sexual interests. This is a vital part of risk assessment according to some scholars. Winder et al. (2015) argue that risk management must involve having CSEM users recognize that their actions are not addictions. If appeals to addiction are “allowed to go unchecked,” they suggest “this will inhibit the offenders understanding and accepting their deviant preferences and risk, potentially exacerbating their likelihood of committing contact offences, if such opportunities arise” (Winder et al., 2015, p. 178). We do not aim to say for certain that participants were “addicted” or not. Most feasible is that some did feel that they were, while others were using this term in an attempt to relinquish control and create a more acceptable narrative, ideas which are explored below.

Discussion: CSEM Offending and “Addiction” in a Broader Context

Across conviction status, countries, settings, programs, timeframes, and methodologies, we found commonalities in the stories of some CSEM users as related to “addiction.” These narratives were similarly structured; they described CSEM use as an “addiction,” or as part of an “addiction” (e.g., to pornography) that included CSEM usage, with offending attributed at least partly to “addiction.” According to these participants, this was marked by an initiation into and subsequent spiraling progression, and with features akin to other addictions. The narratives employed by participants served important functions, which were to make sense of and create coherence across experiences, to give meaning to their lives, and to better understand the world around them and their place within it (McAdams, 2001). Using a larger and more diverse sample, these findings build on the work of Quayle and Taylor (2002), who similarly found “addiction” to be one of six discourses of offending among 13 CSEM users.

Central to participant narratives are accounts, or linguistic devices that individuals employ to explain actions outside the boundaries of normative conduct, used to describe past behavior and inform future actions (Jacobs & Copes, 2015; Mills, 1940; Scott & Lyman, 1968). A given account is important in meaning making because of “its ability to throw bridges between the promised and the performed, its ability to repair the broken and restore the estranged” (Scott & Lyman, 1968, p. 46). Crimes against children are among the most socially detested and punished. Society draws boundaries between “normal,” prosocial individuals and those who engage in criminal sexual behavior, unable to hold these two concepts simultaneously (Marshall, 1996). This results in an othering and labeling of individuals who commit sexual offenses as “monsters,” who are deemed threats to the moral order of society (Waldram, 2009). Thus, for CSEM users, accounts can be crucial to the management of spoiled identity (Goffman, 1963) and shame (Braithwaite & Braithwaite, 2001). We suggest that for participants, an addiction framework served to reframe their transgressive behaviors as more understandable and forgivable to themselves when recounted as something said to be out of one’s control.

Yet narratives are not employed by participants only to explain behavior, but also to construct identities, explore crucial events and issues, and understand themselves and the world (Sandberg, 2016). Initiation to “addiction” served an important function in participant stories, as it identified a “cause” or catalyst that could explain and make sense of events that followed. Temporality and causality are components that imbue narratives with meaning (Sandberg, 2016). For some, this included the need to exercise control or power in a life that felt chaotic, meaningless, or overwhelming. Thus, CSEM became a way to “escape,” avoid, or manage negative emotions and events (Laulik et al., 2007; Marshall et al., 2012; Middleton et al., 2006; Quayle et al., 2006). The descriptions of pornography progression, framed as culminating in a loss of control, were consistent with narratives of addiction to substances. These accounts also draw from larger cultural narratives of pornography addiction as biological and with possibly devastating consequences (Taylor & Gavey, 2019).

On the whole, debates surrounding whether problematic Internet, pornography, or Internet pornography use can be labeled addiction are contentious, and concepts such as “hypersexuality” and “compulsive sexual behavior” have recently been employed by clinicians and academics to describe some of these behaviors. As Taylor (2019, p. 57) asserts, “Pornography addiction now stands as an intelligible subject circulating across popular culture, despite also being a contested category in clinical, academic, and legislative fields.” However, in emphasizing how participants conceptualize their behavior, it is clear that the “addiction” construct remains salient and foregrounded for them. The question that then follows is, what are the implications of defining CSEM offending through an “addiction” framework? We offer two perspectives below, both of which may be applicable but to different CSEM users, with each perspective having distinct potential outcomes.

First, framing CSEM use through an addiction lens has several problematic consequences. The biologizing of addiction has been argued to create a sentiment that absolves the individual of a sense of responsibility, potentially shaping future behavior (Hacking, 1996; Taylor, 2019). Clinically, there is no agreement about whether pornography addiction or Internet addiction are disorders (Kraus et al., 2016; Schell, 2020). Moreover, an appeal to addiction can also convey the notion of limited personal agency, choice, and responsibility (Quayle & Taylor, 2002), suggesting that offending is out of one’s control or unchangeable. Taken to an extreme, CSEM users could then conceivably reach various harmful and counterproductive conclusions: that there is no point trying to stop or to strive for change; that stopping is insurmountable; that reoffending is inevitable; that they are not accountable for harms caused by CSEM; or that progression to contact offending is likely. Therefore, some scholars argue that CSEM users must be discouraged from conceptualizing offending as an addiction, as it could have consequences for their risk (Winder et al., 2015).

These points are salient, given that some participants resisted the “addiction” label because of responsibility and harms that they claimed to now recognize (for a paper specifically focused on victim empathy with Study 1 participants see Rimer, 2019). Children must be exploited to create CSEM. This exploitation can have highly damaging effects above and beyond sexual abuse without recording: in particular, children (and subsequently adults) can be impacted by the permanency of CSEM, and the ongoing invasion of privacy, powerlessness, and victimization with knowledge that countless unknown people will view their images (Gewirtz-Meydan et al., 2018; Martin, 2015; Palmer, 2005). CSEM is not localized to where abuse took place, and thus, trauma continues with victims unable to have closure or identify an end (Martin, 2015). Palmer (2005) suggests that victims can then have low self-esteem, repressed anger, inability to trust, lack of perceived control, and “damaged goods syndrome.” For these reasons, we argue that the harms caused by CSEM cannot be ignored or explained away by appeals to addiction. Subsequently, extra considerations around harm are necessary when defining this behavior through an addiction framework or not, which may not be as relevant or present with other behaviors (e.g., substance use).

From a different perspective, despite academic and clinical controversy, the concept of addiction seemed to serve a function for some participants. While it is possible that some used the label to abdicate responsibility, it is also possible that some used it to identify actions and areas for change. This is particularly apt because participants identified a progression to CSEM, acknowledging that they were not always engaging with it. It therefore stands to reason that they could recognize the possibility of returning to a pre-CSEM existence. Their narratives could situate illegal behavior and make sense of past criminal selves, but they could also be transformative and create paths to desistance (McAdams, 2001; Maruna, 2001). In this way, regardless of whether pornography or CSEM use meets the criteria for addiction, it may nevertheless be a framework or purposive metaphor with which these individuals make sense of their past behavior and manage their future actions (Briggs et al., 2017). By creating accounts, some may have been making sense of their experiences by crafting a powerful force that could be understood as causal, and therefore possibly manageable and controllable (as opposed to unmanageable and uncontrollable): as one stated, “you don’t know you have a choice until you realize you have a choice” (Study 1, P52).

Following this line of thinking, to achieve behavioral change, it is necessary to find ways to motivate individuals, to channel this motivation, and to sustain change while avoiding relapse (Webb et al., 2010). Perhaps for some participants, internal and external controls on individual behavior can be strengthened using an addiction framework. To an extent, sexual offending treatment already does this through the common practice of relapse prevention planning, which is borrowed from addictions therapy (Webster, 2005) and involves attempting to alleviate triggers for offending by identifying hypothetical but possible risky moods, emotions, and situations that once recognized can be responded to, changed, or adjusted (Seto, 2013; Waldram, 2012). Participants could potentially learn to manage their behaviors and identify triggers that could lead to CSEM use. Through awareness of the emotional states and negative life events that are associated with offending, they could utilize external social controls such as treatment providers, religious or community institutions, friends, or family. Thus, framing CSEM use this way may be a conduit for cognitive behavioral change and increased social supports. What is key is the ability for individuals to recognize when internal controls are loosening or weakening, and then to seek out formal controls to monitor and shape/redirect behavior, thus not using “addiction” as a way to limit responsibility but rather the opposite: as a mechanism of knowledge, prevention, and turning points (Sandberg, 2016). Narratives are not one-dimensional stories; they have the power to transform and structure action, as individuals can become the stories they tell (Bruner, 1990).

Child sexual exploitation material and addiction are complicated topics fraught with clinical, social, and personal consequences for both those who are victimized and those who commit offenses. We cannot resolve definitional debates; rather, we aim to present multiple and sometimes incongruous viewpoints to consider for researching and framing this behavior, toward the ultimate goal of reduced offending and victimization.

Limitations

This paper’s methodology has several limitations. We combined results, and there were differences between samples. Study 1 was in a community agency where individuals were seeking help and were living in society awaiting trial. Study 2 took place in prison, which resulted in a setting to collect data and a sample that was qualitatively different. The samples were in two countries, which may reflect cultural differences in sexual offending behaviors and treatment. Yet, despite these differences, we found comparable “addiction” framings. Second, the samples were self-selected, which may skew the findings. Third, we did not have access to treatment or police files, and it may be that participants were not candid or were influenced by social desirability. However, the purpose was to understand how they framed their experiences. Narratives are inherently subjective, but they offer rich data on meaning making and the self (Presser & Sandberg, 2015). Fourth, we cannot know if addiction framing occurred naturally or as a result of programming effects. However, regardless of how framing occurs, it is a potentially powerful mechanism warranting exploration. Fifth, due to data access restrictions, each dataset was coded by one researcher for their respective study. As such, intercoder reliability could not be measured (although this is often not the goal of thematic analysis), and it is possible that inconsistencies occurred. However, as noted in the Method section, to limit inconsistencies the researchers underwent a systematic and iterative process that strengthened communication, collaboration, and agreement. Sixth, participants’ use of the term “addiction” was fluid. It was not always clear if they were referring specifically to CSEM offending as an “addiction,” or to its place within a broader pornography “addiction,” or as part of an “addiction” to the Internet, or a combination. Their statements and quotes do not always differentiate these, and therefore, we are not always able to disentangle such nuances. However, regardless of the exact specifics for each person, the concept of “addiction” was an important and consistent explanatory element in participant narratives, and thus a worthy line of analysis. Future research could aim to specifically differentiate the kinds of “addiction” to which CSEM users may refer, and to determine the place of CSEM within each type of offending narrative. Lastly, all participants were male, and most were white, which is consistent with the extant literature. Studies of CSEM users have focused on samples from North America, Europe, Australia, and New Zealand, which has led to selection bias (Seto, 2013). Lack of diversity in sampling limits the scope of knowledge regarding CSEM use among other populations such as minorities, women, and those in non-Western countries. The experiences, perceptions, and narratives of these users are critical to a comprehensive understanding of CSEM offending.

Conclusion

Using results from two studies, this paper explored notions of “addiction” among individuals who committed CSEM offenses. Labeling offending through the concept of “addiction” allowed for accounts that situated these behaviors into a knowable framework. Participants described their initiation into “addiction” and the spiraling progression that followed as they claimed to lose control of behavior. Such accounts may be used in an attempt to absolve responsibility and blame, or may provide mechanisms by which an individual can engage in effective stigma management and self-regulation, positioning oneself in a way that may be conducive to inspiring and sustaining behavioral change. Future research should continue to examine the accounts of individuals who have engaged in CSEM offending, especially through qualitative inquiry. Interpretations of their own behavior by those who engage in offenses is one of the most understudied aspects of criminality (Jacobs & Copes, 2015). Understanding and analyzing the narratives of these individuals can provide insight into treatment, supervision, and prevention of CSEM offending, as well as child protection more generally.

Footnotes

Acknowledgments

Thank you to the anonymous reviewers and Associate Editor for their helpful feedback and openness to interdisciplinary research. Thank you to all participants for taking part in the studies. Thank you also to Sarah-Ann Burger and Pearl Rimer for commenting on previous versions of this paper. The authors take responsibility for the integrity of the data, the accuracy of the data analyses, and have made every effort to avoid inflating results.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Study 1 was supported by the Commonwealth Scholarship Commission in the UK [Commonwealth Scholarship to J.R.R], the Social Sciences and Humanities Research Council of Canada [Doctoral Fellowship to J.R.R], and the Royal Anthropological Institute [Sutasoma Award to J.R.R]. Study 2 was supported by the Michigan State University School of Criminal Justice [Start-Up Research Funds to K.H.]. Funders were not involved in the research or manuscript preparation.

ORCID iDs

Jonah R. Rimer

Karen Holt

References

American Psychiatric Association . (2013). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Association.

Babchishin

K. M.

Hanson

R. K.

Hermann

C. A.

(2011). The characteristics of online sex offenders: A meta-analysis. Sexual Abuse, 23(1), 92–123. https://doi.org/10.1177/1079063210370708.

Billieux

Schimmenti

Khazaal

Maurage

Heeren

(2015). Are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research. Journal of Behavioral Addictions, 4(3), 119–123. https://dx-doi-org-s.web.bisu.edu.cn/10.1556/2006.4.2015.009.

Blaszczynski

(2006). Internet use: In search of an addiction. International Journal of Mental Health and Addiction, 4(1), 7–9. https://doi.org/10.1007/s11469-006-9002-3.

Braithwaite

(2001). Part IV: Creating institutional spaces for shame management. In Ahmed

Harris

Braithwaite

(Eds.), Shame management through reintegration (pp. 315–331). Cambridge University Press.

Braun

Clarke

(2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa.

Braun

Clarke

(2019). Reflecting on reflexive thematic analysis. Qualitative Research in Sport, Exercise and Health, 11(4), 589–597. https://doi.org/10.1080/21596.76X.2019.1628806.

Briggs

Gough

das Nair

(2017). Losing control in sex addiction: “Addict” and “non-addict” accounts. Sexual and Relationship Therapy, 32(2), 195–209. https://doi.org/10.1080/14681994.2016.1276551.

Bruner

(1990). Acts of meaning. Harvard University Press.

10.

Campbell

J. L.

Quincy

Osserman

Pedersen

O. K.

(2013). Coding in-depth semistructured interviews: Problems of unitization and intercoder reliability and agreement. Sociological Methods & Research, 42(3), 294–320. https://doi.org/10.1177/0049124113500475.

11.

Carnes

(1983). Out of the shadows: Understanding sexual addiction. CompCare.

12.

Carnes

(1989). Contrary to love: Helping the sexual addict. CompCare.

13.

Carnes

P. J.

Delmonico

D. L.

Griffin

(2007). In the shadows of the net: Breaking free of compulsive online sexual behavior. Hazelden Publishing.

14.

Cash

Rae

C. D.

Steel

A. H.

Winkler

(2012). Internet addiction: A brief summary of research and practice. Current Psychiatry Reviews, 8(4), 292–298. https://doi.org/10.2174/157340012803520513.

15.

Cooper

Delmonico

D. L.

Griffin-Shelley

Mathy

R. M.

(2004). Online sexual activity: An examination of potentially problematic behaviours. Sexual Addiction and Compulsivity, 11(3), 129–143. https://doi.org/10.1080/10720160490882642.

16.

Diamond

(2009). Pornography, public acceptance and sex related crime: A review. International Journal of Law and Psychiatry, 32(5), 304–314. https://doi.org/10.1016/j.ijlp.2009.06.004.

17.

Duffy

Dawson

D. L.

das Nair

(2016). Pornography addiction in adults: A systematic review of definitions and reported impact. Journal of Sexual Medicine, 13(5), 760–777. https://doi.org/10.1016/j.jsxm.2016.03.002.

18.

Dunn

Seaburne-May

Gatter

(2012). Internet sex addiction: A licence to lust? Advances in Psychiatric Treatment, 18(4), 270–277. https://doi.org/10.1192/apt.bp.109.007492.

19.

Elliott

I. A.

Beech

A. R.

(2009). Understanding online child pornography use: Applying sexual offense theory to Internet offenders. Aggression and Violent Behavior, 14(3), 180–193. https://doi.org/10.1016/j.avb.2009.03.002.

20.

Garrison

D. R.

Cleveland-Innes

Koole

Kappelman

(2006). Revisiting methodological issues in transcript analysis: Negotiated coding and reliability. The Internet and Higher Education, 9(1), 1–8. https://doi.org/10.1016/j.iheduc.2005.11.001.

21.

Gewirtz-Meydan

Walsh

Wolak

Finkelhor

(2018). The complex experience of child pornography survivors. Child Abuse & Neglect, 80, 238-248. https://doi.org/10.1016/j.chiabu.2018.03.031.

22.

Goffman

(1963). Stigma: Notes on the management of spoiled identity. Prentice-Hall.

23.

Griffin-Shelley

(2014). Sex and love addicts, who sexually offend: Two cases of online use of child pornography (child sexual abuse images). Sexual Addiction & Compulsivity, 21(4), 322–341. https://doi.org/10.1080/10720162.2014.966936.

24.

Griffiths

M. D.

(2012). Internet sex addiction: A review of empirical research. Addiction Research and Theory, 20(2), 111–124. https://doi.org/10.3109/16066359.2011.588351.

25.

Hacking

(1996). The looping effects of human kinds. In Sperber

Premack

A. J.

(Eds.), Causal cognition: A multidisciplinary debate (pp. 351–383). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198524021.003.0012.

26.

Henshaw

Ogloff

J. R. P.

Clough

J. A.

(2017). Looking beyond the screen: A critical review of the literature on the online child pornography offender. Sexual Abuse, 29(5), 416–445. https://doi.org/10.1177/1079063215603690.

27.

Holt

Kissinger

Spickler

Roush

(2021). Pornography use and sexual offending: An examination of perceptions of role and risk. International Journal of Offender Therapy and Comparative Criminology. Advance online publication. https://doi.org/10.1177/0306624X211049183

28.

Jacobs

B. A.

Copes

(2015). Neutralization without drift: Criminal commitment among persistent offenders. British Journal of Criminology, 55(2), 286–302. https://doi.org/10.1093/bjc/azu100.

29.

Kafka

M. P.

(2001). The paraphilia-related disorders: A proposal for a unified classification of nonparaphilic hypersexual disorders. Sexual Addiction & Compulsivity, 8(3–4), 227–239. https://doi.org/10.1080/107201601753459937.

30.

Kafka

M. P.

(2010). Hypersexual disorder: A proposed diagnosis for DSM-V. Archives of Sexual Behavior, 39(2), 377–400. https://doi.org/10.1007/s10508-009-9574-7.

31.

Kirmayer

L. J.

Raikhel

Rahimi

(2013). Cultures of the internet: Identity, community and mental health. Transcultural Psychiatry, 50(2), 165–191. https://doi.org/10.1177/1363461513490626.

32.

Kraus

Rosenberg

(2014). The pornography craving questionnaire: Psychometric properties. Archives of Sexual Behavior, 43(3), 451–462. https://doi.org/10.1007/s10508-013-0229-3.

33.

Kraus

S. W.

Voon

Potenza

M. N.

(2016). Should compulsive sexual behavior be considered an addiction? Addiction, 111(12), 2097–2106. https://doi.org/10.1111/add.13297.

34.

Krone

(2004). A typology of online child pornography offending. Trends and Issues in Crime and Criminal Justice, 279, 1-6. https://www.aic.gov.au/publications/tandi/tandi279.

35.

Laulik

Allam

Sheridan

(2007). An investigation into maladaptive personality functioning in internet sex offenders. Psychology, Crime & Law, 13(5), 523–535. https://doi.org/10.1080/10683160701340577.

36.

Marshall

W. L.

(1996). The sexual offender: Monster, victim, or everyman? Sexual Abuse, 8(4), 317–335. https://doi.org/10.1177/107906329600800406.

37.

Marshall

L. E.

O’Brien

M. D.

Marshall

W. L.

Booth

Davis

(2012). Obsessive compulsive disorder, social phobia, and loneliness in incarcerated internet child pornography offenders. Sexual Addiction & Compulsivity, 19(1–2), 41–52. https://doi.org/10.1080/10720162.2012.665291.

38.

Martin

(2015). Conceptualizing the harms done to children made the subjects of sexual abuse images online. Child & Youth Services, 36(4), 267–287. https://doi.org/10.1080/0145935X.2015.1092832.

39.

Maruna

(2001). Making good: How ex-convicts reform and rebuild their lives. American Psychological Association.

40.

McAdams

D. P.

(2001). The psychology of life stories. Review of General Psychology, 5(2), 100–122. https://doi.org/10.1037/1089-2680.5.2.100.

41.

Middleton

Elliott

I. A.

Mandeville-Norden

Beech

A. R.

(2006). An investigation into the applicability of the Ward and Siegert pathways model of child sexual abuse with internet offenders. Psychology, Crime & Law, 12(6), 589-603. https://doi.org/10.1080/10683160600558352.

42.

Mills

C. W.

(1940). Situated actions and vocabularies of motive. American Sociological Review, 5(6), 904–913. https://doi.org/10.2307/2084524.

43.

National Center for Missing and Exploited Children . (2022). Child sexual abuse material (csam). National Center for Missing and Exploited Children. https://www.missingkids.org/theissues/csam.

44.

Neutze

Grundmann

Scherner

Beier

K. M.

(2012). Undetected and detected child sexual abuse and child pornography offenders. International Journal of Law and Psychiatry, 35(3), 168–175. https://doi.org/10.1016/j.ijlp.2012.02.004.

45.

Nowell

L. S.

Norris

J. M.

White

D. E.

Moules

N. J.

(2017). Thematic analysis: Striving to meet the trustworthiness criteria. International Journal of Qualitative Methods, 16(1), 1–13. https://doi.org/10.1177/1609406917733847.

46.

O’Connor

Joffe

(2020). Intercoder reliability in qualitative research: Debates and practical guidelines. International Journal of Qualitative Methods, 19, 1-13. https://doi.org/10.1177/1609406919899220.

47.

Palmer

(2005). Behind the screen – children who are the subjects of abusive images. In Quayle

Taylor

(Eds.), Viewing child pornography on the internet: Understanding the offence, managing the offender, helping the victims (pp. 61–74). Russell House.

48.

Pies

(2009). Should DSM-V designate “internet addiction” a mental disorder? Psychiatry, 6(2), 31–37. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2719452/.

49.

Presser

Sandberg

(Eds.). (2015). Narrative criminology: Understanding stories of crime. New York University Press.

50.

Quayle

Taylor

(2002). Child pornography and the internet: Perpetuating a cycle of abuse. Deviant Behavior, 23(4), 331–361. https://doi.org/10.1080/01639620290086413.

51.

Quayle

Taylor

(2003). Model of problematic internet use in people with a sexual interest in children. CyberPsychology & Behavior, 6(1), 93–106. https://doi.org/10.1089/109493103321168009.

52.

Quayle

Vaughan

Taylor

(2006). Sex offenders, internet child abuse images and emotional avoidance: The importance of values. Aggression and Violent Behavior, 11(1), 1–11. https://doi.org/10.1016/j.avb.2005.02.005.

53.

Reid

R. C.

Garos

Carpenter

B. N.

(2011). Reliability, validity, and psychometric development of the Hypersexual Behavior Inventory in an outpatient sample of men. Sexual Addiction & Compulsivity, 18(1), 30–51. https://doi.org/10.1080/10720162.2011.555709.

54.

Reid

R. C.

D. S.

Gilliland

Stein

J. A.

Fong

(2011). Reliability, validity, and psychometric development of the Pornography Consumption Inventory in a sample of hypersexual men. Journal of Sex and Marital Therapy, 37(5), 359–385. https://doi.org/10.1080/0092623X.2011.607047.

55.

Reinarman

(2005). Addiction as accomplishment: The discursive construction of disease. Addiction Research & Theory, 13(4), 307–320. https://doi.org/10.1080/16066350500077728.

56.

Rimer

J. R.

(2017). Internet sexual offending from an anthropological perspective: Analysing offender perceptions of online spaces. Journal of Sexual Aggression, 23(1), 33–45. https://doi.org/10.1080/13552600.2016.1201158.

57.

Rimer

J. R.

(2019). “In the street they’re real, in a picture they’re not”: Constructions of children and childhood among users of online child sexual exploitation material. Child Abuse & Neglect, 90, 160–173. https://doi.org/10.1016/j.chiabu.2018.12.008.

58.

Rimer

J. R.

(2021). Discipline as prevention: Psychoeducational strategies in internet sexual offending group programs. International Journal of Offender Therapy and Comparative Criminology, 65(15), 1607–1628. https://doi.org/10.1177/0306624X20952389.

59.

Sandberg

(2016). The importance of stories untold: Life-story, event-story and trope. Crime, Media, Culture, 12(2), 153–171. https://doi.org/10.1177/1741659016639355.

60.

Schell

(2020). Internet addiction and cybercrime. In Holt

Bossler

A. M.

(Eds.), The Palgrave handbook of international cybercrime and cyberdeviance (pp. 679–703). Palgrave Macmillan. https://doi.org/10.1007/978-3-319-90307-1_26-1.

61.

Scott

M. B.

Lyman

S. M.

(1968). Accounts. American Sociological Review, 33(1), 46–62. https://doi.org/10.2307/2092239.

62.

Seto

M. C.

(2013). Internet sex offenders. American Psychological Association. https://doi.org/10.1037/14191-000.

63.

Shapira

N. A.

Lessig

M. C.

Goldsmith

T. D.

Szabo

S. T.

Lazoritz

Gold

M. S.

Stein

D. J.

(2003). Problematic internet use: Proposed classification and diagnostic criteria. Depression and Anxiety, 17(4), 207–216. https://doi.org/10.1002/da.10094.

64.

Steel

C. M. S.

(2009). Child pornography in peer-to-peer networks. Child Abuse & Neglect, 33(8), 560–568. https://doi.org/10.1016/j.chiabu.2008.12.011.

65.

Taylor

(2019). Pornography addiction: The fabrication of a transient sexual disease. History of the Human Sciences, 32(5), 56–83. https://doi.org/10.1177/0952695119854624.

66.

Taylor

Gavey

(2019). Pornography addiction and the perimeters of acceptable pornography viewing. Sexualities, 23(5–6), 876–897. https://doi.org/10.1177/1363460719861826.

67.

Taylor

Quayle

(2003). Child pornography: An internet crime. Brunner-Routledge.

68.

Waldram

J. B.

(2009). “It’s just you and Satan, hanging out at a pre-school:” Notions of evil and the rehabilitation of sexual offenders. Anthropology and Humanism, 34(2), 219–234. https://doi.org/10.1111/j.1548-1409.2009.01039.x.

69.

Waldram

J. B.

(2012). Hound pound narrative: Sexual offender habilitation and the anthropology of therapeutic intervention. University of California Press. https://doi.org/10.1525/j.ctt1pp5zw.

70.

Walton

M. T.

Cantor

J. M.

Bhullar

Lykins

A. D.

(2017). Response to commentaries: Recognizing hypersexuality as a psychosexual behavioral problem and advancing the sexhavior cycle of hypersexuality. Archives of Sexual Behavior, 46(8), 2279–2287. https://doi.org/10.1007/s10508-017-1111-5.

71.

Webb

T. L.

Sniehotta

F. F.

Michie

(2010). Using theories of behaviour change to inform interventions for addictive behaviours. Addiction, 105(11), 1879–1892. https://doi.org/10.1111/j.1360-0443.2010.03028.x.

72.

Webster

S. D.

(2005). Pathways to sexual offense recidivism following treatment: An examination of the Ward and Hudson Self-Regulation Model of relapse. Journal of Interpersonal Violence, 20(10), 1175–1196. https://doi.org/10.1177/0886260505278532.

73.

Widyanto

Griffiths

(2007). Internet addiction: Does it really exist? (Revisited). In Gackenbach

(Ed.), Psychology and the Internet: Intrapersonal, interpersonal, and transpersonal implications (2nd ed., pp. 141–163). Academic Press.

74.

Wilson

(2017). Your brain on porn: Internet pornography and the emerging science of addiction. Commonwealth Publishing.

75.

Winder

Gough

Seymour-Smith

(2015). Stumbling into sexual crime: The passive perpetrator in accounts by male internet sex offenders. Archives of Sexual Behavior, 44(1), 167–180. https://doi.org/10.1007/s10508-014-0302-6.

76.

Wolak

Liberatore

Levine

B. N.

(2014). Measuring a year of child pornography trafficking by U.S. computers on a peer-to-peer network. Child Abuse & Neglect, 38(2), 347–356. https://doi.org/10.1016/j.chiabu.2013.10.018.

77.

World Health Organization . (2020). ICD-11 for mortality and morbidity statistics. https://icd.who.int/browse11/l-m/en#/http%3a%2f%2fid.who.int%2ficd%2fentity%2f1630268048.

78.

Wortley

(2012). Situational prevention of child abuse in the new technologies. In Quayle

Ribisl

K. M.

(Eds.), Understanding and preventing online sexual exploitation of children (pp. 188–203). Routledge.

79.

Young

K. S.

(2008). Internet sex addiction: Risk factors, stages of development, and treatment. American Behavioral Scientist, 52(1), 21–37. https://doi.org/10.1177/0002764208321339.

80.

Young

(2009). Internet addiction: Diagnosis and treatment considerations. Journal of Contemporary Psychotherapy, 39(4), 241–246. https://doi.org/10.1007/s10879-009-9120-x.