The virtual simulation of child sexual abuse: online gameworld users’ views, understanding and responses to sexual ageplay
This paper explores cultural understandings of virtual sexual ageplay in the online world of Second Life. Online sexual ageplay is the virtual simulation of child abuse by consensual adults operating in-world with child computer characters (avatars). Second Life is primarily governed by Community Standards which rely on residents (the users of Second Life) to recognise sexual ageplay and report it, which requires an appreciation of how residents view, understand and construct sexual ageplay. The research presented drew on 12 months of resident blog posts referring to sexual ageplay: 263 total, with 91 residents. The analysis of this talk explores the cultural understandings of this banned behaviour and beliefs about the nature of Second Life which underpin residents’ likelihood to report sexual ageplay and so comply with the Community Standards. In considering these issues the paper is able to highlight issues regarding the unique cultural position of abuse against children and key concerns which underpin the reporting behaviour of residents. Key considerations relate to defining online sexual activity and child avatars; the moral status of ‘reporters’, and sexual ageplay as a form of edgeplay; belief in the harmfulness of sexual ageplay and its relationship to real world behaviours.
KeywordsVirtual child abuse Sexual ageplay Ageplay Second Life Moral
Sexual ageplay refers to adults engaging in sexual activity when one or more are role-playing as a child. In real life, such behaviours between consenting adults may be considered as unusual, a fetish, or even deviant, but they are not illegal because they are all consenting adults (Richards 2015). However, although virtual sexual ageplay is essentially the same behaviour but online, it is regarded as much more problematic: as the simulated sexual abuse of children in online environments (Reeves 2013). Online sexual ageplay is, therefore, regarded as evidence of a sexual interest in children as the adult players of child computer characters (known as avatars) choose to use them to act out scenes of child sexual abuse either between other adults playing as children, or adults playing as adults, or adults playing as other types of avatars. Thus, while online sexual ageplay, in itself, has no direct involvement with real life (RL) children and is only a simulation or pretence of child sexual abuse, it is considered as having indirect harms. This is particularly in light of gathering international concern round the anonymity and freedom of the online world which facilitates social networking and the sharing of real and digitally-created child sexual abuse images, as well as the potential harms of viewing fantasy images of such abuse related to child sexualisation as well as the normalisation of a sexual interest in children, which is argued to increase the likelihood of real world offending (c.f., Levy 2002; Kierkegaard 2008; Meek-Prieto 2008; Wilson 2009; Luck 2009; Nair 2010; Reeves 2013).
Consequent to this view of virtual sexual ageplay, there is increasing consideration of the legal status of this online behaviour. Sexual ageplay can arguably be encompassed under legislation on fantasy images of child abuse (Reeves 2013). Some jurisdictions, such as England and Wales though the Coroners and Justice Act 2009, have made possession of fantasy images1 illegal based upon arguments that relate such imagery to the normalisation of child sexuality, exploitation and abuse, as well as correlate it with other criminal activity, such as grooming of children and possession of real life child abuse images. Whilst the status of sexual ageplay as an ‘image’ that can be possessed is contestable, being more of a social interaction; the underlying principles for criminalisation (or non-criminalisation) remain the same: the level of indirect harm to children caused, the boundaries of criminal law, and the relationship of online fantasy behaviours to real world action. Because of the reliance on criminalising behaviours that have only indirect harm (with little current evidence base), other jurisdictions, including the US, have refrained from criminalising fantasy images to date because of the lack of direct harm to children (Herczeg 2014). The cultural status of fantasy images has also impacted on the legal status. For instance, in Japan erotic anime and manga2 are popular in mainstream culture, and possession of images of child abuse was criminalised only as recently as 2014 (although creation and distribution had been illegal since 1999), with possession of fantasy images and pseudo-photographs exempted from this prohibition, despite global pressure (Takeuchi 2015). This global uncertainty about the legal position of sexual ageplay has resulted in virtual gameworlds being required to prohibit the behaviour to ensure their users’ compliance with relevant jurisdictional laws, as well as to maintain good social order and online behaviour.
This paper will explore the online gameworld of Second Life (SL) to exemplify how people using online games or environments understand, view and respond to sexual ageplay. It is thus beyond the remit of this paper to engage in a moral, research evidence or legal debate about the harmfulness or wrongness of sexual ageplay. Exploring the cultural understandings of the issue from the users’ perspective is particularly important given the reliance of such online environments on these users to identify and report such behaviour which contravenes the rules of the environment. Their definitions and understanding of what sexual ageplay is and entails, views on the ‘wrongness’ of the behaviour, as well as on the efficacy of reporting procedures are fundamental to efforts to curb such online behaviours.
Virtual sexual ageplay in Second Life
Sexual ageplay has been of particular concern within massively-multiuser online games (MMOG) such as SL due to the fluid and anonymous nature of these virtual social environments; being flexible playing platforms in which users (hereafter referred to by their preferred moniker of residents) can socially interact with other residents in real-time through highly manipulatable avatars in ways not unlike in the real world, but unfettered by social and physical boundaries (Zhou et al. 2011). Residents can have one or more avatars (known colloquially as an avi or ava) that are created to represent anything or anyone they wish. This can be inanimate as well as animate objects, real or mythical, like the resident (or an aspect of themselves) or not. Residents can change the appearance of avatars through changing the nature and look of their features, size, shape, clothes, colour and so on, so that each avatar is a highly constructed and often artistic image. Thus, the capacity of residents to ‘play’ with and try out new realities and identities is vast and one of the attractions of SL type worlds. However, this very freedom or ‘creative empowerment’ (Stoup 2008) that residents prize is what allows sexual ageplay. Thus, since initial police attention on SL uncovered ‘paedophile’ and abuse imagery distribution networks in 2005, simultaneously discovering related sexual ageplay groups dubbed ‘virtual paedophile’ or ‘virtual child prostitution’ rings, Linden Lab (the Californian company that owns SL) has both denied the existence of sexual ageplay in SL but also, through the Community Standards, banned such virtual behaviours in any SL region (known as a sim), as well as all child avatars from adult only or ‘mature’ (M) SL sims. Thus M sims have widened in definition from being for RL adult residents only to RL adult residents and non-child-like avatars.3
The governance of these Community Standards is complex as SL operates through allowing residents to buy and create their own sims on the SL grid and so use the virtual space as they wish. Thus, SL is made up of a patchwork of co-created, individually owned and operated sims under the umbrella of Linden Lab, but within which each sim owner can set their own rules and accessibility (Stoup 2008). A resident who does not own or operate a sim can, thus, access others as they wish, unless they are for private groups, so long as they abide by the individual sims’ ground rules. These tend to not be very prescriptive as that is against the ethos of SL and SL residents, but sim owners and operators have the right to eject any resident from their sim. Sexual ageplay is one of the only virtual behaviours that are explicitly prohibited across the grid, and sim owners’ have responsibility to report any concerns regarding sexual ageplay and to eject child and child-like avatars and residents from adult-only M sims. Other residents can also report any concerning behaviour to the sim owner/manager or to Linden Lab. Thus, policing of these Community Standards is dependent primarily on SL residents reporting behaviour they regard as against the standards or the sim rules (Stoup 2008).
This reporting action of SL residents is dependent on the level of collective agreement in the social norms, the Community Standards, and prescribed penalties for breaches. Stoup (2008) considered the normative nature of SL and compared it with other RL communities that are primarily self-governed through a democratically emergent set of social norms. He argued that the environment of SL is so large, diverse, lacking in community reciprocity and agreement on what the rules should be, as well as characterised by poor monitoring and policing, and inefficient penalties for code violations that SL will always fail to be adequately managed through a social rule based code.4 Key to this lack of community or collective agreement on social norms, as Stoup (2008) notes, is not only the vastness of SL but the range and diversity of residents. This relates to not only their RL location and legal contexts, but their diversity of cultural and moral associations, beliefs and understandings in respect to ‘child’ sexual activity, online sexual activity and online morality. Thus, given that SL residents are primarily responsible for the recognition of sexual ageplay and for the reporting of it as contrary to the Community Standards, it is imperative to understand how SL residents in general understand and regard sexual ageplay and sexual ageplay in the particular context of SL. Confusion, misunderstanding, non-interventionist perspectives, ‘bystander’ attitudes, belief in the harmlessness of sexual ageplay or ineffectiveness of the reporting process and sanctions would potentially all undermine the likelihood of SL residents reporting any concerns or instances they may be aware of. Despite the importance of SL residents’ views no research could be found exploring this topic; a gap this paper seeks to shed some light upon and in doing so explore the wider issues of online policing strategies that rely on public identification and reporting.
The aim of the study is to explore the in-world cultural meanings attached to sexual ageplay and consider the ways in which such online behaviours are constructed by users of SL. Their understandings are significant in highlighting how such socially undesirable behaviours, when enacted as simulated virtual behaviours, may be considered within the context of an online adult environment which is valued for its capacity to facilitate anonymous and transgressive fantasy role and sex play ‘safely’, and often in socially disconnected ways from the real world and personhood of the user. These perspectives serve to help appreciate the particular social context in which sexual ageplay takes place and may indicate socially prevalent cultural attitudes which serve to facilitate, rationalise or excuse such online behaviour as well as support (or hinder) resident reporting of online sexual ageplay.
To enable this exploration it was particularly pertinent to this study that the direct perspectives of the residents of SL were accessed in such a way that minimised researcher and social desirability bias, given the contentious and value-laden nature of the topic. Consequently, a data collection technique was utilised that ensured that the talk of SL residents be natural and un-solicited in order to be confident that their talk be as truly reflective of their personal views on sexual ageplay as possible. Therefore, to explore the understandings of sexual ageplay amongst SL residents a qualitative grounded theory analysis (GTA)5 was undertaken on all threads referring to sexual ageplay posted in SL forums and blogs during a 12 month period.6 This approach allowed meaning, understanding and theorising to emerge from the data (Papathanassis and Knolle 2011), which, crucially, was free-talk and not the result of particular lines of questioning (as in semi-structured interviews, for example) nor limited other than by the search terms used.
Second Life forums and blogs are publicly viewable open access online areas but in which only paying subscribers may post and respond to threads about aspects of SL play. Residents post under their chosen SL username and no other identifying information is available to non-subscribers. Although SL resident usernames are fictitious and anonymous to public viewers who cannot access their profiles, they may be representative of the resident’s RL self and are linked to a sustained SL existence and social presence that makes it appropriate that quotes from the blogs are further anonymised through the use of pseudonyms to protect their SL identity and in compliance with British Society of Criminology and the host University’s ethical guidelines. Both the forums and blogs are searchable via a Boolean search term facility with search refinement options, including by time frame. Second Life forums and blogs were systematically searched in a 12 month period for all instances of ‘ageplay’ and ‘age play’. These terms were used as they encompassed the possible typing formats, further ‘sexual’ was not used to ensure the search was not overly prescriptive and all instances where sexual ageplay may be alluded to were recovered. In the forum search results are presented by thread, whereby if the search term is found in any post in that thread the whole thread is returned in the results. As posts within threads were often not comprehensible without reference to the whole thread and earlier posts, the thread was the main unit of analysis for this study, with posts as a sub-unit. The blog search returned only individual posts already recovered in the forum search and so is not referred to further in this paper.
The forum search recovered a total of 51 threads, comprising 540 posts, when duplicated threads were removed and exclusion criteria applied this resulted in a final sample of 20 threads, comprising 263 posts from 91 individual avatar accounts. The detail of this sampling process is illustrated in Fig. 1. Notably, the majority of threads excluded were so because they only included advertisements for role play areas or sexual partners with the statement: no ageplay. As a result an analysis of the meaning and discourses could not be undertaken with these posts, but consideration of the significance of the statement is included in the discussions below.
The majority of the final sample were from ‘people forum’, with 13 threads (197 posts): these were derived from the sub-forums ‘general discussion’ (7 threads, 156 posts), ‘lifestyle and relationships’ (5 threads, 20 posts) and ‘your avatar’ (1 thread, 21 posts). The remainder were from the ‘adult content forum’ (5 threads, 59 posts), ‘commerce forum’ (1 thread, 2 posts) and ‘land forum’ (1 thread, 5 posts).
Analysis was undertaken with all posts in each thread using qualitative grounded theory approach in which emergent themes are derived from line-by-line coding of the issues arising from iterative readings of the data set (Papathanassis and Knolle 2011). These themes (or codes) are then categorised and analysed to uncover underlying discourses of meaning and how these themes relate to each other and wider social discourses. Initially eighteen themes were identified, these were categorised into five order one broad themes encompassing the remaining themes categorised as order two sub-themes. The order one themes relate to: defining and understanding sexual ageplay; evidence of sexual ageplay; moral standards; the relationship between the real-life (RL) self and SL self; prevention of sexual ageplay. Table 1 illustrates all the emergent themes.
Emergent themes for ‘sexual ageplay/age play’
Order 1: Main themes
Order 2: highest level sub-themes
Defining and understanding sexual ageplay
What is a SL ‘child’?
What is a RL child?
Evidence of sexual ageplay
Tainted by association
Sexual ageplay as defining ‘unacceptable’ online sexual behaviour
Ageplay as edgeplay
The relationship of the RL and SL self
Link to real world offending
Unique environment of SL
Policing and prevention
Mechanisms of prevention
Order two theme categorisation was based on primary association to the order one theme, and are not necessarily mutually exclusive. For example, a zero-tolerance approach is categorised under prevention, but is built upon moral standards as it was talked about in terms of how ‘wrong sexual ageplay’ is. These themes encompass SL residents’ cultural understanding and ‘knowing’ of sexual ageplay, which underpin their recognition of sexual ageplay, willingness to engage with the Community Standards and report, and beliefs about how best to police and prevent sexual ageplay: thus, providing the basis of understanding the cultural and social context of self-policing and residents’ likely compliance with reporting rules.
Defining and understanding sexual ageplay
SL residents who discussed or mentioned sexual ageplay broadly shared an understanding of what ageplay is as “Adult/child AVI sex.” (SLjoy), highlighting that it is not about RL children but rather “using child avatars to engage in inappropriate sexually-themed role play” (Lashka) through “adults controlling child avs engage[d] in pixel sex with adults controlling adult avs” (Skli). Nevertheless, on occasion, this was conjoined with behaviours that related, at least indirectly, to RL children, for example, Ember wrote that they viewed the sale of sexy lingerie for child avatars in the SL market place as “age play pedo paraphernalia” and that “if any of those pictures advertising those sex outfits were in rl a bunch of people would be going to jail, obviously.” Ember, was equating the sexualised image of a child avatar in SL to the sexualised image of a child in RL (which would be illegal both in England and Wales, and in Linden Lab’s jurisdiction, even if digitally created), however, by definition, it would be more appropriate to equate the market place outfits to a sexualised image of a RL adult dressed as a child, which, of course, sets the image in a very different moral as well as legal context.
These definitions are clear in identifying sexual behaviour between avatars as central to the definition, and not just the representation of age. Despite this, some residents did conflate in their writing ageplay and sexual ageplay. For example, Koala described an experience in which a sexualised submissive “boy child avatar” propositioned their avatar for sex, ending the post with “Maybe I just don’t get the whole child avatar thing in SL with adults behind the keyboard.” Similarly, Chemix linked ageplay and sexual ageplay in commenting on “the ageplayers and pedos that are all over”. Furthermore, when discussing sexual ageplay it was not uncommon for residents to shorten the phrase to just ageplay, prompting some residents to point out the difference:
Delicia: What do I do when I stumble on ageplay in progress? [Asks questions on how to report.]
Jewel: You should report it if you’re pretty sure one of the participants was a RL minor or was roleplaying a minor. [Details given on how to report.]
Jared: Assuming the OP [original poster] was talking about sexual ageplay…No mention was made of such. Nonsexual ageplay need not be reported.
Delicia: Yes, it is sexual ageplay I am referring to. I am aware that nonsexual ageplay is completely tolerated in second life.
However, in SL, sexual ageplay need not actually involve interactions with a second party at all, as some residents explained: “a child avatar may not use adult furniture [sex-play furniture], but also may not be NEAR adult furniture or images” (Tempest). Of course ‘near’ is “somewhat open to interpretation” (Rose X) as is “what adult furniture is..is a bed automatically adult? What if no one is on it [the adult furniture]?” (Pheonix). Some SL residents enjoy family role-play and in those contexts an SL child being naked or near their adult SL parents’ bed is acceptable so long as there is no sexual context, which some residents state as not just sexual chat, dress or conduct, but also being aroused, although whether this is the SL avatars, or RL residents, or both, is unclear. Thus, recognising the boundary between ageplay and sexual ageplay is acknowledged to be very difficult in some contexts. To complicate matters further, residents discussing the ‘wrongness’ of the child avatar lingerie referred to earlier agreed this was an example of sexual ageplay due to being a sexualised image of a child avatar, although acknowledging it was a “grey area” (Jemerie) as no sex act was actually taking place or implied in the advertisement.
Intrinsic to defining and recognising sexual ageplay, therefore, as Jewel notes above, is recognising the relevant avatar as representing a child, or as predominantly human child-like (as in the case of furries, vampires, angels and the like). This may be considered a ‘loophole’ in which sexual ageplay could be occurring but without meeting these particular definitional requirements. For example, the post below appears to be advertising for a sexual ageplay relationship, but by tying this into a furry ‘pet’ master/sub relationship it attracted no attention or comment:
im a red fox looking for a loving furry master that is ok with ageplay and just will love me i don’t want real love i have a mate he just cant play i just want them to love me as a pet and sub but we can get personal but like i said i have a mate but i will love u as a master im looking for a k9 fur but bunnys will wirk or otters (FoXXy)
Consequently, it is important to the strategic efficacy of primarily relying on SL residents to report sexual ageplay incidences that residents have a collective understanding of not only what constitutes sexual ageplay, but also on what is a child avatar, which is dependent on appreciating how age is performed and represented in SL. Unlike traditional online communication, age in SL is indicated through visual and verbal cues such as: youthful dress; skin complexion, tone and smoothness; avatar attractiveness; body development, height and size; choices of activity and movement; as well as text and verbal language associated with youth culture (Martey et al. 2015). Notably, because SL residents, if they have humanoid avatars, tend to portray themselves as comparatively youthful and attractive, often being positive self-representations, the general age profile of SL avatars is youthful in comparison to the RL population. For example, in Martey et al. (2015) SL study, the average RL age of the 201 participants was 32 years, with 24% being over 45 years of age, however, none of the avatars had the normal visual cues of older age (such as greying hair, wrinkles, body shape). Further, RL younger residents tended to have taller avatars. Thus, the age of the avatar is little reflection on the age of the resident, nor can avatar age be easily understood from the avatar itself if not clearly a human child in appearance, as ‘tiny’ avatars (SL adults who are very small), animals, furries and other such fantastical avatars complicate this significantly. Therefore, recognising a SL child is not simple and was highly debated in the forum posts as the standards do not define ‘child avatar’. These discussions centred around the key age-indicators of skin tone, dress, behaviour, body type and height, acknowledging that, because of the constructed nature of avatars, the “definition for child ava would have to be very “state of the art”” (Catsya) based on the “overall impression people get from your appearance” (Charlz). Furthermore, age-indicators were often in conflict, causing confusion. For example, in endeavouring to determine the representation of the age of a sexily dressed avatar DarkLord noted the “very mature face of the model. While the body appear to be very child-like, the face (and makeup) appear to be that of a mature (age of consent) woman.” This often results in child/adult avatar hybrids than can be difficult to define and understand in terms of age as they represent “TEENS with jiggly parts much to big for most women” (VLee) as well as adult avatars that are dressed as babies or children (for example, in nappies/diapers). The later was broadly considered to not constitute ageplay at all as the avatar was intended to represent a SL adult.
VLee’s experience in trying to find a child avatar to role play family life with indicates a further confusion in terms of the avatar size and relationship dynamic. VLee explained how they found someone to role play as a child in a family context, but that this ended when the other party tried to initiate a sexual relationship with them. VLee described the other avatar as a ‘little’ who went on to request a ‘daddydom/littlegirl (DD/LG) relationship. VLee eventually blocked them and described this as an attempt at sexual ageplay. This account resulted in a number of rebuttals in defence of ‘littles’ and DD/LG relationships. Most articulately from Phoenix who wrote:
Littles are adults […] They are NOT child AVs. EVER. Daddy Dom/Mom/little relationships are a part of BDSM [bondage, domination, sadism and masochism], not part of parent/child RP [role play]. [….] Do NOT lump DD/LG relationships into paedo territory. Thank you.
The confusion stemming from hybridised age-indicators is highlighted in Felix8’s response:
I don’t see much difference outside of the shape. The talking is the same, the spread of power is the same, the clothes are the same. Just the shape.
It is evident from these discussions that size can be both an indicator of age and of the emotional or power status of one avatar to another. As Phoenix later writes:
DD/lg is a relationship between two adults. When a little is in little mode: as in she is feeling childlike, scared, needy or just in need of comfort, the Daddy) or Mommy) does not start a sexual moment.
Thus, size is a complicated issue that is further confused by the artistic trends in avatar design within SL:
Make sure it’s really sexual ageplay and not just a realistic proportioned (180 cm) [just under 6 ft] avatar. The norm in many places in SL is sadly 7 ft + which makes people who are realistically sized sometimes look like children. (Trior)
One problem is that “standard height, full sized” is pretty meaningless in RL, and entirely so in SL. There is no standard for height. For some reason, 5′3″ is sometimes thrown around as a cut-off point. We’re told that under 5′3″ is too short. [….] For most women in the world 5′3″ is too tall! (Mellow)
Of course, this is further complicated by RL difficulties in determining a ‘child’. As residents may originate and/or be located across the world, and so their legal and cultural understandings of the age, appearance and associated behaviours of a ‘child’ may vary considerably (c.f. Pasura et al. 2012), resulting in a wide scope of child-related representations (although it should be noted that SL operates under the legal code of California, the home of Linden Lab). For example, Jellie posted asking for advice when they were reported for sexual ageplay for “mistakenly” saying they were 17 in SL when they are really 67 in RL (they deliberately wanted to role play a teenager, but believed 17 was acceptable). Harmon8 commented: “I think she meant to put 17 because in most states depending on her location 16 is the actual age of consent.”
The main conclusions from this discussion are that whilst there is a broadly shared and agreed definition and understanding of what sexual ageplay is, this is very difficult to operationalise in the virtual context where none of the normal RL age-indicators are reliable. Importantly, what constitutes ‘sexual’ ageplay and child in practice are highly contested and even where there is agreement this is difficult to recognise in SL play when small avatars that “dress in sweet lolita, kawaii & pre-pubescent fashions” simultaneously claim “I am ALWAYS an adult” (Trixie). The difficulties of defining a child or child-like avatar, thus, may make it extremely difficult for SL residents to recognise when sexual ageplay is and is not taking place, as well as providing potential defences for accused residents.
Evidence of sexual ageplay
As can be gleaned from the above discussions, some SL residents had come across activities that they understood to constitute sexual ageplay. Five of the 20 threads retrieved were primarily a question asking what to do to report or avoid sexual ageplay overtures by other residents. Commonly these threads recounted accessing a region related to ageplay in some way (most commonly family role-play scenarios) where the resident was then invited to engage in sexual ageplay:
I was checking out beaches and was tpd [transported] to a family nude beach where child and adult avis were naked. It creeped me out and I left once I realized it was a nude beach for all ages. I got several IMs wanting me to be a naughty mommy. (Sheena)
All posters reported that the invitations were both unwanted and made them feel angry or disgusted:
I have open a club on this sim… But after I bought the land, I turned on my mini map, and seen a bunch of avatars in one part of the sim. Ok… So I decided to check it out. It was a kid area… I was there maybe a minute when I get an IM from some guy in a boy child avatar and hitting on me. I thought I was going to puke, plus he was wearing a “Collar” and I checked out his profile seeing that he was also collared to a mistress as a child avatar. That is when I had to leave, before I was running to bathroom to throw up. (Koala)
Although, such encounters were reported as not uncommon, they were normally described as isolated instances. A common response to such posts was residents expressing surprise that others have come across such behaviours:
I must be hanging out in all the right places.. 8 years in SL and I have yet to come across sexual ageplay. […] (Phoenix)
5 years for me and I’ve been to some seedy places. Never seen anything like this. (Nox)
The disparity between some residents’ stated experiences and others led those that had not commonly come across sexual ageplay to make judgments on assumptions about the provocative behaviours of the other residents:
I suspect the OP [original poster] is visiting places that encourage it, kinda looking for trouble. (Poloma)
Thus, residents often constructed the experiences of others regarding unwanted sexual ageplay as stemming from their own actions and decisions, considering them blameworthy in respect to being in the ‘wrong’ places or in some way inviting this attention; with this reflecting negatively on their personal character: “why would you even want to be there?” [sim involving suspected sexual ageplay] (Skli). That residents posting about sexual ageplay, clearly stating it was unwanted, were nevertheless stigmatised as a result of this contact sheds light on how sexual ageplay is viewed through a moral as well as regulatory lens. In so doing negative attributes are applied to those who are associated with the disapproved social behaviour (sexual ageplay) and they become discredited in their community (Goffman 1963). That this is happening to residents who are not directly involved in sexual ageplay, but only coming into contact with it indicates a transference of stigma through which residents become tainted by association; with other residents attributing negative characteristics to them based solely on their unwitting contact with such universally socially disapproved behaviour.
This attribution of transferred stigma highlights how people thought to be engaged in sexual ageplay are othered: constructed as separate and different to the norm (Young 2007). For the most part this is through constructing sexual ageplayers as akin to paedophiles and sexually abnormal, with residents’ expressed reactions being visceral: referring to “revolting” (Draxx) “sick groups” (Aaas) that make them want to “puke” (Koala). In doing this residents are defining acceptable and unacceptable behaviour and conduct in SL. Sexual ageplay is:
[…] just about the only kind of behaviour pretty much universally condemned in Second Life. (Lashka)
Thus, sexual ageplay is understood to be a form of ‘edgeplay’: sexual play that transgresses or at the limit of social acceptability (Butts 2007). Sexual ageplay is stated as being the defining behaviour which sets this boundary. For example, as mentioned in the sampling, many adverts were retrieved that were seeking other residents to engage in sexual play with, but which stated that sexual ageplay was not permitted:
Roleplay and Sex without Limits (except child/age play) […] No Child Avies/No Age Play… never ever (Phor)
This boundary norm creation is a construct recognised as evolving from the diverse cultural interaction of SL residents:
While a common mode of thought is that SL is what you make of it, the greater truth is that SL is what we make of it. “We” as in all of us, everyone, and all at once. (Stormi)
The boundaries of the freedom of expression in SL are very significant as SL is valued for its openness and freedom to experiment with sexual interests as well as identity and social interaction without judgement: “you don’t get to choose how other people enjoy Second Life” (Viking girl). This leads to SL residents arguing that “combating intolerance is a cornerstone of Second Life’s Community Standards.” (Flaxx), and that “decent [SL] places are filled with nice accepting people.” (SLover). This acceptance and tolerance extend to the simulation of a wide range of sexual and violent behaviours that in RL would be unacceptable and criminal, from role-playing sex slaves, torture, (adult) incest, and rape to humiliation and prostitution: “every form of human depravity rages on in world.” (Jellie). For example, Flame defended the right to be ‘kinky’ in SL and that “SL is a relatively safe place for people to play” and that her own interests meant that:
I’ve been incinerating people in SL since the day I discovered fire, some more than once. I get great pleasure out of it.
The response to such an admission—the simulated sexually-motivated murder of others—led not to condemnation or stigmatisation but light-hearted banter:
I’m well-aware you want to burn me down, and am okay with that ϑ (Viking girl)
This moral boundary of acceptability between virtual sexual murder (the incineration of avatars, who can come back to life) and virtual sexual ageplay (as understood to be representing virtual child sexual abuse) reflects the ‘gamer’s dilemma’ as posed by Luck (2009) regarding why some behaviours that are considered morally taboo and legally sanctioned in RL may be acceptable or even required in online environments whilst others are not; questioning the moral difference between the two groups of online behaviours. For example, assaulting, killing and even sometimes torturing other computer characters or avatars is not only accepted, but normal and intrinsic to many MMOG gameworlds, despite their moral and legal position in RL. However, virtual paedophilia, necrophilia, bestiality and rape have been generally regarded as beyond the realms of acceptability, even in virtual simulation. However, even these moral absolutes of acceptable online sexual behaviour are being eroded in SL. Resident’s discussions of incest role-play help to further illuminate the boundaries of sexual morality in SL. When Fallyn asked for any suggestions for family role-play incest regions some other residents argued this was against the standards in respect to sexual ageplay, but this conflict was quickly resolved when it was pointed out that it was strictly between adults and this was not against the Community Standards as it would not make sense as the avatars have no biological relationship to each other (due to not being biological beings). Thus, incest role play was acceptable because it was only considered as an in-world behaviour and not linked to RL or the RL-self of the residents in anyway, and did not involve children or the image of children.
This begs the question of why would virtual murder, torture and incest be morally acceptable, but not virtual paedophilia (as which sexual ageplay can be regarded)? The question is complex and most academics conclude that morally there is no real difference in the online nature of the behaviour beyond the culturally shifting notion of public disgust and abhorrence, and so we need to look to other issues such as psychological harm (c.f. Young and Whitty 2010), or morally reconceptualise sexual ageplay online as child pornography, regarded as intrinsically immoral as it sexualises inequality (c.f. Bartel 2012).7 These positions draw a distinction between virtual murder (for example) and virtual paedophilia, saying that the former is either (or both) not psychologically harmful or not a legally and morally prohibited image, and therefore acceptable in virtual space. (Young and Whitty 2010, temper this by arguing it is only acceptable in virtual space where these sorts of behaviours are collectively accepted by the online community and it is an agreed ‘status function’.) Bartel’s (2012) position of moral (if not legal) difference based on it being an image of child pornography and sexual inequality is considered by Luck and Ellerby (2013). They argue that equating virtual paedophilia with child pornography is not always appropriate because in gamespace it may be part of the game structure to commit such acts, therefore the player is not choosing this virtual behaviour voluntarily and/or not taking enjoyment from the intrinsic act; which they argue is fundamental to Bartel’s understanding of what child pornography is. Partridge (2013) takes issue with this claim, arguing that just because something is pornographic, does not make it necessarily pornography.8 However, if players do take enjoyment from the act of virtual paedophilia, why does it not follow that taking enjoyment from virtual murder (as in the case of Flame, above) would also be immoral?9
This question is considered by Young (2013, 2016) who explored the role of player motivation in determining the moral status of virtual murder or virtual paedophilia. He highlighted the difference between the motivation of the avatar in carrying out the act in-game, and the motivation of the player (either for choosing to make the avatar act in that way, or choosing a game which includes those actions). Young (2013) points out that it is intuitive for us to make negative attributes about the player based on their motivation, but that we often infer player motivation from the avatar’s (possibly incorrectly). For example, an avatar acting out virtual murder or child abuse could, conceivably, be required (or encouraged) to do this through a game structure, thus, relieving the player from attributions of personal choice and enjoyment in the act—although such an inference may be incorrect. Ali (2015) similarly argues that the gamers’ dilemma could be partially solved by reframing the dilemma itself: that virtual murder and virtual paedophilia are not actually considered morally different in the first place, if you control for contextual differences. The difference he highlights is that virtual murder tends to take place in the context of legitimised violence (war, for example), but also in the context of furthering a game. Consequently, those virtual behaviours which align with the game structure and design are considered to be outside of the player’s moral decision-making, and thus acceptable as they do not indicate any relationship to the player’s motivations nor, therefore, relate to the possibility of psychological harm to the player. In arguing this case, Ali (2015) draws on the work of Partridge (2013) who also points out that not all virtual murder is acceptable either, and that the moral status of virtual murder is founded on the moral status of such acts in the real world (taking into account the lack of direct harm caused virtually). Thus, some virtual murders are more acceptable than others, with player’s more likely to morally judge other players negatively for virtual deviant behaviours where they conceive of the game (and acts within the game) as reflective of RL (be that the deviant motivations and desires of the player or broader social problems of, for example, inequality and discrimination). The player’s motivations, therefore, can be understood (however incorrectly) in light of a game structure which allows us to divorce the choice to commit virtual murder from the RL desires of the player. In contrast, however, virtual paedophilia is not a requirement of game structures and so is understood as inferring the RL desires of the player.
In respect to SL, however, there is no structured gameplay (beyond that potentially created by the residents themselves), Ali (2015, p. 273) suggests that in such gameworlds (or simulation games) that neither virtual murder nor virtual paedophilia would be considered acceptable because “the act here reflects the sorts of acts the gamer finds desirable.” But, as can be seen from the talk and responses to Flame, above, this is not the case in SL. Here moral distinctions are made between virtual murder and virtual paedophilia (as enacted through sexual ageplay) despite that both are unstructured and un-coerced by the gameworld. Thus, indicating that Ali’s thesis, at least, is not the basis of SL residents’ moral frameworks. Further, Partridge’s (2013) distinction based on targeting and discrimination does not seem reflective of the talk of SL resident’s either. Rather, those that highlight child pornography (Bartel 2012) and link moral judgments to views on the connection between the resident’s motivation and potential indirect harms to the resident or RL children, appear to better reflect residents understanding of sexual ageplay. However, as Young (2013, 2016) points out, such consideration of the player’s motivation does not explain the selective acceptability of virtual murder, for example, and not virtual paedophilia (sexual ageplay), unless also coupled with Ali (2015) distinction.
Young (2013) argues, however, where the player does gain personal enjoyment from such a virtual act, this may be not because of the act per se but rather, because of the thrill in transgressing social norms. This is highlighted in SL discussions around the acceptability of simulated rape and rape role-play, which show how embedded considerations of power and control are to SL culture, with some residents arguing that ‘playing’ with power and control is part of why many people spend time on the SL grid but that it is nevertheless a consensual arrangement in which both parties retain control and “one partner chooses to ‘submit’ while the other chooses to ‘lead’” (Viking girl). Rape-play was, therefore, seen by many residents as an acceptable fantasy extension of domination and submission sexual relationships which, despite outward appearance, remain founded on mutual respect and shared control. Notably no-one argued this about sexual ageplay relationships.
Of course, these claims of power-play as being between equals were challenged by residents pointing out that predominately the dominant partner (or rapist) plays a male avatar and the submissive (or victim) a female, and thus rape-play is actually indicative of, and maintaining, patriarchal structures of male dominance. Its relative frequency, taken with the sexualisation of the female body image in SL and prevalence of virtual prostitution, is regarded as reflecting RL culture, but able to be more freely and overtly expressed in the virtual world. Those arguing against the acceptability of such play claimed, therefore, that residents wanting to be degraded in SL only did so because of RL social patriarchal structures, or individual psychological and emotional issues they suffered from, and so should not be exploited. Again, other residents strongly argued against this perspective, invoking the freedom and tolerance embedded within SL:
Whats wrong with two consenting adults exchanging money for [sexual] service? No matter if you would like to participate in it or not or if you find it disgusting, its none of your business or your place to judge. (Jemrie)
From exploring the boundaries of virtual sexually acceptable edgeplay in SL two fundamental bases of moral positioning can be identified: most edgeplay is acceptable because it is regarded as consensual play, either not indicative of RL society or the RL behaviours and desires of the resident. That which transgresses the boundary is regarded as indicative of problematic RL social structures or the residents RL behaviours and desires. This still leaves the question as to why sexual ageplay online is so universally condemned as it is consensual behaviour, nor is it argued to directly harm the residents involved. Further, it is not considered to reflect broader problematic social power structures (for example as rape-play may be argued to), but rather only on the ‘perverted’ individual pathology of the residents. The moral ‘otherness’ of sexual ageplay thus, appears to lie in what residents believe it reveals about the RL self and motivations of the residents playing the avatars and their sexual interest in children, which leads to questions of how residents regard the relationship of the RL and SL self.
The relationship of the RL and SL self
Outside of discussions of sexual ageplay the relationship between the RL and SL self is very contentious and highly debated with residents variously describing SL sexual relationships as “playing” (Marlyne) and “it is a fantasy” (KrashD), but others arguing that SL “mirrors” RL (Bookworm) and that RL and SL interests are related: “In SL they seek for power to compensate for their powerless, insecure characters.” (Flaxx). It seems, therefore, that some residents, at least, are concerned about the motivations of people who choose to ‘play’ in SL behaviours which are indicative of what they consider to be abnormal sexual interests in the resident, not just the avatar. Thus, in respect to sexual ageplay:
Ever thought about why LL has forbidden Age Play (Child Pornography Roleplay) in SL? No real child is harmed if two adults play child pornography in SL… Because there is so much more to it. It is not about whether a real child is hurt or not. It is about the idea of having sex with a child. Every sane person would agree that this idea is sick and wrong. Also it is known that such fantasies sometimes are the beginning of the plan of a crime. Such fantasies are not harmless at all. (Periate)
Periate sums up why sexual ageplay is considered unacceptable and immoral by many residents: because it is indicative of a RL sexual interest in children that could potentially lead to real-world offending. Although there is some debate about just how directly related sexual ageplay and real world offending is (Reeves 2013), no resident defends sexual ageplay (though this may be because the act is against the Community Standards), however, Periate goes on to argue that the same is true of rape and torture groups in SL as they devalue women, and other residents do disagree openly about this, calling it only fantasy role-play even when they have just agreed with the wrongness of sexual ageplay:
Child pornography is definitely sick and wrong. Nobody disputes that. [….now referring to rape and torture play] As long as they stick to fantasies, and don’t act them out in RL, nobody gets hurt. It’s not your job, or anybody else’s, to dictate what goes on in other people’s brains. (Drey)
From Drey’s post it seems that perhaps the moral distinction is not based on perceptions of the indirect harm caused to the residents playing child sexual abuse, rape or torture, despite the arguments put forward. Instead the distinction stems from the simulation of child sexual abuse in SL being understood as a criminal offence, because it is regarded as a form of child pornography, and the latter (simulated rape or torture) are not because they are considered as a form of adult pornography, even if some residents consider them obscene or distasteful. Further, that sexual ageplay is considered akin to RL images of non-consensual sexual activity between an adult and child, and the others are consensual, even where consent is pretended to not be present: i.e. SL rape images are akin to RL image not of rape but of rape-play. This denies that SL sexual ageplay is actually also more like RL images of a pretend child consensually acting out child abuse-play, which is not illegal. Thus, from the SL forum posts it is clear that there is little logical distinction between sexual ageplay and other forms of edgeplay, but there is a strong moral and emotional distinction. This separation is validated by the Community Standards which do not prohibit consensual edgeplay in adult sims between adult residents, other than specifically sexual ageplay.
Policing and prevention
As a result of residents widely agreeing with the moral wrongness of sexual ageplay, and that such virtual behaviour is indicative of harmful RL desires or future criminal behaviours, broad support for a zero-tolerance in-world approach existed not present in discussions about other edgeplay behaviours: “First do be aware that ageplay is OFF THE MENU HERE.” (Lyra). For example, in discussing the suspension of a resident for breaching the Community Standards, support was given along with advice on how to appeal until it emerged that the breach was for presenting as a child in an adult sim which led to a withdrawal of support: “if you were banned for ageplay, well, bye.” (Bleeper) as well as hostility and sarcasm: “I shall now have to resort to wallowing in the wealth of tears which have been induced by your departure.” (Drey). This led to many of the threads discussing sexual ageplay to include support and guidance to “go there and Abuse Report it” (Dotti) coupled with indicators of moral condemnation: “if you have to live near that creep, just report him/she/it.” (Snowflayke).
This zero-tolerance approach coupled with the difficulties of operationalising the definition of sexual ageplay, or ‘child-avatar’ in practice, as discussed above, resulted in a risk-averse culture in which child-avatars were often uniformly banned from adult M sims, irrespective of the sexual, or not, nature of the area or the avatar’s behaviour, as described by Jenka, a manager of several adult regions:
We are strict about excluding child avatars from our Adult regions, since we’re very conscious of LL’s policy disallowing age play […] and we don’t want to put either our visitors or the estate owners at risk. [when asked about the look of a particular avatar] the pictures seem to show an avatar who looks sufficiently close to 15 or 16 to worry me. […] we’d politely ask you either to change your avatar or leave.
Similarly Koala explained their child avatar policy:
But I also have a “No Child Avatar” policy with my club…And sure enough, had a kid pop in, told she had to leave, she tried to argue with me on it. I banned her really quick.
This risk averse culture was believed to stem from Linden Lab’s own response to allegations of sexual ageplay: “LL have an itchy trigger-finger on this issue […as…] they have to be cautious in the first instance” (Viking girl). As a result residents advised each other, when wondering how to respond to potential sexual ageplay, to “Just file the AR [abuse report] and let the company deal with it.” (Buzzedoff), “A report will at least get the REAL judge to make a decision” (DarkLord). Conversely, risk averse behaviours resulted in individual residents feeling the need to defend themselves as not being child-avatars prior to any allegations being made:
In my Profile I have stated in several places that my avatar and me in RL is 18+. Here is what my Appearance Pick in my Profile reads:
I am ALWAYS an adult who [describes how presents as childlike]. By expressing myself this way DOES NOT mean i am a child. At ALL times i’m an adult mentally, emotionally and physically! […] i may look small & you are right, i’m petite in RL (4 ft 11in 100lbs) & represent my avatar as best i can here. Yes, i have realistic sized breasts. Just cos i don’t have some massively sized breasts doesn’t mean anything! We are have different sized bodies. accept it please! (Trixie)
Similarly Cressida explained that they would make their avatar appear more mature in adult sims, particularly where a lot of sexual activity took place in order to avoid allegations of sexual ageplay. Cressida described feeling like other residents could report them for sexual ageplay even when the avatar “wasn’t doing anything, or wasn’t engaging in sexual activity”, thus, describing a synopticon world resulting from the self-policed nature of SL and moral status of sexual ageplay. This could result in tensions between residents with some arguing that residents who regularly or frequently reported potential abuses were “busybodies” who should “Let LL police it’s own world” (Skli).
Other residents proposed mechanisms to try and make the policing and prevention of sexual ageplay more efficient and effective, without relying on residents. For example, Catsya suggested that avatars could be pre-approved, and that all child avatars should be reviewed for approval before allowed on SL and then only to be used in G(eneral) sims, coupled with algorithms to identify when child avatars are in pre-defined positions or places and automatically report these as abuses. Catsya recognises that this is not necessarily a popular idea in a forum which residents prize for the freedom it represents, but argues it is acceptable in respect to sexual ageplay as it is against the Community Standards and breaches the RL law in Linden Lab’s jurisdiction. Only two residents responded to this post, both to ‘roll their eyes’ to indicate they felt it was an over-reaction.
Contrary to Stoup’s (2008) more general work on self-policing and collective agreement of social norms in SL, through analysing residents’ natural talk in forums it was evident that there is evidence to support that many of these foundations required to support SL residents in identifying and reporting sexual ageplay are present: there is a shared strong condemnation of the behaviour, coupled with an appreciation of the rules banning it, broad resident support for these rules and a willingness to report suspected sexual ageplay, and to encourage reporting by others. However, these foundations are undermined by resident difficulties in always understanding what ‘sexual’ constitutes or recognising ‘child’ avatars, as well as moral condemnation of those who appear to come across sexual ageplay regularly or gain a reputation as a frequent abuse reporter. This has resulted in a culture which is overly risk averse to suspected incidences of sexual ageplay; with many residents blanket-banning child-like avatars from adult sims or condemning all residents with child or teen avatars, irrespective of their behaviours online. Such risk-averse and intolerant reactions results in some residents criticising these indiscriminate moral positions: pointing to the freedom valued in SL and the harmlessness of fantasy-play. However, such alternate positions on the moral wrongness of sexual ageplay were very rarely indicated, and were vocally condemned by other residents.
Underpinning these shared cultural values about sexual ageplay is a common understanding of sexual ageplay online as a form of virtual paedophilia rather than as a form of sexual fetish between consenting adults. As such, in keeping with previous work on the morality of sexual and violent deviance online, sexual ageplay is almost universally denounced as wrong, ‘sick’ and abnormal. This construction results in normative moral positions based on an understanding of sexual ageplay online as both a form of child pornography and also indicative of the true RL desires and motivations of the resident considered dangerous due to being part of a fantasy rehearsal cycle preparing the resident for RL offending against a child. It is this common belief in the relationship of sexual ageplay online to the RL resident that sets it apart from other forms of online sexual and violent behaviours. Thus, similar edgeplay behaviours such as rape-play, murder-play, incest-play are broadly culturally accepted because they are constructed as only consensual fantasy and so not as reflecting or affecting the RL resident and their future behaviours. Thus, for the residents of SL, normative moral positions on virtual deviant behaviours are founded on the perceived relationship to RL (reflecting, in part, arguments around the gamer’s dilemma of Bartel 2012; Partridge 2013; Ali 2015; Young 2013, 2016), but this connection is more readily assumed in respect to sexual ageplay than other virtual deviant behaviours, leading to questions about why sexual ageplay has this unique position even when the context of the behaviours are identical. However, as noted, when residents also believe that there is a connection with other forms of edgeplay and RL problematic social structures or behaviours, such as for rape-play on occasion, then these residents also tend to express the wrongness of the play online. But these views are not widely culturally shared within the SL community as it is uniquely for sexual ageplay.
This unique normative moral position of sexual ageplay within the otherwise ‘free’ and socially accepting virtual world of SL indicates that more work is needed to explore the reasons why sexual ageplay is regarded as abhorrent whilst other similar sexually fetishistic deviant edgeplay behaviours, which could be argued to indicate RL criminal and harmful propensities in residents, are not, even when all such behaviours are within non-game structure virtual environments. Perhaps using interview techniques to explore the findings observed in this paper with residents, such research would help to uncover the basis for attitudes in society broadly (not just online) which serve to morally separate offending against children from other interpersonal violence. Related to this is a need to explore the extent to which residents regard their, and others, avatars as extensions or reflections of their RL self. These questions are essential in further exploring how to develop the efficacy of virtual gameworld or broader online strategies to support residents as the frontline in identifying and reporting sexual ageplay.
Fantasy images are distinct from pseudo-photographs or digital images which are criminalised in a wider range of jurisdictions (including, for example, the US) and were first criminalised in England and Wales through the Criminal Justice and Public Order Act 1994 which amended the Protection of Children Act 1978 to extend the definition of indecent images of children to include digitally created images. These images are normally required to be virtually indistinguishable from a real photograph of abuse. Often legislation prohibiting such images was primarily argued on the grounds of investigative need: where photographs and pseudo-photographs were difficult to tell apart and were introducing reasonable doubt in prosecutions (Akdeniz 2008).
This includes the manga subgenre of Lolicon (Lolita Complex) which sexually depicts children and adolescents.
Notably, it cannot be for RL adult residents and SL adult avatars only as some avatar types cannot be referred to in age-terms easily, this includes ‘furries’ (animal/human hybrids), animals, mythical creatures and the like, but can also be objects.
Possible sanctions by Linden Lab range from informal warnings to termination of accounts, and possible reporting to RL police. The latter of course has its own set of complications relating to issues such as police and legal jurisdiction, evidence collation, user identification and inter-service working.
GTA is an analytical approach first set out by Glaser and Strauss (1967) and is founded in symbolic interactionism. Through GTA, theory is derived from the data by undertaking a constant comparison analysis involving coding the data by categorising it into themes (or codes) based on patterns of meaning in the data. These themes are further iteratively analysed (and coded) for conceptual meaning and relationships between codes. Thus the emerging theory is ‘grounded’ in the data. GTA is used primarily in exploratory studies of previously neglected or new research phenomenon to gain “knowledge of social realities […] achieved through careful observation of behaviour and speech practices” (Stark and Trinidad 2007: 1374).
Dates are not given to help preserve the anonymity of the posters.
It is not within the scope of this paper to do more than briefly review the Gamer’s Dilemma debate; this will be the central subject of a forthcoming paper. Please see my earlier paper (Reeves 2013) for a discussion on the complexity of the legal position of sexual ageplay in England and Wales.
Partridge (2013) argues that virtual paedophilia is immoral but not based on this reasoning, but rather because of the targeted discrimination of children. Young (2016) counter-argues this by pointing out that even if one conceived of a game where the player indiscriminately sexually assaults all classes in society, including children, this sexual assault of children would still be regarded as particularly more inappropriate than against adults (especially if one could choose not to assault them). Further, he points out, there is still a form of discrimination being enacted, if only against those near the player’s game location.
It is apparent that many of these arguments relate, for the most part, to role-playing games in which the player is acting-out a constrained fantasy-role in gamespace with game structures, rules, purposes and so on. Although there is much discussion of the blurring of the boundaries between these online roles and the RL self (c.f. Seung-A, 2012; Nagy and Koles 2014; Schultze 2014), this is much more of a concern in respect to non-gamespace online MMOG worlds such as SL because the world is created by players and they have almost infinite choice as to how to use and interact in that world via their self-chosen and created avatars.
Compliance with ethical standards
Conflict of interest
The author declares that they have no conflict of interest.
- Akdeniz, Y. (2008). Internet Child Pornography and the Law: National and International Responses. Aldershot: Ashgate.Google Scholar
- Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.Google Scholar
- Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Jamaica: Pelican.Google Scholar
- Herczeg, J. (2014). Actual problems of possession and viewing of child pornography in internet. JURA, 1, 70–80.Google Scholar
- Meek-Prieto, C. (2008). Just age playing around? How second life aids and abets child pornography. North Carolina Journal of Law and Technology, 9, 88–109.Google Scholar
- Stoup, P. (2008). The development and failure of social norms in Second Life. Duke Law Journal, 58, 311–344.Google Scholar
- Takeuchi, C. L. (2015). Regulating Lolicon: Toward Japanese compliance with its international legal obligations to ban virtual child pornography. Georgia Journal of International & Comparative Law, 44, 195–236.Google Scholar
- Wilson, R. F. (2009). Protecting virtual playgrounds: Children, law and play online: sex play in virtual worlds. Washington and Lee Law Review, 66, 1127–1174.Google Scholar
- Young, J. (2007). The vertigo of late modernity. London: Sage.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.