Keywords

1.1 Virtual Murder: The Current State of Play

Within single-player video games (hereafter, video games), it is permissible to engage in simulated murder. By murder, I mean the intentional and unlawful killing of an individual. Indeed, it is far from hyperbole to say that a large percentage of violent video games contain acts of simulated killing, many of which would be categorized as murder or as otherwise unlawful if performed for real. To illustrate, Cunningham et al. (2011) report that from a total of 1117 video games sampled, 672 were identified as non-violent and 445 violent (based on the Entertainment Software Ratings Board’s (ESRB’s) ratings and content descriptors). Of the 445 violent titles, 113 were considered to be extremely or, as Cunningham et al. refer to them, ‘intensely’ violent. Moreover, Prigg (2009) reports that, on the first day of its release, the video game Call of Duty: Modern Warfare 2 sold 4.7 million copies in the USA and UK alone, outselling the previous best video game – Grand Theft Auto IV – by some distance. Both the Grand Theft Auto and Call of Duty series are held to be extremely violent games. (Before proceeding, a point of clarification: reference to ‘violent video games’ should be understood as short-hand for video games whose content contains simulated violence.) Call of Duty: Modern Warfare 2 became infamous for its airport massacre scene, and Grand Theft Auto IV permits the gamer’s character to have sex with a prostitute before mugging or even killing her. The popularity of violent or even extremely violent content does not appear to be waning. As Haynes (2015) notes:

In 2015, we saw some of the most violent video games ever released. Plus, older violent games such as Gears of War: Ultimate Edition and Resident Evil: The Definitive Edition were re-released with visual upgrades that intensify the more violent moments, including blood and gore splattering (p. 1).

When describing the current state of play (meaning those games currently available to age-appropriate persons in the UK and USA), enacting murder is not only permitted but a common occurrence; some might even say ‘positively encouraged’. In Manhunt 2, for example, I (in the form of an avatar) can bludgeon to death a stranger with a kitchen utensil. Postal 2 allows me to set someone on fire while they are alive, douse the flames by urinating on them, before beating them to death with my boot and a shovel. More recently, the video game Hatred has courted controversy through its seemingly relentless enactment of random murder (Campbell 2014). In contrast, the current state of play does not permit video games to contain enactments of paedophilia.Footnote 1 One quick and easy way to account for this discrepancy is to point out that virtual child pornography, which would include the virtual enactment of paedophilic acts, is illegal in many countries, including the UK and, with qualification, the USA.

Before discussing the legality of virtual paedophilia (both for the purpose of clarification and as a means of informing the moral debate to come), one might ask with some incredulity: why would anyone want to do that? By ‘that’, I mean why would anyone want to play a game in which they can simulate paedophilic activity and therefore, to all intense and purposes, play at being a paedophile? The intuition underlying this question and the incredulity with which it might be asked seem to appeal to player motivation. Crudely put, one might suspect that there is something wrong with someone who wants to play at being a paedophile; that their motivation to enact paedophilia stems from the fact that it vicariously satisfies, and is therefore a symptom of, their desire to engage in actual paedophilia. Or perhaps, one fears the risk of enacting this activity within a game; that, somehow, repeatedly engaging in such simulations may lead one to acquire a taste for what the simulation represents (a kind of slippery-slope argument). Of course, some people may question the motivation of individuals who play a game like Postal 2 in which one can enact all kinds of extremely violent acts. Returning to the earlier example, they may ask with equal incredulity why anyone would want to play a game in which it is possible to set someone on fire, urinate on them to douse the flames and then beat them to death. Is enacting this kind of activity likewise a symptom of some other desire: namely, to engage in actual murder? Although there will be dissenters, I suspect the majority response would be ‘no’. It is, however, a question I will return to.

1.2 The Gamer’s Dilemma

Virtual murder is permitted in the UK and USA, even when enacted with the level of violence depicted in video games like Postal 2 (as one example among many). Given this, consider the words of Morgan Luck when introducing the gamer’s dilemma:

Is it immoral for a player to direct his character to murder another within a computer game? The standard response to this question is no. This is because no one is actually harmed as a result of a virtual harm. Such an outlook seems intuitive, and it explains why millions of gamers feel it is perfectly permissible to commit acts of virtual murder. Yet this argument can be easily adapted to demonstrate why virtual paedophilia might also be morally permissible, as no actual children are harmed in such cases. This result is confronting, as most people feel that virtual paedophilia is not morally permissible. (Luck 2009, p. 31)

According to Luck, the dilemma gamers face – or indeed anyone faces who has a view on the selective prohibition of video game content (Young 2013b) – is that any appeal to rudimentary arguments avowing ‘no harm’, used to rebut criticism of our intuitions over the permissibility of virtual murder, can also be used to challenge any intuitions we may have about the impermissibility of virtual paedophilia. If the claim is that no actual harm occurs as the result of virtual murder then, likewise, why should it not be claimed that no actual harm results from virtual paedophilia? Given the permissibility of the former, why prohibit the latter? What justifies our contrary intuition, here? Where our intuitions are shown to be inconsistent or seemingly without support, at least after a cursory examination, the gamer (or any other interested party) is faced with a dilemma. If one wishes to achieve parity, either one prohibits virtual murder and virtual paedophilia (resulting in the unfortunate consequence of prohibiting an activity many gamers intuitively feel is acceptable and indeed enjoy enacting: namely, murder) or one permits each of these activities (thereby creating a different unpalatable consequence: allowing the enactment of paedophilia, which many would find repugnant). Of course, one could simply admit to having inconsistent and, it would seem, indefensible views about different virtual content; indefensible, that is, outside of an appeal to the popularity of certain intuitions.

Appeal to intuition is not a sage strategy, however (something we will return to in Section 2.1); a conclusion Luck himself acknowledges. Indeed, much of Luck’s original paper on the gamer’s dilemma sets out to examine “whether any good arguments can be produced to reconcile the intuition that virtual murder is morally permissible, with the intuition that virtual paedophilia is not” (2009, p. 31), thereby making such seemingly inconsistent intuitions defensible through evidence and/or argument. Luck concludes that there are none.

1.2.1 A Brief Overview

Since the introduction of the gamer’s dilemma, a number of ways of resolving it have been suggested, and debate continues over their respective success. In what is to follow, I will consider each of these arguments in turn and present various responses to them: mainly in relation to competing or absent empirical findings (where certain findings are required to support an argument) or through the identification of internal inconsistencies and/or conceptual incoherence within the argument itself. On completing my critical review in which, to a greater or lesser degree, I identify problems with all previous attempts at resolving the dilemma, I present my own thoughts on how we might approach finding a solution.

Chapters 2 and 3 will be taken up with the different ways in which Luck tries to resolve the dilemma, none of which he finds wholly convincing. In his original paper, some of his suggestions are given only cursory treatment, I therefore expand on the reasoning Luck uses in each case. My aim is to provide further support for the conclusions he draws and although, in places, I disagree with the manner of his argument, I nevertheless concur with his overall dissatisfaction with the suggested means of resolving the dilemma. In Chapter 4, I consider Christopher Bartel’s attempted resolution (Bartel 2012). Here, I present a systematic critique of each of the premises on which he grounds his argument. I find each problematic in its way. In my appraisal, I draw on recent (i.e. 2013) responses to Bartel’s paper from Stephanie Patridge and also Morgan Luck and Nathan Ellerby. In Chapter 5, I consider Patridge’s reply in more detail, and offer some critical thoughts on her position. In many respects her argument is promising; although not without its problems, as I discuss. I also consider Rami Ali’s work on the gamer’s dilemma. Ali (2015) offers an original approach which, again, shows promise – particularly his thoughts on different contexts – but like all previous attempts is not without its problems.

With the exception of Ali, all other attempts at resolving the dilemma have accepted Luck’s claim that there is a difference in our intuitions over the permissibility of virtual murder and virtual paedophilia. If we likewise accept (for now) this claim as our starting point, then what forms the basis for this difference? Are our intuitions tapping into and therefore describing some independent moral fact – in a moral realist sense – or are they indicative only of a difference in our moral attitude towards these respective virtual enactments: an attitude that neither describes nor derives any moral authority from putatively independent moral truths? If moral realism is true then it appears unable to inform attempts at resolving the gamer’s dilemma, as I hope to show in my critical review throughout Chapters 2–5.

In Chapter 6, I therefore adopt an anti-realist approach and, in doing so, present my own thoughts on how the gamer’s dilemma could be resolved. I discuss constructive ecumenical expressivism: a meta-ethical approach to moral utterances which I have previously applied to virtual gaming content (Young 2014, 2015b). I argue that constructive ecumenical expressivism provides insight into what our moral intuitions amount to and therefore why there is a difference between our moral attitude to virtual murder compared to virtual paedophilia. Once the nature of this moral attitude is understood (in terms of the basis for its formation), differences that exist between our attitude towards different virtual content can be articulated in morally relevant terms, whether in the context of the gamer’s dilemma specifically or selective prohibition more generally. Constructive ecumenical expressivism not only proffers a means of resolving the gamer’s dilemma but, importantly, is robust enough to be co-opted as a normative ethic applicable to all forms of virtual gaming content.

I would like to finish this chapter by saying something about the legality of virtual child pornography, predominantly within the UK and USA. Initially, to illustrate ways in which legislation is similar or differs between these two countries, but more importantly to make the point that the focus of this book is on the morality of video game content irrespective of its legal status. In other words, irrespective of the legality of virtual paedophilia, what arguments are there for or against its moral prohibition, and are these able to differentiate between virtual paedophilia and virtual murder in a morally relevant way? That said, I believe that an understanding of some of the key legal arguments for and against virtual paedophilia will prove to be of use when debating the morality of certain activities within video games.

It is also worth noting that I consider a detailed examination of the different ways theorists have attempted to resolve the gamer’s dilemma to be crucial to an understanding of what Whitty et al. (2011) refer to as symbolic taboo activities (STAs): basically, the virtual enactment of all activities deemed to be taboo (qua illegal and/or immoral) in the real world, such as assault, torture, rape, murder, paedophilia (including incest), bestiality, necrophilia and so on. As alluded to above, what we will learn by considering arguments for and against the selective prohibition of virtual paedophilia will, in turn, provide a platform for further discussion on the morality of STAs more generally and, in the case of constructive ecumenical expressivism, perhaps point the way to what an agreed normative approach to policing all video game content might look like.

Before discussing any of this, however, I will present a brief exposition of the legal status of virtual paedophilia (for now, under the umbrella term ‘virtual child pornography’), noting similarities and differences between the legislation of the USA and UK, respectively. I intend to discuss the US position first and in more detail simply because (1) more academic literature is available on US legislation, particularly in relation to freedom of expression, (2) it provides a good comparison with UK and other countries’ legislation and (3) current debate on the criminalization of virtual child pornography, which I wish to use to inform my discussion on the morality of virtual paedophilia, is largely based on US legislation. First, however, a point of clarification is required.

1.2.2 The Homogeneity of Virtual Murder

Bartel (2012) queries what he perceives to be Luck’s treatment of violent video games containing enactments of murder: that they are essentially treated (by Luck) as the same; as part of one homogenous group. Bartel claims that, morally, gamer’s will approach acts of killing, including murder, within games in different ways. I accept that gamers may well do this, depending on context. This context may include the reason for the killing within the narrative/gameplay (e.g. self-defence, revenge; see Ali’ s work in Section 5.3; Hartmann et al. 2010), the availability of options as determined by the game mechanics (i.e. whether different outcomes are available to the player or whether actions and/or moral constraints are imposed on them qua their character; see Bartel 2015; Bartle 2008; Pohl 2008; Vanacker and Heider 2012; Zagal 2009), the level of violence and graphic realism (Barlett and Rodeheffer 2009; Krcmar et al. 2011; Wood et al. 2004; Zumbach et al. 2015) and so on. But the fact remains that all of these acts, in whatever context they are presented, in virtue of the gamer being able to choose to engage with them or not (even if ‘not’ ultimately means exiting the game), are permitted. It is this fact that makes all forms of unlawful killing part of a homogenous group: they are all enactments of something that is prohibited in the real world and yet permitted within the gameplay. In numerous other ways, they may differ, and this may impact on the player’s psychological and moral appraisal of the enactment (Sicart 2009). Nevertheless, it is their permissibility tout court that keeps them part of the same group; and where this group contains the intentional and unlawful killing of another person, as it does here, I will refer to these acts collectively as virtual murder. Having said that, in Section 5.2, I will consider the importance to the gamer’s dilemma of Patridge’s distinction between run-of-the-mill virtual murder and more extreme enactments, such as those described in games like Postal 2.

1.3 The Legal Status of Virtual Child Pornography in the USA and UK

Child pornography typically involves the sexualized image of a child (or children) which often includes the child engaged in some form of sexual activity. Where this is the case, the image amounts to a record of an actual event (in effect, a sexual assault) involving at least one actual child. In contrast, in the case of virtual child pornography, what is accepted is that the image of the child is computer generated, meaning that its creation did not involve an actual child, nor is it intentionally meant to represent a particular child, living or dead. Consequently, objections to virtual child pornography cannot appeal to any kind of argument based on abusive production (Sandin 2004). Given my interest in the gamer’s dilemma, the example of virtual child pornography I intend to focus on in this and further discussion (but not to the exclusion of other examples), is the virtual representation (qua computer-generated image) of a child engaged in sexual activity with an adult.

1.3.1 US Legislation

In the USA, the 1996 Child Pornography Prevention Act (CPPA) was the first attempt by the US Congress to respond to the digital era by alluding (rather than making explicit reference) to the virtual sexual imagery of children within its definition of child pornography. The new definition sought to criminalize not only that which depicts actual sexual activity involving a minor (in the case of the USA, someone below 18 years of age) but also that which appears to depict a minor engaging in sexual activity, or conveys the impression that a minor is involved (Bird 2011; Rogers 2009; Russell 2008). In 2002, however, a ruling by the US Supreme Court (in the case of Ashcroft v. Free Speech Coalition; based on a 6-3 decision), directly challenged the CPPA, claiming that aspects of the legislation were overbroad and therefore unconstitutional, insofar as they prevented freedom of expression (Kosse 2004; Mota 2002). Thus the US Supreme Court ruled that whilst “it remains illegal to make, show or possess sexually explicit pictures of children…[there is] no compelling reason to prohibit the manufacture or exhibition of pictures which merely appear to be of children” (Levy 2002, p. 319). Moreover, with regard to images of a purely digital origin – that do not involve any actual minors and therefore do not amount to a record of an actual crime – the Supreme Court ruled that as the US child pornography laws were implemented to prevent the victimization of children, and as there is no victim in cases of virtual child pornography, there is no compelling reason to restrict such freedom of expression (however, see Goldblatt 2012, for an attempted rebuttal of this claim). It is important to make clear, though, that the 2002 ruling did not affect the continued prohibition of ‘morphed’ images: namely, images of real children which have been integrated with some other image or in some way altered in order to create child pornography (Karnold 2000).

The Supreme Court did acknowledge that computer-generated images may lead to actual instances of child molestation, but they ruled that, at present, there is no evidence to suggest that a causal link between these images and actual abuse is anything other than contingent and indirect (Williams 2004). They reasoned as follows:

  1. 1.

    Virtual child pornography is not intrinsically related to child sexual abuse in the way actual child pornography is and so cannot be linked to any actual crime.

  2. 2.

    Any connection with actual child sexual abuse is indirect and contingent and so cannot be said necessarily to be connected to any future child abuse.

  3. 3.

    Prohibition of virtual child pornography cannot be based on the possibility that it will cause harm to some children.

In response to this ruling, in 2003, the US Congress introduced the PROTECT Act (which stands for Prosecutional Remedies and Other Tools to End the Exploitation of Children Today).Footnote 2 The PROTECT Act sought to clarify the overbroad nature of terms within the CPPA (like appears to be or conveys the impression) by seeking to prohibit virtual images that are indistinguishable from or virtually indistinguishable from actual images of children. The measure of whether a virtual image is indistinguishable from an actual image of a child (or virtually indistinguishable) is based on the extent to which an ordinary person is able to tell the difference between the two. The PROTECT Act does not therefore prohibit drawings, cartoons, sculptures and paintings of child sexual activity per se, given that such imagery is distinguishable to the average person. For the same reason, it does not criminalize (inter alia) plays and films such as Romeo and Juliet or Titanic or American Beauty, which depict adult performers appearing as minors engaged in sexual activity; thereby alleviating a previous criticism levelled at the original 1996 CPPA: that it was overly restrictive.

The PROTECT Act (section 1466A) does, however, limit the permissibility of such representations where they are considered to be obscene or ‘hard core’ (Bird 2011). In other words, regardless of their distinctiveness from any imagery of actual children and therefore regardless of the medium used (meaning that drawings, paintings and so on, are included), if a virtual image of a sexualized child or of a child involved in sexual activity is judged to be obscene, then it is deemed to be a form of child pornography subject to prosecution under the law. Indeed, as Kornegay (2006) notes: perhaps “an obscenity offence is the most appropriate way of proscribing content not produced with actual children” (p. 2167).

In the USA, obscenity is based on accepted contemporary community standards (the Miller test); basically, what a typical community would find obscene. What counts as obscene in the USA, then, is “not based on fact or policy, or harm done, but rather on a specific moral worldview” (Russell 2008, p. 1494).

To be obscene, as the law defines such a status, is to belong to a legal class of things, which varies over time and space. This is because attitudes and views about what is appropriate and offensive change over time in communities. (White 2006, p. 31)

Specifically, the law criminalizes:

…a visual depiction of any kind, including a “drawing, cartoon, sculpture or painting” that “depicts a minor engaging in sexually explicit conduct and is obscene” or “depicts an image that is, or appears to be, of a minor engaging in…sexual intercourse…and lacks serious literary, artistic, political, or scientific value” (18 USC §1466A) (Samenow 2012, p. 19).

Permitting a visual depiction that might otherwise be prohibited under an obscenity ruling as long as it is considered to be of serious literary, artistic, political or scientific value is likewise not without its problems. It is perhaps a matter for conjecture whether the following examples would or should fall foul of the PROTECT Act:

  • In the USA in the 1990s, the work of photographers Jock Sturges (e.g., The Last Days of Summer and Radiant Identities) and David Hamilton (e.g., The Age of Innocence), which typically involves nude adolescent models, were accused of violating child pornography legislation, although attempts to prosecute failed (Moehringer 1998).

  • In 2007, artists Zoe Hartnell and Sysperia Poppy created artworks for their online gallery, The King Has Fallen, depicting erotic dolls in what has been described as a Victorian “Gothic Lolita” style. After growing controversy over the depictions, the gallery was taken offline by the artists (Lichty 2009).

  • In 2009, the Tate Modern in London was embroiled in controversy when it decided to exhibit a piece by artist Richard Prince entitled Spiritual America (see Adler 1996). The artwork is a photograph of a photograph of actress Brooke Shields, aged 10. She is depicted naked with oiled skin and heavy make-up, staring directly at the camera in what has been described as a provocative pose. The photograph was displayed away from the other exhibits, behind a closed door, with a warning that some may find the artwork ‘challenging’ (Singh 2009).Footnote 3

1.3.2 UK Legislation

In the UK, even though it is accepted that sexual images of actual children and virtual children are not the same, the 2003 Sexual Offences Act and the 2009 Coroners and Justice Act in many respects treat them as if they are (See Ost 2010, for a detailed discussion). Under the UK law, no distinction is made regarding their criminality. As section 6A.1 of the Sexual Offences Act (SOA) states:

The SOA [Sexual Offences Act] 2003 makes amendments to the Protection of Children Act 1978 and the Criminal Justice Act 1988. It is now a crime to take, make, permit to take, distribute, show, possess, possess with intent to distribute, or to advertise indecent photographs or pseudo-photographs of any person below the age of 18 (emphasis added).

The Coroners and Justice Act (65:2) also broadens the definition of ‘image’ to include a moving or still image produced by any means. Pseudo-images and images produced by any means are therefore taken to include cartoons, drawings and computer-generated images (as well as other material) which depict, or appear to depict, a child (someone under the age of 18) engaged in some form of sexual activity (see also Section 84(7) of the Criminal Justice and Public Order Act 1994 which states that a ‘pseudo-photograph’ means an image, whether made by computer graphics or otherwise howsoever, which appears to be a photograph). Section 6A.3 of the SOA does acknowledge some difference between actual and pseudo-images, which should perhaps be reflected in sentencing, but at the same time recognizing the potential for equivalence under the law in more serious cases:

6A.3 Pseudo-photographs should generally be treated as less serious than real images. However, they can be just as serious as photographs of a real child, for example, where the imagery is particularly grotesque and beyond the scope of normal photography (emphasis added).

Given that no children are involved and therefore directly harmed in the production of virtual or pseudo-images, in the case of UK legislation, what is driving harsher sentencing, although not criminalization per se, is the degree to which the imagery is judged to be obscene (in this regard it is similar to the PROTECT Act). As Williams (2004) notes, in the case of virtual child pornography: “the criminal law is linked to the indecency of the image depicted and not to the harm suffered by the child” (p. 246).

The UK Obscene Publications Act 1959 determines something to be obscene:

[I]f its effect or…the effect of any one of its items is, if taken as a whole, such as to tend to deprave and corrupt persons who are likely, having regard to all the relevant circumstances, to read, see or hear the matter contained or embodied in it. (Section 1:1)

Therefore, rather than the measure of obscenity being rooted in some form of offence principle reflecting community standards (as in the USA), classification is based on whether the material is likely to deprave or corrupt those who have access to it. In other words, what is considered obscene is couched in social pathology such that there would be a tendency towards ‘moral and physical harm caused to vulnerable persons by exposure to obscene writings and images’ (Hunter et al. 1993, p. 138). Potentially, this could lead to what McGlynn and Rackley (2009) refer to as cultural harm (see Cappuccio 2012, for more detailed discussion on this issue; see, also, Section 5.1).Footnote 4

1.4 Some Closing Remarks

In bringing this chapter to a close, I would like to reiterate my earlier point that the focus of this book is on the morality of virtual enactments within video games, with particular emphasis on representations of murder and paedophilia, irrespective of their respective legal status. That said, there will be times when moral discussion may coincide with published legal argument and jurisprudence. Where this is the case, reference will be made to the similarity between the two accounts without seeking to use this similarity to convince the reader of the importance of the particular moral position. Without wishing to sound trite, it is my view that the merits of a moral argument should be determined by the quality of the argument itself, including, where applicable, the strength and validity of the evidence it may draw on, and not on the extent to which it aligns itself with a particular legal position. Nonetheless, it is not my intention to extol the virtues of this viewpoint while overlooking much of the good work that has been done debating the legality of virtual child pornography, and therefore ignoring the value to be had from drawing on legal argument to inform and illuminate moral debate.

In conclusion, although Section 1.3. provided only a rudimentary outline of some of the legal positions and arguments regarding the criminalization of virtual child pornography, hopefully, what has been made clear is that, hypothetically (given no commercially made games are yet available), where those players who engage in virtual paedophilia are adults, and do so willingly, much discussion has gone into whether such activity should be criminalized and, at present, as we have seen, different countries hold different views. Matters of legality aside, then, what is of interest and what will ground the discussion to come is how we might respond to the following questions:

  1. 1.

    Irrespective of whether it is legal to do so, is engaging in virtual paedophilia something that should be considered morally wrong?

  2. 2.

    In light one’s response to (1), is virtual murder liable to the same moral outcome for the same moral reason(s)?

How we answer these questions will likely determine whether the gamer’s dilemma can be resolved.