Introduction

There is a widespread intuition that playing computer gamesFootnote 1 is rightly subject to moral evaluation. Of course, saying this does not mean to pass a negative moral judgement on gaming as such or on certain genres that are commonly the target of media hysteria, such as first-person shooters.Footnote 2 However, it is quite easy to name examples of real computer games that even gamers will find morally appalling. A game that has become notorious in this context is RapeLay (Illusion 2006).Footnote 3 The main protagonist of the game is a serial sex offender and as such the player is supposed to virtually harass and rape a mother and her two daughters.

Let us call someone who believes that playing computer games can be immoral a moralist about computer games. The moralist must face what has been called the amoralist challenge (Patridge 2011). It is epitomized by the laconic statement “it’s just a game!”. The amoralist might elucidate his comment by pointing out that, after all, morality is about the real world, about things we really do. But within the virtual realm of computer games, there are only virtual crimes that have no real victims and produce no real harm. Even the amoralist can concede that things are different when real people are involved, e.g., when real players are being bullied in a multiplayer game. For single-player games, however, he will insist that if there is no real harm done, there can be no real moral responsibility. If the amoralist is right, then our moralist intuition about playing games such as RapeLay is the product of a category mistake. It consists in transferring the concept of morality, which is applicable to the real world only, to something which is not real but only a game.

The roots of this amoralist challenge to computer games lie in the general characteristics of all forms of play, characteristics that have been pointed out and discussed in Johan Huizinga’s Homo Ludens (1949). Play is “not ‘ordinary’ or ‘real’ life” (8). Rather, it consists in temporarily stepping outside of the ordinary into some sort of conceptually (and sometimes also spatially) demarcated “magic circle” (10) wherein “the laws and customs of ordinary life no longer count” (12). Thus, playing is a phenomenon sui generis, one that transcends ordinary life and the categories applicable to it, thus placing it beyond “good and evil” (6).Footnote 4 Of course, computer games were not known to Huizinga, but as games they do have the characteristics just mentioned; it is these characteristics that allow for the amoralist challenge to arise at all.

In the literature on the ethics of computer games, there are two types of moralist answers to the amoralist challenge. The first one is consequentialist and argues that, different from what the amoralist wants us to believe, there are real morally significant consequences that follow from playing certain games. The second type of moralist rebuttal to the amoralist challenge is expressivist. It sees the gamer’s involvement in certain types of games as an expression or a symptom of a moral flaw of character. In the following, I want to show that while both moralist strategies fall short of meeting the amoralist challenge, it is nevertheless possible to sustain the intuition that there is something morally wrong about games like RapeLay. I claim that we can save our moralist intuition if we locate the moral wrongness not in the activity of gaming but rather in the games themselves. The amoralist, I will argue, is right only to the extent that playing games is not in and of itself morally objectionable. By arguing for the so-called endorsement view, I will defend the position that a game can rightly be called immoral if it prompts us to transfer the abhorrent worldview of a fictional setting to the real world, thus rendering the amoralist slogan “It’s only a game!” false.

I will proceed in four steps. In the first two sections, I will take a look at the consequentialist and the expressivist answers to the amoralist challenge and argue that they both fail. In the third section, I will present and defend the so-called endorsement view about the morality of computer games. According to it, if a game endorses a certain worldview, it can rightly be subjected to moral evaluation. Finally, in the fourth section, I will turn to the question if the gamer has any moral obligations when playing games that are immoral according to the endorsement view. In my answer, I will discuss the special ontological role that gamers have in bringing games into existence. This role, I claim, is not sufficient to say that the moral responsibility of the gamer differs categorically from that of the recipient of classical representational arts such as movies or literature.Footnote 5

The consequentialist answer to the amoralist challenge

Let us begin by reconstructing the consequentialist argument against amoralism. The consequentialist line of thought comes mainly in two flavors, i.e., a utilitarian and a virtue ethical one.Footnote 6 Of course, virtue ethics is not in itself consequentialist like utilitarianism, but—as will become apparent below—the virtue ethical argument against amoralism rests upon the belief that playing certain games is bad because it has a negative effect on our character.

The utilitarian might try to show that playing computer games that contain violent or abusive gameplay will lead to real life actions that have an overall negative impact on the well-being of everyone involved. If playing shooters led to real mass killings, and playing racing games led to massive road rage, it might stand to reason that the harm resulting from gaming would outweigh its economic and recreational benefits. The problem with such an argument is that there is no empirical evidence for its premise that playing certain games causes violent real life actions.Footnote 7

Let us thus turn to the consequentialist argument proposed by virtue ethics. According to an influential paper by Matthew McCormick (2001), playing violent games is not wrong because it leads to harming other people, but because we harm ourselves. By engaging in virtual representations of immoral acts, we are accustoming ourselves to vicious behavior, thereby damaging our character.Footnote 8 Again, this is a hypothetical consequentialist claim. If playing certain games has the consequence of ruining one’s character, then this activity is to be condemned by a virtue ethical approach.Footnote 9 However, as for the utilitarian argument, there is no sufficient empirical evidence for the claim that gaming will alter our character for the worse. Surely, gaming can induce some short term emotional states, but that will not be enough for the virtue ethicist’s argument to go through.Footnote 10

Of course, this does not mean that there is not an important virtue ethical point to be made about engaging in virtual worlds and living well. However, the question central to virtue ethicists like Sicart (2009) and Søraker (2010) is not the question discussed here, i.e., whether virtual acts themselves can be morally good or bad, but rather what kind of player we should be if we want gaming to be a successful practice, one that is “conducive [...] to our well-being” (Søraker 2010, 12). Pointing out the empirical problem of a virtue ethical response to the amoralist challenge is not meant to diminish the valid question how we can be “virtuous players” (Sicart 2009, 92–94) who reflect on their gaming activity and enjoy games that further such reflection, thereby creating an ethically meaningful gaming experience. Even if one were to accept the amoralist view as defined in this paper, the ethical question concerned with virtuous gaming and living well would not be obsolete, because even if individual virtual actions are not straightforwardly immoral, it is an open question whether gaming (what and how we play) is in furtherance or hindrance of the good life.Footnote 11 That, however, would be the topic of a different paper.

Coming back to the question whether the consequentialist attempts to meet the amoralist challenge are successful, I believe that—the discussed empirical considerations apart—there is something else that is unsatisfactory about them. Even if there were better empirical evidence supporting the hypotheticals put forth by the utilitarian and the virtue ethicist, it would not be enough to dismantle the amoralist’s statement that games are just games. For what the amoralist seems to point to is that there is no inherent conceptual connection between the virtual actions of gaming and morality. This does not exclude the possibility that there might be non-virtual aspects of gaming that are rightly subject to moral assessment. What makes the amoralist position so disturbing is that it challenges our intuition that, e.g., virtual rape is in itself (i.e., regardless of the consequences) morally repugnant. The consequentialist arguments employed by the utilitarian and the virtue ethicist relocate the moral wrongness of virtual actions outside of the virtual realm, thereby effectively supporting the amoralist claim that there is nothing inherently wrong about virtually raping a non-player character. But this is exactly where our moralist intuitions seem to lie. It is hard to see how our moral indignation towards the fact that someone is carrying out virtual rape should be the result of our reasoning that this will lead to something morally bad in reality rather than it being something bad in itself as a virtual action.Footnote 12

The failure of the consequentialist answers to the amoralist challenge should compel us to rephrase the challenge itself. Since the amoralist might even concede that the effects of gaming can be rightly subject to moral judgement without giving up his core position, we should understand the latter as a thesis about the intrinsic connection between gaming and the possibility to evaluate it morally. The amoralist’s claim is the following: Gaming in itself, i.e., the mere carrying out of virtual actions, is never rightfully subject to moral judgement at all. If there is something morally problematic about playing certain games, it is only something that is extrinsic to gaming in and of itself.

The expressivist answer to the amoralist challenge

Because the consequentialist line of thought does not capture the intrinsic nature of our moral judgments about playing certain computer games, we need to look for a different, non-consequentialist account of our moralist intuition. Stephanie Patridge (2011) has proposed such a non-consequentialist answer to the amoralist challenge, one that can be called “expressivist”. It rejects the idea that playing certain games is bad because of the consequences. Instead, the expressivist’s main proposal is that playing certain games reveals something about our character, namely that it is morally flawed; or in Patridge’s own words (2011, p. 305): “There must be something wrong, antecedently, with anyone who would engage in such an activity for pleasure, independent of the consequences that might accrue to herself or others.”Footnote 13 She asks us to consider the following scenario in support of her thesis:

If you are skeptical of this point, I invite you to imagine what you would think of your friend should you find her coming out of the virtual reality suite announcing “I just had great time in there. You can even have sex with virtual children. But hey, no worries, they aren’t real.” (Patridge 2011, p. 305)

What is morally wrong in this case, according to Patridge, is revealed by the joyful remark of the player (“I had a great time!”) to something that, as a virtual representation of a heinous act, does not merit such a reaction. As Patridge (2011, p. 306) points out, “we should not, it seems, enjoy representations like that.”

I want to argue that Patridge’s argument is only valid if we suppose a certain sense of “to enjoy”. But this is not the only one possible. There is, I believe, another sense in which it is possible “to enjoy” the fictional representation of a morally degenerate action without any moral reproach. In short, I believe that “to enjoy” can be understood in a weak (or rather minimal), and a strong sense. The minimal sense of “to enjoy” just means to be captured or to be fascinated by something, while the strong sense of the word means to experience something as cheerful or fun. There are many games we enjoy playing without them being fun, even ones that are not suspected of being morally offensive. Take That Dragon, Cancer (Numinous Games 2016) for example.Footnote 14 In this game, we use point and click gameplay to explore the story of a family and its youngest child who has been diagnosed with cancer at the age of one. It’s a deeply moving, sad, and enriching experience. Yet, it would be a terribly misguided thing to say that it is a fun game.Footnote 15 If we apply the minimal sense of the word, it is far from clear why it should be a moral flaw to enjoy playing games that contain virtual representations of immoral actions. After all, people have watched Pier Paolo Pasolini’s Salò, or the 120 Days of Sodom (1975) with utter disgust and—like Austrian director Michael Haneke—ranked it nevertheless among the most important cinematic experiences of their lives.Footnote 16 Thus, they must have enjoyed it in the minimal sense of the word, even if our ordinary use of “to enjoy”, which tends towards the strong meaning, would probably prohibit us from saying that “we enjoyed Salò”.Footnote 17

Surely, in order for someone to enjoy a piece of fiction in the minimal sense of the word, he must take a certain interpretive stance towards it. Grant Tavinor (2009, p. 161) has argued (with reference to the works of Noël Carroll) that this means that we may be forced to “adopt systems of beliefs or affective attitudes [...] of a morally questionable nature”Footnote 18 if we are to understand a fictional world. This seems correct, but only if “to adopt” in the above quote means something like “to entertain an idea in a strictly as if mode of thinking”. Clearly, in order to understand a story of the inner workings of the Mafia, like in the game Mafia (Illusion Softworks 2002), we must use our imagination to try to understand the mobsters’ ethical system (or lack thereof). Such efforts of the imagination might even teach us important ethical lessons about our own real life values.Footnote 19 However, this does in no way imply that we have to adopt the values of the Mob for our own lives beyond the realm of make-believe. We might, of course, choose to do so, and this choice would rightly be subject to moral evaluation. Yet, there is no intrinsic connection between imaginatively engaging in the moral set-up of a fictional game world and adopting it for one’s own life.

Now, if we acknowledge that we might enjoy things in the minimal sense of the word without enjoying them in the strong sense, then we must reject the expressivist argument against amoralism. Let us take a closer look at the expressivist answer to the amoralist challenge to see why this is so.

The expressivist argues that:

  1. 1.

    Playing a computer game necessarily means enjoying it.

And that:

  1. 2.

    Enjoying representations of immoral acts is immoral.

Therefore:

  1. 3.

    It is immoral to play games that involve representations of immoral acts.

The argument does not go through because of the double sense of “to enjoy”. The first premise is true only for the weak sense of “to enjoy”. If someone voluntary plays a game in his free time, we can legitimately suppose that he enjoys it in the minimal sense, but have no right to suppose anything more substantial about the specific nature of his enjoyment. The second premise on the other hand is only acceptable if we understand “to enjoy” in the strong sense of “to have fun”.

To be clear, the point is not that there is no sound expressivist argument against enjoying certain games (like RapeLay) if one uses “to enjoy” in the strong sense of the word; there clearly is. Therefore, I believe that a charitable reading of expressivists should ascribe to them the use of “to enjoy” in a strong sense.Footnote 20 The above argument is thus not aimed against expressivism as such, but its purpose is to point out that the reasonable expressivist point against enjoying certain games (in the strong sense of “to enjoy”) is limited insofar as it cannot be used in an argument against amoralism as such. To put it differently, the point of the above argument is that we cannot assume that playing a game necessarily means enjoying it in the strong sense, and with only the latter being (sometimes) morally problematic, the moral fault cannot lie in the act of gaming as such.Footnote 21 If playing certain games is morally wrong, then it is wrong because of the specific mindset or mood of the player. However, there is no conceptual connection between playing something and playing it in a certain mood or with a certain mindset. Acting is a fitting analogy to drive this point home. Should we suppose that an actor playing a murderer, a sex offender, or a war criminal is actually having morally inappropriate reactions to his fictional deeds? Of course, we should not. We may very well suppose that the actor enjoys the pretense as such, but from that we may not infer that, e.g., he thinks it was “really fun” to pretend to murder or rape. Playing a game is of course very different from acting in many ways, but it is similar insofar it consists of actions that are merely pretense. If an actor can play a bad guy without us presuming that he is doing so with a morally inappropriate inner reaction or attitude, we should not treat the gamer, who is in charge of the virtual actions of a virtual bad guy, any differently. The amoralist is therefore correct insofar as the act of gaming is concerned. It is never in itself wrong to play a game. Yet, it might be wrong to play a certain game with a certain attitude or mindset. Let us take a closer look at this possibility.

The endorsement view

The observation that there seem to be certain fictional representations we should not enjoy makes sense only if we talk about enjoyment roughly along the lines of “cheerful fun”. But even if this intuition, relying on the strong sense of “to enjoy”, is more convincing, the question remains why we should not have a good time playing certain games. Let us look at a recent example from the world of computer games. In Hatred (Destructive Creations 2015), our goal as players is to engage in a mass shooting of unarmed civilians. We can regain our health only by triggering brutal execution scenes with our innocent victims begging for their lives. The moralist claims that we are right to morally condemn anyone whose reaction to playing Hatred is one of joyful enchantment, but wherein exactly lies the moral fault of the player?

The amoralist will insist that if Hatred is just a game, there is no reason to make any moral demands on the enjoyment of something that is, after all, not real but only virtual. He might even concede that there are adequate and inadequate reactions to certain game content but still deny that these questions of adequacy are of a moral nature. The amoralist might say for games what could be said for movies and literature as well: Whoever laughs at tragedy and is shaken by comedy has an insufficient grasp of those genres. He thus shows a lack of aesthetic judgement but does not make a moral mistake.

I believe this line of amoralist reasoning should be rejected because it implausibly severs all ties between the fictional and the non-fictional. First of all, there is a rather strong intuitive pull towards the idea that our emotional reactions to fictional representations are themselves not only fictional but real and therefore rightly subject to moral (and not only aesthetic) evaluation.Footnote 22 As Christopher Bartel (2015, p. 291 f.) has suggested, we at least seem to have the moral responsibility that our emotional reactions to certain fictional representations are not real-life endorsements of them. The second point—the one I would like to focus on in the following—is concerned not with the endorsing attitudes that players may have towards certain representations but with the endorsement character of games themselves.Footnote 23 Sometimes a piece of fiction is not merely fictional because, on a pragmatic level, it also endorses a normative view about the real world.Footnote 24 As such endorsements, pieces of fiction can be subject to moral evaluation. Let us call this the endorsement view. It finds support in our treatment of propaganda art. The German film Jew Süß (Harlan 1940) is a prime example. It has been praised for its aesthetic qualities by the likes of Michelangelo Antonioni (Friedländer 2006, p. 126). Yet, no one in his right mind would doubt that it promotes anti-Semitic prejudice and resentment. A movie is thus not necessarily just a movie, but it can also be an endorsement of a morally reprehensible worldview. The same goes for certain games. KZ Manager (The Missionaris 1989), originally programmed for the Commodore 64, is a resource management game; its main objective is to optimize the murderous activities of a Nazi death camp. Its ludic qualities (or rather its lack thereof) aside, KZ Manager is not just a game, in the same way that Jew Süß is not just a movie. It glorifies and thereby endorses the real mass murder of Jews.Footnote 25

The endorsement view differs significantly from a theory that was proposed by Patridge and that might look similar at first glance. According to Patridge (2011), games are morally objectionable if they contain representations that have an objectively offensive meaning within a given society. Because these meanings are specific to a society, they can be called “social meanings”, and because they cannot be “overturned” by individual intentions, they can be called “incorrigible”.Footnote 26 Patridge (2011, p. 304) gives us the following example of such an “incorrigible social meaning”: “Let us consider a cartoon image of Obama eating watermelon. In the context of the contemporary United States, such an image is properly interpreted as a racial insult.” It is a contingent social fact that the depiction of African–Americans eating watermelons has become an icon of the racist notion that blacks are easy to please half-wits who need not freedom but only a watermelon to be happy. Despite its contingent character, this imagery—if used in the United States—is offensive to blacks and objectively so. After all, it is not up to an individual to choose whether the imagery in question really “targets groups of individuals” or not.

I believe that Patridge’s view is flawed. The use of representations with an incorrigible, morally offensive social meaning can indeed be a reason to judge that a game is immoral. However, culturally insensitive imagery is neither necessary nor sufficient for the immorality of a game. Consider Hatred once more. We are right in objecting to this game because it glorifies the indiscriminate murder of innocent people. We are led to believe that it endorses the morally abhorrent worldview that it is somehow desirable to murder because there is no ludic or narrative element that would thwart its main gameplay mechanic of merciless and brutal mass murder. No reference to culturally insensitive imagery is needed to understand why Hatred is morally degenerate.

Also, culturally insensitive imagery might be included in a game without automatically rendering it immoral. Patridge is right in saying that no mere mention of the amoralist motto “it’s just a game” will rid certain imagery of its meaning. However, any such imagery alongside with any single virtual action can be subverted or modified by the overall narrative context of the game, or certain gameplay mechanics. In the roleplaying game series The Witcher, we explore a fantasy world that is ridden with discrimination, prejudice and racism. While playing the main protagonist Geralt of Rivia in the latest installment of the series, The Witcher 3: Wild Hunt, we are constantly confronted with discriminatory comments by non-player characters who call Geralt a “mutant” or a “freak” and make it clear that as a witcher he counts less than a human. Now, I believe that the insults that Geralt faces can be said to have a morally condemnable, incorrigible social meaning as construed by Patridge. Does this in itself make The Witcher 3: Wild Hunt (CD Projekt RED 2015) a game that endorses discrimination and/or racism? Such an allegation seems very far-fetched, because even though discrimination and racism are part of the fictional world of The Witcher 3, it is clearly not embraced by the main protagonist Geralt of Rivia. To the contrary, playing as and identifying with Geralt, the player himself experiences the hurtful nature of (racial) discrimination. Also, some of the main quests revolve around fighting racist institutions like the witch hunters who are after Geralt’s friend Triss Merigold for the racist reason that she is a sorceress.Footnote 27 What this example is supposed to illustrate is that not everything that is featured in a piece of fiction—be it a game, a movie, or a piece of literature—is something that is endorsed by it. Representation and endorsement have to be kept apart. Otherwise we would have to claim that because Schindler’s List (Spielberg 1993) contains representations of Nazis murdering Jews, it endorses the Holocaust.Footnote 28

The latest installment of the Grand Theft Auto (GTA) series, GTA V (Rockstar North 2013), is another great example for the need of a distinction between representation and endorsement. In this game, the player can switch between three playable protagonists, all of them criminals with a morally dubious character. One of them, Trevor, is even plainly sociopathic. In GTA V, the player may play scripted missions that advance the main story line, or he may just explore the huge game world without any objectives. Freely roaming the fictionalized version of Los Angeles, Los Santos, the player can choose to—but does not have to—commit immoral acts, e.g., he can run over pedestrians, shoot police officers, or beat up drivers and steal their vehicles. In the story mode, however, the player is forced to commit theft, murder, and even torture if he wants to beat the game. In a now infamous scene, while playing as the sociopath Trevor, one of the many competing government intelligence agencies pressures our character into repeatedly torturing a man suspected of terrorist ties. From this description, it should become clear that the fictional world of GTA V is characterized by a certain morality or rather immorality. In the world of GTA V, the playable protagonists, and many of the shady characters and government agencies they encounter believe that murder, theft, and torture are acceptable (or even fun, in the case of Trevor) if they are necessary to advance one’s own personal interests. It is safe to say that GTA V is the representation of an immoral world.Footnote 29 However, it would be wrong to infer from this that the game is also the endorsement of such an immoral world. Rather, anyone who has actually played the game will attest to its dominant satirical character. Whether it is the incompetent, ruthless, and power hungry government agencies competing against each other, or the shallow, greedy, and fame and sex crazed inhabitants of Los Santos, the world of GTA V can easily be identified as a satirically exaggerated version of our world, the real one.Footnote 30 But as satire GTA V does not endorse what it portrays, rather it ridicules it.

Therefore, it is misleading to take certain gameplay sequence like the torture scene in GTA V in order judge the moral character of the game.Footnote 31 Not only can we not infer any moral fault of the player by simply witnessing him playing a certain scene out of context, but we cannot even judge the morality of that piece of gameplay, let alone the game in its entirety. This, however, does not mean that we cannot judge a single gameplay scene to be immoral. We can. It is just that its immorality is only to be established after having considered the entire game and the context it provides. Games are holistic entities. By “holistic” I simply mean that we can only discover the meaning of an element of a game, i.e., make sense of it, with regard to the game as a whole. The meaning of a game (or a piece of literature or movie) is what we try to uncover when we ask questions like: What is it really about? What are the central ideas that are being expressed or explored? Has it any bearing on the reality of our lives? Does it promote a certain worldview? The endorsement of a certain normative or evaluative stance is thus part of such a meaning. The point about the holistic nature of games relevant to our question is therefore a fairly simple one: If we want to talk about the moral character of a game or a certain piece of gameplay, i.e., find out its moral meaning, we must take a look at the game in its entirety instead of just going by the (seemingly) morally offensive nature of an isolated representation.Footnote 32

Take the infamous airport scene form Call of Duty: Modern Warfare 2 (Infinity Ward 2009) as an example. In the mission “No Russian” of the single-player campaign, the player controls a US army ranger who has gone undercover and infiltrated a nationalist Russian terror group. The group launches a savage terror attack on a Moscow airport and the player may choose for his character to actively participate in the killing of innocent civilians at the airport, or he may abstain from the shooting.Footnote 33 No matter what the player does, at the end of the attack, the terrorists kill his character, revealing that they had known about his real identity all along. I believe that this massacre scene may rightly be called immoral because it fails to supply what Tavinor (2009, p. 165) calls “redeeming or mitigating context” for its representation of immoral actions. In the case of computer games such redeeming or mitigating context consists of narrative or ludic reasons for the representations of immoral actions. If there are no such reasons, we are led to believe that the game actually wants to endorse the represented content. This is exactly the case in said mission of Modern Warfare 2. There are simply no good narrative reasons for this scene. The game is set up as a conventional story about good counter-terrorists against evil terrorists. If one were to follow the logic of the story, the player’s character should actually attempt to stop and kill the terrorists right there at the airport. Nothing in the narrative or the gameplay points towards the perspective of the victims of the terror attack.Footnote 34 There is no ethical reflection coming from the player’s character, no signs of inner conflict or disgust that could have easily been introduced through some voice-overs. Also, there is no satirical backdrop to the game or this scene. Not only is the massacre scene not redeemed or mitigated by the context, it is actually in stark conflict with the context provided by the fictional game world. Because it is not a credible part of the game world, we are led to believe that is an endorsement of something that transcends the confines of the fictional realm.

Before moving on to the last section, I want to briefly address two critical questions that may arise at this point. The first one simply asks why we should think that the endorsement of an immoral worldview is morally wrong. The answer is twofold. First of all, endorsing an immoral worldview is itself something actual, i.e., it is beyond the virtual realm of what is represented in a game. Thus, the endorsement view overcomes the problem that this paper is concerned with, namely the amoralist challenge towards gaming. Now, one could of course continue to ask what makes the actual and not merely virtual act of endorsing an immoral worldview morally bad. Indeed, it is beyond the limits of this paper to try to trace back to its underlying principle the moral wrongness of such an endorsement. Yet, I am skeptical about the necessity of such an endeavor, because I believe it to be a rather uncontroversial, commonsensical claim that, e.g., if someone were to endorse slavery, he should be condemned morally. For clearly, the actual support of the (re-)establishment of a grave injustice in the real world is rightly subject to moral criticism.

The second question has to do with the way I have described the endorsement view, namely that games themselves can be said to endorse a certain worldview. This might seem strange because games are cultural products that have certain meanings—i.e., they represent, express, explore or endorse certain ideas—only because there are human subjects that have created these games and thereby bestowed this meaning on them. Is it strictly speaking not the game developers behind games like RapeLay, Hatred or KZ Manager, and not the games themselves, who endorse a morally problematic worldview? It is certainly correct to say that without game developers there would be no games, but it is also the case that once games are created they speak for themselves. By that I mean that we can find out what kind of worldview (if any) is endorsed by a game simply by analyzing it. We do not need to talk to the developers. Once a game has been published, the developers have no special authority in determining its meaning. Imagine the developers of Hatred claiming that they did not intend the game to be a glorification of murder. Even if we were to believe this claim, this would not change anything about the fact that Hatred indeed does glorify murder, which is easily confirmed simply by looking at the actual gameplay. Because games speak for themselves in the way just outlined, I believe it is justifiable to speak about games as such being endorsements of certain worldviews.

The gamer’s moral obligation

So far I have argued for two things: first that the amoralist is right in claiming that the activity of playing a game cannot in itself be morally wrong, and second that he is wrong in claiming that games are not subject to moral evaluation because they are “just games”. Games are not “just games” in a twofold sense. First of all, they may contain representations that require certain emotional reactions or attitudes on the side of the player. Our reactions to representations of immoral acts should not be ones that express their endorsement in real-life. Second of all, and this has been at the center of the so called “endorsement view” discussed in the previous section, games themselves sometimes transcend the boundaries of the fictional by endorsing a morally problematic worldview. As such endorsements, games do warrant certain reactions by the player, which in turn can rightly be morally evaluated. Herein lies a partial truth of the expressivist argument discussed above. It may not be able to meet the amoralist challenge about gaming, but when it is combined with the endorsement view about games, it tells us something about how certain games must be played if the player is to avoid moral reproach. Even though it is not intrinsically bad to play games like Hatred or RapeLay, they should not only not be consumed as fun games, but they should also be treated with moral contempt; at least in the case of adult players, the lack of disapproval for something morally blameworthy may not be excused as mere moral immaturity, for this kind of immaturity is itself a moral flaw of character. It may be legitimate to play such games—e.g. as part of the research for an article about the ethics of computer games—but only if the player is aware and critical of the moral status of the game.

The endorsement view about games can be used for other forms of representational art as well, especially movies and literature. If a movie or a piece of literature endorses a certain worldview, its moral character is to be judged by the very content of its endorsement. Whoever consumes (reads, watches, plays) such a movie, piece of literature, or game, is morally responsible for reacting adequately to the moral status of the object he consumes. But there seems to be something unsatisfying about treating games on a par with movies and literature. After all, to many it seems worse—at least prima facie—to play a game like RapeLay than to watch an equally immoral movie. A likely first shot at an explication might be that playing is worse than watching because watching is passive while playing is active. But since we have already argued that the amoralist is right in claiming that no virtual action can be in and of itself immoral, we must look for another explanation why activity strikes us as being worse than passivity when it comes to immoral fictional content.

It might be argued that the basis for our intuition that playing an immoral game is somehow more problematic than watching an immoral movie lies in the peculiar ontological status of games. As Sicart (2009) has pointed out, games are ontologically incomplete objects as long as they are not actually being played by someone. Computer games consist in a set of rules (captured and enforced by the programming code). These rules determine the game world, which is presented to the player by the graphics and sounds of the game, and its basic gameplay mechanics, i.e., the possible ways of interacting with the game world (Sicart 2009, p. 28 ff.). But as such the game is still ontologically incomplete for it needs the player to be actualized. There is no actual gameplay without the activity of the player (Sicart 2009, p. 54 f.). It would thus seem that by playing an immoral game the player is complicit in bringing into existence an entity that endorses an immoral worldview.Footnote 35

Now, one might argue that games are not unique in this respect because movies and literature also rely on an activity on the part of the recipient. After all, reading and watching are things that we do. Kendall Walton (1990) has famously argued for this point, claiming that the representational arts are all based on the activity of make-believe. There are movies that reflect on this active complicity of the viewer. Michael Haneke’s Funny Games (1997), e.g., has several scenes in which the two sadistic killers turn ad spectatores, e.g., by winking at the camera, thereby signaling to the audience that—by watching—it is actively in on the crimes.

The difference between games, and movies and literature therefore cannot be cast as one between co-creating morally relevant content and passively consuming it. However, there are two other ways in which one might try to spell out the idea that the player is more active than the ordinary recipient of the representational arts. The first one makes an analogy between player and actor, the second one works with an analogy between player and screenwriter.

Let us first consider the player-actor analogy. Of course, there are many differences between actors and players. The analogy that I am concerned with here lies only in the additional kind of activity that separates both playing and acting from watching and reading: Both the player and the actor carry out pretend actions, the actor on stage or on set, and the player in the virtual realm. Is this additional kind of activity morally relevant? When it comes to moral evaluation, we do judge actors more severely than those who watch them. Again, Jew Süß is a good case in point. Intuitively it seems like the actors play a more fundamental ontological role in bringing a movie into existence than the viewers do. Without the actors, no viewer could participate in the make-believe of the movie. Even if the viewer is mistaken in believing that he is in no way responsible for bringing the movie he is watching into existence, the actor’s activity is more fundamental for its ontological constitution. Therefore, it seems, that the actor’s moral responsibility for bringing something morally objectionable into existence has to be more fundamental as well. The reason why we believe that the actor’s moral responsibility is more fundamental than that of the viewer is that the actor is responsible for presenting the endorsement of an immoral worldview to others. Acting is an inherently public activity. By making a public appearance in a movie like Jew Süß, the actor tacitly consents to exposing the audience to the endorsement of an immoral worldview. The moral fault of the actor, I believe, lies in this willing exposure of others to an immoral endorsement. It is at this point that the analogy with playing games breaks down. Since gaming is not inherently public (even though it can be, e.g., if someone is streaming a game on Twitch), we cannot judge gamers who play an immoral game in the same way we judge actors who are part of an immoral movie.

Let us turn to the second analogy, the one between actor and screenwriter. One might have objected to our first analogy by pointing out that actors usually just follow the script and have no room to decide which actions to take, while at least in some games we seem to be able to decide how to act or even what kind of character we want to be.Footnote 36 Is the player therefore (at least sometimes) not more of a screenwriter than an actor? I think the answer has to be “no”. Our decisions as players are themselves part of an utterly deterministic virtual world which is determined by the developers and screenwriters behind the game.Footnote 37 Players are not actively creating the game the way a screenwriter is creating a screenplay. They are always following a predetermined path and their choices are only about which predetermined path they want to follow. Therefore, they can never change the fundamental character of the game. They cannot make an endorsement of a morally problematic worldview stop being what it is through their virtual actions.

The upshot of the discussion of these two analogies is that the player-actor analogy does not yield a reason why we should judge players more severely than recipients of the representational arts for consuming immoral works of fiction. The player-screenwriter analogy breaks down completely, for players do not share a screenwriter’s creative freedom and the special moral obligations that might come along with it.Footnote 38 I believe we should therefore acknowledge that there is no reason why playing immoral games would be morally worse than watching immoral movies or reading immoral books. There is thus no special moral obligation that would pertain only to gamers but not to readers or moviegoers.

Conclusion

I have defended gaming against charges of immorality by agreeing with the amoralist claim that the activity of gaming is in and of itself amoral. This defense does not, however, amount to a sweeping moral exoneration of games and gamers. I have argued that games themselves can be immoral if they are endorsements of an immoral worldview, and that gamers have the moral obligation to react adequately to the moral status of game. An approving and even a missing disapproving emotional and/or cognitive reaction to an immoral game is itself immoral. At the end of my paper, I examined the question whether the gamer’s moral obligation does differ categorically from the moral obligation of the recipient of the representational arts. It turned out that it does not.