Monday, February 11, 2008

The Torture Game

More recycled content, technically the second half of the last post. If you're not going to skip this, read that one first.

Four recent, commercial games have directly dealt with the issue of torture: The Punisher, State of Emergency 2, The Godfather and Reservoir Dogs. This list is not exhaustive, but these titles demonstrate some of the ways torture has been approached in existing games. Of these four titles, The Punisher is the most explicit, and is the central subject of my investigation. As such it receives the most attention, but all four offer useful insight on the subject.

The Punisher, it must be noted, is not merely a game, but part of a multimedia franchise. Originating as a villain in an issue of Spider-Man, the character known as Frank Castle—alias The Punisher—has been a persistent figure in the Marvel Comics universe for thirty years. Volition's videogame adaptation of The Punisher was released in 2004 to coincide with the theatrical release of the film of the same name. Both the film and game adaptations drew heavily on the work of Garth Ennis, who had recently revitalized interest in the character among comic readers. Ennis' particular take on The Punisher is substantially more complex than the simple-minded vigilante previous writers had crafted, and the Punisher videogame is so thoroughly steeped in the work of Ennis that it cannot be read in isolation from that work. Panels from Ennis' books provide a substantial part of the game's reward system, and serve as indexes, pointing to the larger narrative of which the game is a part. That narrative guides the game mechanics, and the game's ethical framework compels the player to kill in a variety of ways, none of which should be unfamiliar, symbolically or mechanically, to any action game enthusiast. What is comparatively new is The Punisher's treatment of torture.

The Punisher's so-called “torture engine” is a mini-game of sorts. Frank puts his victims in a dangerous, frightening and/or painful situation that is not immediately lethal, and he must keep them sufficiently intimidated without killing them. The controls vary with every method of torture, but all rely on subtle manipulation of an analog stick. At first glance, torture appears to function as an interrogation technique. Certain characters possess special information that can only be extracted through torture. However, this information is never essential to Frank's mission, but only supplementary: a skilled player can easily get by without it. Moreover, very few characters have any useful information to be extracted, yet nearly all can be tortured. In spite of torture's lack of value for interrogatory purposes, it is nevertheless a crucial play mechanic, and players cannot easily avoid engaging in it.

The Punisher is not an open-ended play-space like Second Life, and players are not expected to do things merely because they can. Rather, the game encourages torture (makes it "ethical") by connecting it to two incentives: the acquisition of points, and the unlocking of hidden content. Points feed directly back into the gameplay experience, as players exchange them for skill and weapon upgrades. Scripted, location-sensitive tortures provide the largest point bonuses, but any enemy character within grabbing distance can be exploited for this purpose, and an execution is never as profitable, in terms of points, as an execution preceded by torture. In addition to the points, torture will randomly cause Frank to have flashbacks. These flashbacks are presented to the player as a panel of comic art from Ennis' Punisher stories accompanied by an appropriate voice sample; for example, an image of Frank holding a dead family member juxtaposed with a terrified criminal screaming “I have a family!” These flashbacks, once unlocked in the main game, can be viewed from the title menu, and contribute to overall completion of the game, much like the side-quests in the recent Grand Theft Auto games. For the player, the reward for the (frequently challenging) act of torture is non-diegetic. Points have no meaning at the narrative level, and it's unclear why Frank would want to suffer flashbacks to painful moments in his life. Thus, in terms of the game's internal world, it would be tempting to refer back to George Orwell's 1984: “The purpose of torture is torture.” More accurately, though, the purpose of torture, in The Punisher, is a “bonus round” of sorts, a chance to allow the player to demonstrate skill in exchange for points. If torture is a “mini-game,” it is easy enough to “fail” by accidentally killing the victim. The player loses points for killing a victim in the course of torture, even though he or she would gain points for killing the same person in a more conventional fashion. The game takes no notice whether or not the victim has given Frank whatever information they have. The rules are simply that killing is rewarded, torture is rewarded, but accidental killing during torture is punished. These are the ethics of torture in The Punisher, and they make sense at a purely mechanical level. At a narrative level, they are internally inconsistent, and thus the narrative and ethics cannot be integrated into a moral argument about torture.

State of Emergency 2 is the little-known sequel to the controversial State of Emergency, which places players in violent street combat against a fascistic corporate dictatorship. The original game incorporates contemporary political debates about globalization into its narrative, but squanders its potential for legitimate discourse through simple-minded play mechanics.

The sequel adopts a more linear, story-based approach to revolution that includes a mini-game in which players interrogate suspects. The interrogator is “Spanky,” a former gang member and Hispanic stereotype, and the interrogation consists of repeatedly punching a captive. In terms of play mechanics, interrogation is a timing game, in which players must hold the proper button and release it at the proper time—release the button too early and Spanky will not punch hard enough to cause sufficient pain, release the button too late and Spanky will punch too hard and kill the captive. In contrast to the calculated brutality of the torture seen in The Punisher, the State of Emergency torture scenes are somewhat cartoonish. The famously graphic violence of the original State of Emergency, which allows players to blast non-player characters (NPCs) apart with explosives and then use the charred body parts as weapons, has been toned down considerably in the sequel, and one wonders why torture was included at all if gratuitous violence were a concern. As it stands, the torture scenes are among the least violent and disturbing action scenes in the game.

The Godfather is the high-profile videogame adaptation of the world described in the Mario Puzo novel and Francis Ford Coppola films. Though not explicitly mirroring the plot of the novel or films—the protagonist is a new character not found in either—the ubiquity of The Godfather in popular culture makes it unlikely that players will come to the game unfamiliar with the Corleone dynasty. As with The Punisher, the game narrative must be read in context of the larger text of which it is a part.

Intimidation is a major factor in the gameplay of The Godfather. The most common use of intimidation is against shopkeepers, to encourage them to hand over protection money. Unlike the previous examples, the player need not resort to physical pain for this purpose, although the game allows a great deal of realistic physical violence. If a shopkeeper is being particularly stubborn in his refusal to pay, smashing his cash register might be more effective than choking him or shooting him in the kneecap. Simply placing someone in your gunsights for several seconds will often do the trick. Consistent with the gangster ethics detailed in the novel and films, the game engine generally rewards players for finding ways to intimidate without resorting to direct bodily harm.

Finally, Reservoir Dogs is the videogame adaptation of the 1994 Quentin Tarantino film of the same name. Similar to The Godfather, torture is used not for interrogation, but rather for intimidation. Though the game gives players the option of blasting their way through all obstacles, earning a “Psychopath” rating in the process, the more cerebral “Professional” track requires a more measured use of violence, both threatened and enacted. Taking human shields, and therefore threatening hostages with lethal violence, is sufficient to disarm security guards, but will result in a standoff with actual police. Police will also drop their weapons, however, if the player pistol-whips the hostage in front of them—but even this is ineffective against large numbers of police. When surrounded, players who have charged up the avatar's “adrenaline” can perform a “signature” move, beating the hostage into unconsciousness and likely disfiguring him or her in the process.

These “signature” moves are unique to each character, from Mr. Blue's cigar to Mr. Blonde's trademark straight razor, though the most brutal violence happens off-screen. A “signature” move will make every cop in the vicinity lay down their weapons in surrender. The game's ethics, in this case, cannot possibly be developed into a moral argument, simply because the they make no sense whatsoever at the narrative level. Beating and disfiguring a civilian should, logically, make the character more likely to be shot by police, not less. In addition, unconscious hostages drop to the ground and cannot be picked up. Thus, by performing a “signature” move, the protagonist reveals to the police that he is violent, unpredictable and dangerous, while simultaneously releasing his human shield. The torture techniques described by Mr. White in the film, or enacted by Mr. Blonde, would have made some degree of sense in terms of the narrative, but the torture found in the game, while superficially similar, does not.

In all these games, some common elements exist. First, the games' ethics, which compel the player to torture, are not explicitly out of sync with the protagonists' motivations. From the protagonists' perspective, torture is justified by the moral “gray area” of the situations in which they find themselves, be it organized crime, insurrection, or vigilantism. We are given no reason to believe that the protagonists themselves believe torture to be immoral, at least under the given circumstances. It is worth noting that three of the games I've discussed, The Punisher, The Godfather and Reservoir Dogs, are adaptations of existing works, and each inherits a nuanced morality of violence from the worlds' origins in film, novels and comic books. The player is not called upon to accept or reject the protagonist's actions as moral, and the circumstances in which the protagonists find themselves are defined as extraordinary and largely unrelated to “real life.”

Second, the morality of torturing an innocent is never addressed. The Punisher cannot torture an innocent person who was simply in the wrong place at the wrong time, because these people do not exist in the game. (Innocents exist, but they are clearly marked, and the player cannot make Frank torture them.) In The Godfather and Reservoir Dogs, the player is an anti-hero at best, but there are no judgments on when it is moral to torture, just when it is ethical in terms of gameplay.

Third, when torture is applied for the purpose of interrogation, it is universally effective. The tortured party will invariably “crack,” given the right circumstances. When they do, they will invariably give the protagonist correct information.

Fourth, the actions of the player have no long-term effect on the overall “war effort.” It is hard to imagine how it could, given the genres in which it takes place. The mafia and the fascist thugs of the games in question are not in a position to become more brutal due to the avatar's actions.

Fifth, the experience of having intentionally inflicted pain on a defenseless human being has no long-term effect on the mental health of the protagonist. Again, this is to be expected, since the modeling of avatar's mental states is still very rare in videogames. (Silicon Knights, of Blood Omen fame, made some progress on this front with Eternal Darkness, albeit in a less "serious" supernatural fashion.)

These games clearly demonstrate that videogame designers have developed the conceptual tools necessary to model the act of torture, but not its consequences. By carefully integrating the rule system and narrative, and by explicitly addressing those elements found lacking in the games I've described, it is possible to design videogames that make coherent moral arguments about, and more specifically against, torture in a way that would not be possible in any other medium. I here propose a model for such a game.

The best genre for such a game would be a single-player strategy game that alternates between macro-management and micro-management, similar to Microprose's X-COM: UFO Defense. Time will need to be somewhat fluid in the game, which would suggest a turn-based approach, but there's no reason parts of the game couldn't be designed for real-time strategy. The player commands a military unit in occupied territory, under constant threat of attack from local guerrilla forces. To prepare for or prevent these attacks, the player must gather information, make arrests, interrogate suspects, and use the new information to coordinate attacks or make more arrests. Like X-COM, gameplay will be cyclical in nature, and will end when either the guerrillas successfully wipe out the player's unit, or when public support for the guerrillas wanes and order is restored. These are only end conditions, however—it might be necessary, depending on the argument the designers seek to make, for true, non-diegetic victory to be independent of military success. Most importantly, the morality espoused in the narrative must be consistent with the ethics of gameplay.

As the game begins, players are given some initial intelligence from a variety of sources concerning planned attacks, and suggesting suspects. Players must then travel to a given location and attempt to arrest a suspect, using a minimum of force. After all, killing a suspect before he can make himself useful is a failure at both military and moral levels. Assuming the suspect can be arrested and returned to base successfully, the interrogation phase begins.

The interrogation process is the most significant portion of the game. Consequently, the game rules must acknowledge the issues ignored by the games I've discussed. The rule system, after all, will determine the ethics of gameplay, compelling gamers to play in a certain way, and the narrative cannot be allowed to disconnect from these ethics. Thus, characters must express differing opinions on the morality of torture in general. Establishing the opinions of NPCs can be handled in a number of ways, and designers need not resort to overlong cutscenes, but they will need, at the very least, well-written dialogue that is both semi-random and likely to be encountered by players. In addition, the game must include the possibility of bad intelligence, and it must be possible, even likely, for players to make false arrests. Whether or not the suspects actually know anything, many will lie and give false information as the torture becomes increasingly brutal; conversely, some will protest their innocence through any level of torture, and some will simply say nothing.

Players will be allowed to detain suspects for as long as they choose, torture them in any way provided by the game designers, and execute them at will. All of these actions must directly affect the rest of the game. The guerrillas might gain popular support, and become more numerous and better armed, depending on who the player arrests, how the suspects are treated, and whether they are released, detained indefinitely, or executed. In addition, as a result of the player's actions, suspects could become increasingly less likely to allow themselves to be arrested, opting instead to shoot it out with the player's troops or blow themselves up to evade capture.

In addition to the effects of the player's torture on the effectiveness of the mission, there must also be consequences to the torturer. This can best be accomplished by having a single interrogation specialist character with greater narrative depth than most other characters: in the context of the interrogation sequences, the specialist is the protagonist. While much of the game's dialogue can be semi-random, the interrogation specialist must have more tightly scripted dialogue, and more of it. If the game is to have a narrator of any kind, the interrogation specialist is the logical choice. As torture becomes more frequent and more brutal, the specialist will become increasingly unhinged. Torture will become more difficult to accomplish, as the protagonist increasingly “ignores” the player's controller input, increasing the number of so-called “accidents.” As the protagonist moves from torture as a means to an end to torture as an end unto itself, he will become less effective at extracting information. The less brutal methods of interrogation will cease to be available to players. Eventually, it will become impossible for players to do anything with suspects except brutally torture and kill them, and doing so will only hasten the victory of the guerrillas.

These are the basics of the game, the elements common to any meaningful argument against torture. From there, three specific arguments can be made. The specific mechanics of the game, such as the probabilities of arresting an innocent person or extracting false confessions, will be dependent on the designers' intended argument. The first is a rather Machiavellian claim that torture is an effective tool for a counter-insurgency, but must be used sparingly, so the benefits of useful information outweigh the costs of increased enemy resistance and deaths of innocent victims. This argument defines what is good as what wins the war, and treats torture as an evil to be engaged in only for a greater good. For this argument, torture must make the game easier to complete; refraining from torture as much as possible must bring a greater difficulty and a greater reward. Nonetheless, the only win condition is military victory, and no moral rule is more important than that.

The second argument is that torture is simply counter-productive. For this argument, the variables must be set so the costs of torture are overwhelmingly larger than any possible benefits. Consequently, it must be impossible to complete the mission using torture as a strategy, and victory must be easiest when the player repudiates torture entirely. Again, this argument ties morality with military victory, and the most moral solution is also the most practical. This argument could also be made satirically by separating the win condition from military victory, and rewarding the player in non-diegetic ways for continuing to torture even as it dehumanizes the protagonist, kills innocent people, and allows the guerrillas to take over the country. The world will be decisively worse than when the player began the game, the mission will have failed miserably, but the player will be assured, through a high score or bonus content, that they've done the right thing. The sheer absurdity of such a game would be a powerful argument against torture.

The third argument differs from the first two by designing the game's ethics to serve an anti-torture morality completely divorced from military victory. The mission may succeed or fail, but such success is not taken into consideration in terms of the player's reward. Rather, the game must encourage players to torture by offering powerful short-term benefits, and reward them for resisting the temptation, both with non-diegetic rewards such as points and unlocked content, and a well-constructed narrative that makes it clear that, win or lose, soldiers who refrain from crimes against humanity can at least look themselves in the mirror with their sanity intact.

These are, as I like to say, loose thoughts. I can't design this stuff, and don't know if it would work, assuming we can all agree on what constitutes "working" in this context. But it's an interesting possibility, and an interesting way to think about this kind of debate.

No comments: