I said, ”Footsteps in movies.”
“Footsteps in movies never sound real.”
“They’re footsteps in movies.”
“You’re saying why should they sound real.”
“They’re footsteps in movies,” she said.
—Point Omega (Scribner, 2010)
The preceding conversation from Don Delillo’s novel Point Omega occurs between a documentary filmmaker, Jim Finley, and the daughter of a man who is the subject of Finley’s latest documentary in progress, a woman named Jessica Elster. It is probably no surprise that a documentarian would raise the issue of how reality may or may not be successfully depicted in film. Representing reality as authentically as possible would seem to be the bread and butter of most documentary filmmakers.
Jessica’s response, though, is an interesting one and probably a useful one for the documentarian to consider. Film will always be artificial, always be representational, despite one’s intention to be as real or as truthful as possible. Jessica’s comment suggests that the medium of film, the world of representation, is a world of its own, a world that has its own ontology. That ontology is probably largely forged around an audience’s perception of what is (the film, not their direct experience of what a film might document). In other words, the ontology of film is based on its audience’s expectations of what constitutes the reality of film, not what constitutes reality.
One doesn’t expect to hear the sound of footsteps in a film, one expects to hear the sound of “footsteps in a movie” in a film.
While certainly video games are an audio-visual medium (so similar expectations about the representation of sounds or images might be expected by gamers in terms of what audio and visual elements should sound like or look like), this notion got me thinking about what might constitute the “footsteps in a movie” in a game as a result of it being a game.
Stealth games, like Thief, the Batman: Arkham series, or Metal Gear Solid have always contained an element to me that never quite seemed real, guard routes. While the AI governing guards patrolling an area has grown more and more sophisticated over the years, still, guard movement always feels programmed (well, because it is). Hence, guard movement never feels real.
In the earliest Metal Gear Solid games, generally speaking guards followed certain simple surveillance patterns (not unlike the movement of an oscillating security camera, which simply swoops back and forth, back and forth over the same area in the same predictable way because its oscillation is timed). A guard might follow a corridor, reach a fork, take a left, stop at the next intersection, turn fully around and follow his route back to his starting point, before, of course, doing it all over again and again and again and always over the same course of time.
“Getting caught” in such a game is making the mistake of simply not timing one’s movements properly. The rule for success in such a game is: don’t be in that guard’s field of vision at the wrong time. The guard moves, again, like a machine, not like a human.
Also, once again, guard movements have gotten better, less seemingly mechanistic over the years by adding different elements into their program, like following multiple programmed routes, stopping briefly for a smoke at some point in their route, or even moving somewhat more randomly as a result of sights or sounds that distract him. However, even the latter type of movement requires that the guard responds in certain scripted ways and returns to his old patrol route in a fairly predictable way (see Batman: Arkham Asylum for some clearly obvious examples of this idea).
Further though, the idea of predictability is actually rather essential to games like Metal Gear or Batman or Assassin’s Creed. If guard movement were truly more life-like, more unpredictable, the game would no longer feel like a game because it would lack a means of “playing well” under the circumstances, non-scripted, random guard movements. Stealth games are games that are beatable because they are programmatic, because these games are made “fair” for the player, allowing them a way of predicting outcomes and playing within the boundaries of a situation. If guards moved like real guards, what would suffer as a result is the game as a game because sneaking around would become next to impossible in some situations or extremely dull. You might have to wait in an area for hours at a time before an opportunity arose to move unseen and undetected. In other words, guards in video games should not act like guards in real life. Guards in video games should act like guards in video games, like machines that follow artificial rules.
Indeed, that is what they are. They are not guards. They are guards in video games.
Since rule systems govern games, creating balance and fairness, it is rather essential to the ontology of video games, that representations in video games act like what they are representing—but in a video game.
Perfect simulations, often feel less like games because they lack our expectations of fairness. Imagine, for example, a game in which one simulated World War II in which each player took on the role of one of the countries dominantly involved in the war, say, the United States, France, Great Britain, Poland, Germany, Italy, Russia, and Japan, and in which those countries military resources were “true to life.” Playing Poland in a perfect simulation would be terrible. It would be unfair. If you want such a game to play like a “proper” game, you need Poland with its more limited resources and military technology to have some perk, some benefits that violate the “reality” of the simulation, that makes it possible to win, putting you on an even ground with the other countries and, thus, the other players.
Why should guard routes feel real? Why should countries represent their actual historical situation?
They shouldn’t. After all, they’re guard routes in video games, they’re Poland in video games.