It’s hard to imagine the reaction of spectators in the late 19th century encountering film for the first time. There was of course the well-attested initial shock of seeing the living world made uncanny. Maxim Gorky called movies a window into “the Kingdom of Shadows”—“a world without sound, without color…grey rays of the sun across the grey sky, grey eyes in grey faces, and the leaves of the trees are ashen grey.” But beyond this, the question soon arose in the minds of both filmmakers and filmgoers: what are these things and what can they do? The Lumière brothers regarded their creation as a scientific tool, not an artistic medium, and famously declared it “an invention without a future.”

Some early producers sought to give the Shadow Kingdom a more solid cultural footing by tying it to an established art: the theater. Perhaps movies could claim a pedigree by embalming the gestures of Sarah Bernhardt and shipping them out to enlighten the backwaters. Such films were pretty stiff—which isn’t surprising since they served primarily to emphasize all that film seemed to lack with regard to theater while asserting no positive value beyond a mechanical ability to record motion. A counter-movement was inevitable, asserting that the value of the medium lay in those qualities unique to it—the “cinematic”—which in turn was effectively defined as all that theater can’t do.

Readers of this site are no doubt acquainted with such potted histories of movies. I haul this one out to draw out a parallel with video games, which for all their technological progress seem never quite able to shake that basic question of just what it is they are. As with film, one mode of response has sought to overcome any perceived disreputability by investing in the cultural capital of an older medium—in this case movies themselves—while another has reacted against that argument with an essentialist position, saying video games should be evaluated primarily as games, that whatever meaning they contain is dependent on their play mechanics.

In recent years, gaming culture has concentrated many of these arguments in a single phrase—“the Citizen Kane of games.” A Google search turns up over 400,000 results, and a wide variety of uses. Some see the Kane of games as a coming attraction, others as finally arrived, while still other contingents view the whole prospect as stupid, or at least misguided. For those dissatisfied with the current state of games, the phrase implies some future in which they will “grow up,” and the latent possibilities of the medium will at last reach full flower. Most often, though, the phrase is used defensively: for these writers, games are already great; they’re just waiting for the title that will establish that fact once and for all in the eyes of partners, parents, cultural arbiters—whatever disapproving force one cares to imagine.

Perhaps the impulse to defend video games against the charge that they are at best a waste of time and at worst some soul-sapping incentive to violence and general immorality has its roots in something much older than video games themselves—for whatever else they are, they are a form of play. In his great 1938 study of play, Homo Ludens, Dutch cultural historian Johan Huizinga notes that “This ‘only pretending’ quality of play betrays a consciousness of the inferiority of play compared to ‘seriousness,’ a feeling that seems to be something as primary as play itself.” This is far from any condemnation of play on Huizinga’s part—he recognizes that there is a seriousness in play, that “the inferiority of play is continually being offset by the superiority of its seriousness,” and believes indeed that play is an essential constitutive element of culture. Rather, this sense of inferiority is something he sees as inherent in an activity characterized by its lack of direct purpose and concrete utility in any area outside the localized grounds and rule sets of play itself.

So prophets of the Kane of games have yet another goal in mind—to establish the medium’s bona fides in a cultural sphere already characterized by its lack of concrete utility, that of Art. Consequently, when Roger Ebert declared in a 2005 Answer Man column that video games were definitely not an art, and what’s more never would be, he unleashed what he termed a “firestorm” of response, mostly rude.

I’m not sure what video games are either. I play them, but don’t know how to write about them. The term covers such a wide spectrum of experiences, from simple matching games to sprawling narrative epics, that it’s difficult to make any encompassing statement on the medium—if it’s a medium (singular) at all. Since video games are often built around the mechanics of particular platforms, whether these be PCs, consoles, or handheld devices, with various sub-sectors operating even within these categories, it may be that they ought to be regarded in plurality.

In addition, they’re a constantly moving target. It would be nice to say that video games have reached some decisive turning point, but in fact they seem to be always at some turning point or other, yoked toward imagined futures by advances in programming and technology as well as the perpetual hype-cycle of an enthusiast press that makes its Hollywood equivalent seem sedate by comparison.

So rather than taking aim at any large thesis, I’ll adopt a scattershot approach to a few topics in the field, drawing largely (if sometimes indirectly) on talks I heard at this year’s Game Developer’s Conference (GDC) in San Francisco. And to add a bit of mimetic fidelity to my treatment of a subject that often thrives on conflict, toward the end I’ll take up the Ebert Challenge.

Model Operations

Unlike Huizinga, game developer Will Wright sees a direct utilitarian function in play: through it, people build hypothetical models of the world. “When kids play, it’s basically the scientific method,” he said in a GDC talk, adding that every bouncing ball is an experiment in physics. Wright, the creator of The Sims and SimCity, views games as “a possibility space,” a bounded arena in which to test various models. One way of evaluating a game might look at the size of the possibility space it creates, with extra points for elegance going to those games that establish the largest ones with the fewest number of rules. Here Wright cites the board game Go as an exemplar—a nearly infinite number of possible configurations “with only two rules, one of which is almost never used.” (As a side note, I wish he’d spelled them out. I love Go and play it, however badly, but am unable to boil the rules down much past four.) For Wright, games allow players a safe ground in which to experiment with various schemas, thus enriching their own store of conceptual models of the world. In the adventure of ideas, he notes, “perspectives are often more satisfying than solutions.”

I find these notions attractive, but there’s another side to them that went unacknowledged in Wright’s presentation. Go is an abstract game—black and white stones on a 19-by-19 grid—and though one can read any number of models into it, they remain metaphoric. Wright’s games model real-world systems of considerable complexity—in the SimCity games, players are tasked with city planning and maintenance, from creating infrastructure to setting tax levels. A whole set of ideological assumptions unavoidably enters into this. The player can’t construct an anarchist commune in SimCity since it’s restricted to what Wright once termed “a capitalistic land value ecology.” But even within this particular schema, some critics, such as former Clinton adviser Paul Starr, have seen the game’s rule set as skewed a bit to the right. In an American Prospect article titled “Seductions of Sim,” Starr argues that “The critical problem raised by simulation is the black-box nature of the models” underlying the game play.

Part of the challenge of playing nearly any video game lies in trying to figure out the hidden rule sets, which actions are rewarded or punished within the particular system. But much more than most games, the SimCity series has a pedagogic intent and was often used in American schools in the ’90s. The games are beautifully designed and undoubtedly lead players to think differently about urban environments, to begin to recognize their interlocking rings of systemic organization. But given that Wright’s design philosophy is targeted toward getting himself out of the way as much as possible to allow players some free exploration of their own model sets, it’s a little ironic that his own pet set might exert such a subtly coercive influence.

A similar debate has carried over to Wright’s highly successful Sims franchise, which in its first iteration tied popularity pretty closely to the accumulation of material goods. There, though, the world depicted is sufficiently cartoonish that the games have also been read as satire. Certainly making such a rule set blatant, even garish, is one way of cracking open the black box. Richard Evans, lead AI programmer on The Sims 3, demonstrated another at an earlier GDC talk, with a prototype game provisionally titled SimTribes, in which the player gets to set the basic terms of reverence and taboo for their social units (though that tribal metaphor might open another can of worms).

Meaning and Mechanics

If there is a nexus of meaning in video games, where does it lie? Because most of us are trained to think of meaning in thematic terms, the temptation would be to place it there, at least for games with strong thematic content. But EA designer Søren Johnson insists that the meaning of a game is dependent on the player’s actual interactions with its rule sets and mechanics. When these act to support the theme, both are enriched, but when they operate at cross-purposes, the dissonance can be pretty severe. In a GDC talk titled “Theme Is Not Meaning,” Johnson gave a number of examples of design mismatches, including one from a game he worked on: Will Wright’s Spore. In Spore, players begin with a single-cell organism and over the course of five stages progress from creature to tribe to civilization to (potentially) galactic conqueror. Johnson noted that the game was promoted to both the press and the scientific community as a means of teaching kids some of the basic tenets of evolutionary biology, but the game play is geared toward something else: player creativity. Consequently, pretty much any outlandish figure the player designs can thrive in the game’s universe. To illustrate this, Johnson showed a slide of the 50 most popular creatures generated by Spore’s player community. They include Jesus, Obama, and a trombone. (Early players gravitated toward walking genitals, but EA purged most of these from its servers.)

Since the game play has less to do with evolutionary strictures than careful oversight by the player-god, Johnson said the running joke in the development group was that Spore was actually a game about intelligent design. This suspicion was shared by Science magazine writer John Bohannon in a 2008 review of the game, but he found the truth to be “not even that interesting.” He showed the game to a group of scientists, two of whom  called it “an impressive, entertaining, and elaborate Mr. Potato Head that uses the language of evolution but none of the major principles.” He also sent a copy to an intelligent-design advocate, who thought it closer to Pokémon.

Bodies in Motion

Johnson’s argument can even be carried a bit further, I think, into the realm of phenomenology. The feel of a player’s interaction with a game is an essential element of the game’s mechanics. Strange as it may seem, most video games are a tactile medium, or, to put it more precisely and still more oxymoronically, a medium of displaced tactility. Apart from simulation games like SimCity and abstract games like Tetris, the majority of titles depend on player identification with an avatar, some level of immediate physical empathy with a representative body in space. And much of the pleasure of games lies in their bodily affect, the sensation of fluid movement through complicated spaces. It may seem less surprising that such sensations can travel even through the blocky intermediary of the console controller when one considers that such physical empathy is, I would argue, an important component of movie watching too: an Astaire dance, the perfectly calibrated gesture of a Bresson model, Stan Laurel with a crab down his pants—all of these register first in the muscles.

Recent innovations such as motion control, which aims to create a one-to-one relation between player gesture and movement in the game world, and Microsoft’s forthcoming Natal technology, which recognizes player body motion and supposedly even facial expression, seek to further bridge the divide between real and virtual worlds. It’s one signal of the push-pull tensions of much contemporary media—even as Hollywood blockbusters are spilling out of the screen into 3-D space, gaming systems are exerting a parallel inward draw.

But it could be argued that much of the strength of both media lay in their expressive junctures from life. In the 1920s, Hungarian film theorist Béla Balázs argued against the talking picture on such grounds, saying the artistry of movies lay somewhere in that shadow zone between real and represented worlds, and that bringing movies closer to life threatened to sap their imaginative potential. Few film lovers would take that position today—even in the ’30s directors such as Lang and Vigo were quick to discover new suggestive possibilities in sound—but the idea of the importance of an expressive juncture of some kind remains interesting to contemplate, since such gaps perhaps create a more expansive possibility space for the viewer/player’s own imaginative projection.

There’s also the paradoxical fact that the very graphical advances that allow for more realistic approximations of bodies and environments also increase the threat of inexpressive or even awkward juncture, paving a path into what game theorists call “the uncanny valley.” The term comes from robotics pioneer Masahiro Mori, who noted that people’s emotional response to artificial figures generally follows an ascending curve the greater the representational accuracy—until the figures become nearly (but not quite) lifelike, at which point the graph takes a sharp drop into fear and revulsion. Video games have reached the interesting stage of development where they constantly skirt the borders of this declivity, and frequently take the fall. Since the avatar stands as the player’s immediate link with the game world, and since the perception of the uncanny is often localized in the unalive face, developers commonly situate the player either in a first person perspective, sharing the avatar’s eyes as it were, or in a position close behind the figure.

Restarting the Level

When film critics compare some particular movie to a video game, it’s almost always derogatory, and it’s almost always based on a couple of factors: heavy reliance on CGI and a strategy of attaining a rush of continual, heightened sensation through quick cutting. The first is fair enough. The second is weirdly inaccurate. Many games, especially shooters, do strive to create an ongoing rush of sensation, but they frame it within expansive, unbroken tracts of game world. Valve Software’s influential Half-Life series almost never departs from a single straight line in space and time. Apart from occasional loading pauses for the hardware to clear its cache and prepare the next area, the game is effectively a 12-or-more-hour continuous tracking shot.

Consequently, films like Transformers offer a poor analogy to game spaces. One has to look elsewhere, perhaps to Hitchcock’s Rope, which disguises its cuts to create the illusion of a single back-and-forth movement, motivated by the restless crossings of the film’s characters. Or Sokurov’s Russian Ark, which threads its way through St. Petersburg’s Hermitage in a 90-some-minute take representing the point of view of one figure adrift in time. But the canniest approximation of video game geography belongs, in my opinion, to Gus Van Sant’s Elephant. That film does something remarkable, establishing a sense of intimacy with the kids wandering the school’s hallways not through dialogue or exposition but simply because they stand as our avatars in this threatened and threatening space. It’s a bond forged in large part through the soundtrack, which slides almost imperceptibly between a rich representation of the school’s actual acoustics and something more muted and interior: a hum of preoccupation, some memory of piano. It’s forged too in a play with visual focus and depth of field in which characters are located in grounded spaces that dissolve over the course of the film into rectangles, spheres, and diamonds of light and color, with only the figure we're following in sharp relief. The film includes video games in its uninflected checklist of explanations offered for the Columbine massacre, but the film itself is a kind of video game, an anti-shooter that keeps restarting itself, as if to delay the arrival of guns and slaughter.

Other Worlds

The worlds of games such as Bethesda’s Fallout 3 and Rockstar’s Grand Theft Auto series are huge, open, highly detailed, and remarkably beautiful. Players are pointed toward story nodes, which they can follow at any time they wish, but they’re also allowed to wander at their leisure, stumbling across sidequests and grace notes planted by the designers. The post-apocalyptic Washington D.C. of Fallout 3 is chock-full of micro stories, many of them more interesting than the game’s central plot. Some involve narrative agency on the player’s part; others simply stand as efforts to inscribe some sort of history on the environment (what could have happened in the room that features a garden gnome chained to a bed?)

Grand Theft Auto IV is, among other things, a lovingly rendered tribute to New York. The game’s Liberty City gives its prototype a slight proportional squeeze and refracts its landmarks through a satiric lens (where Kafka’s Statue of Liberty holds a sword, GTA’s proudly hoists her Starbucks cup). It’s a peculiar experience to play the game in a Brooklyn apartment—it inspires an intense nostalgia for the world outside your window.

The actual game-play elements sometimes have a hard time living up to the density of their environments, the very sophistication of the visual design serving to highlight a set of older play mechanics. The translucent arrows that indicate mission start points float strangely above Liberty City’s street grime, and the world is so large and rich, so close to real, it seems like there ought to be more to do than chase cars, set up explosions, or go bowling.

But such “sandbox” games encourage exploration and discovery, and the worlds themselves are such intricate creations that they reward random walks and concentrated attention. GTA 4 made me a tourist again in my own city, a place I’d ceased to see. For filmmaker Phil Solomon, “What gets me is the unnecessary interest in landscapes and even abandoned buildings that are just there for you—they serve no narrative or mission purpose.”

Solomon has mined the environments of GTA 4 and its predecessor GTA: San Andreas to create a haunting series of experimental films. One of these, Last Days in a Lonely Place, remakes Rockstar’s San Andreas into a shifting bardo zone for the recently deceased, desaturating the game’s color and heightening its contrast to suggest a fluid world both expansive and provisional, built piecemeal from memories and old movies.

In making these films, Solomon was also inventing his own game within GTA’s confines, one that combined guerrilla filmmaking with guerrilla warfare. To get one shot in Last Days, a slow track into a movie theater, “First I had to sit there all day and time it until I knew when the marquee would come on—the shot starts with the marquee dead and then it comes on. Then I used cheats to bring in two army tanks to block off the street. Then I had to steal a car and smash out its headlights with my baseball bat—otherwise it creates a spotlight I didn’t want. And sometimes I’d miss and hit the hood and it would catch on fire. So it took several takes to get it right. Then one time I almost had it: I’m driving really slow, really careful, and I’m almost at the booth when a taxi cab—they’re programmed to be aggressive—snuck through the tiny space between the army tank and the wall and committed hara-kiri right in front of me.”

Solomon’s most recent film, Still Raining, Still Dreaming, uses careful slices of GTA 4’s Liberty City, but he found the newer game much less tractable for his purposes. “Partly that’s because there are certain cheats that are missing for me. In San Andreas, there are cheats that let you get rid of traffic. You don’t get that here, and it’s New York City so you have people bothering you all the time. I wanted to shoot Chinatown empty and in order to do that I had to fire my machine gun, clear everybody out. That’s my assistant, my machine gun.”

He has also remade Andy Warhol’s Empire, shortening its original 485 minutes to correspond to GTA’s compressed time cycle: “I learned from looking at [GTA’s] Empire State Building for 45 minutes how beautiful light is in that game. Crime is going on down below but they took care of the cosmic as well, and that flipped me out.”

Solomon is already planning a film that will use the environments of Rockstar’s forthcoming western, Red Dead Redemption, when it’s released later this month. He’s anticipating that the new game’s open plains will preclude any recurrence of GTA 4’s crowd control problems: “I’m hoping that I can just go out and be alone and record the dust.”

Thanks to D. Cairns, Craig Keller, and Mahon McGrath.
“State of Play, pt 2” will be published next week.