Refereed articles

Information articles

Notes on contributors

Print friendly version

Neither gaze nor glance, but glaze: relating to console game screens

Chris Chesher

A heavy set man stands on a blazing Callaghan Bridge, surrounded by the sounds of early morning Liberty City. He's just escaped from an exploding prison van, and fellow escapee ?8-ball? is asking him to drive to a hide-out in the red light district. Then the frantic action stops, and it is now your turn. A huge pulsing blue arrow in the sky asks you to become the shady character on the bridge. First you have to learn how to control this brooding thug: you press the triangle button and he gets into the car. You press the ?X? button and the car accelerates down the bridge. You struggle with the direction controls to keep the car on track as you speed into the city.

This introductory sequence in Rock Star Games? Grand Theft Auto III (GTA3), released for Playstation II in Australia in October 2001, is a typical opening for a console game. A somewhat cinematic back-story sets the scene, and defines the context and motivation for the action in which you are asked to participate. Although the opening mission ? to drive the car to the hideout ? is simple, it's quite demanding at first attempt. There are pedagogical elements scattered through the early sections: explanatory text that appears in the early scenes: park on the glowing blue spot to finish this mission; go through this door to save the game; park your car in the garage to keep a car between saves. The subsequent missions gradually introduce the conventions of gameplay and the world of the game. You learn how to control your seedy character?s movement on foot and in cars. You work out how to start missions by following a ?radar? display in the corner of the screen to the locations where you receive instructions. You experiment with your avatar?s powers by beating people up with a baseball bat, and figure out from bitter experience that it?s best to run away from police. You start to become familiar with your new body, and with the geographical and ethical layout of the simulated city.

After an hour of playing the game you are transfixed, held to the screen. Your hands are aching from pummelling the controller. You are no longer thinking about which buttons to press. Your own actions seem inseparable from those of the character on screen. Your eyes are sore from staring at the screen. You're barely aware of other things in the room.

This is not television. And it?s not cinema. The console game is a distinctive cultural form in need of its own modes of criticism and analysis. It?s undeniable that such games are important in sheer economic terms: they have now overtaken the cinema box office. But more than this, games have become a day-to-day screen experience for millions of people, and it?s time to try to make some sense of them. At the moment, there is a rapidly growing literature on computer games. Many researchers find it useful to start by comparing games with more familiar media forms. Espen Aarseth, who now heads a new Centre of Computer Games Research in Denmark, compares games with an esoteric tradition of literary forms in which the reader is expected to make a non-trivial effort to traverse the text (1997). Brenda Laurel compares computers with theatre to argue that interface designers should stage on-screen events in a manner that borrows from the principles of stagecraft (1991). Lev Manovich?s Language of new media reads computers alongside the history of avante-garde cinema (2001). At their best, such approaches do not simply borrow and apply theories that work for other media, but show where the old concepts start falling apart in the face of a new phenomenon.

In this spirit, in this paper I will ask what differentiates console games from televisual and cinematic forms? What is distinctive about the experience of playing games? How are the relationships people have with these screens different? What new concepts do we need to understand games? While they are clearly different as audio-visual texts and experiences from cinema and television, they are also related to them in complex ways. In fact, analysing computer games may demand a re-evaluation of traditional conceptions of other media.

Visible fictions

One common theme in Media Studies is that any medium tends to structure the ways that people look at it. Looking is a learnt skill, and the conventions of looking are particular to any visual medium. Viewers are expected to look at paintings in a gallery in one way, and watch live theatre in a different way. Ways of looking change historically. Any new visual technology emerges with its own conventions ? its own structures of feeling. It has mechanisms that attract eyes, and techniques to counter distractions that might draw those eyes away. In this way, even traditional media are already interactive, in that they structure the activity of those who encounter them. A work succeeds only when its content and expression anticipate audiences? responses.

John Ellis?s influential 1982 book Visible Fictions is a good example of the media studies tradition that looks at the activity of media consumers. Ellis compares the characteristic positions that spectators take towards Hollywood cinema with the positions that viewers take towards broadcast television. Because he did not anticipate computer games at all, his work becomes a useful starting point for evaluating the new regimes of vision that emerge in the way that console game players relate to their screens. He establishes a generic framework for analysing agent / screen relationships, but leaves the example of games entirely unexamined. My task in this paper is to put the console game alongside more familiar media forms and then to look for the differences and the connections between all three. As Ellis admits, trying to characterise a relationship to any medium as something stable and consistent is necessarily an oversimplification. Different individuals and groups relate to films and television programs in different ways. The same problem of generalisation will be true with different games and different players, whose experiences vary dramatically.

Ellis drew on then recent work in film studies on the spectator's position in relation to cinema (most notably in the journal Screen), informed particularly by psychoanalysis. In the cinema, spectators are positioned as voyeurs, sitting together in a darkened public place, their gaze intently focussed on the screen. This arrangement encourages a psychological state that resembles dream or fantasy, allowing spectators to build a close libidinal identification with characters in the narrative. The spectator experiences a complex sense that they recognise themselves in the images of others on the screen ? often ambiguously in more than one of the characters. At the same time, they take on positions as voyeurs ? with a sense that the image is presented for their pleasure. While they can?t intervene in the film world, they can surreptitiously experience the internal lives of others. As feminist critics have argued, the cinematic gaze is often gendered. It tends to privilege the active agency of the male gaze, while presenting the female as passive object of that gaze. It anticipates the male ?scopophilic? pleasure in looking at women as objects (although more recent feminist work on the gendered gaze has criticised this reading as overly reductive and totalising). The relationship of spectators to the screen is structured around manipulations of gazes ? of characters, camera and spectators.

Television viewers relate to a much smaller screen that is always present in their domestic space, so images are experienced as relatively mundane. Television offers viewers a surrogate day-to-day image of the world that casually makes them complicit in this structured way of looking. This view separates the 'normal' reality of Citizen and family at home from a variety of abnormal worlds on television screens. The image quality of TV images is relatively poor, and at home viewers? attention can often drift, and has to be drawn back regularly, usually through sound -- canned laughter, jingles and stings. Content is segmented so that viewers can easily rejoin the narrative at any point. Television?s characteristic regime of vision is the glance, rather than the gaze.

Writing in the early 1980s, Ellis did not foresee that another culturally significant regime of vision in screen media would emerge over the next two decades. This new media formation, manifest in the console game, operates with a different regime of vision: not gaze, nor glance, but what I will call the glaze. The glaze is a liquid adhesion holding players' eyes to the screen. Players are held to the game in two ways ? with their hands on the controller, and their eyes on the screen. Identification between player and the central on-screen character is direct and visceral. The player takes on the role of a character directly, rather than the cinematic spectator who vicariously identifies with others. Players are glazed into a game world subjectivity, rather than detachedly gazing as cinematic voyeurs, or indifferently glancing at the world through television.

Classical cinema, broadcast television, and console games are distinctively different in the ways in which they structure audience / viewer / player subjectivity. The table below outlines themes that I explore through the rest of this article.

Hollywood Cinema Broadcast television Console games
Spectators Viewers Players
Darkened public space Domestic space & family Atomic individuals in domestic (or other) space
Large, projected, high resolution photographic image Small, lower resolution, luminous image; sound draws attention Simulated images invoked from databases and algorithms
Phantasy, voyeurism Distracted complicity Sadomasochistic fetishism
Libidinal identification with others on screen Watch bizarre world from the security of domestic normality Role play in navigable spaces
Nostalgia: what?s underneath? Liveness: what?s happening? Possible worlds: what can I do?
Gaze Glance Glaze
Hollywood Cinema Broadcast television Console games

The distinctiveness of the glaze is increasingly apparent in the recent generations of games consoles, although it has been on the rise since the mid-1960s. Third person 'over-the-shoulder' games on recent games consoles (GTA3 and GTA Vice City) are exemplary of this regime of action/vision. No longer arcade games, they belong in the home. And unlike many earlier computer games, which remain obsessed by the technology itself, many console games draw on wider cultural references. Early computer games focussed on the computer in the same way that the proto-cinematic experiments of the 19th century foregrounded the apparatus itself before the conventions and technology of cinema stabilised. When Space invaders presented an irresistible hoard of aliens descending on the player?s avatar, it metaphorically performed its own entrance into the world. Flight simulators, of which there have been many iterations, mimic one form of high technology with another. Japanese games developers were among the first to move outside the techno-fetishist milieu, incorporating the aesthetics of manga and anime, with great success in the earlier generation of console games from Sega and Nintendo (Wark, 1994:21-30).

As the games market has expanded, and technologies and cultures of production have matured, games have moved outside the technologically marked audio-visual frameworks, and also beyond the association with children. The three dimensional immersive games released in the early 2000s have drawn upon a wider range of cultural references. The increased capacities of storage and 3D processing power on the Sony Playstation 2 and Microsoft X-Box have made it possible to work with more cinema-realistic images and much more extensive sound. In this context the wider cultural significance of the glaze is becoming more apparent.

While no single game can represent the characteristics of all others, the GTA series manifests many of the key characteristics of the emerging conventions of the glaze. The earlier games in the series, GTA, GTA London 1969 and GTA2, feature relatively simple graphics and a top-down view. The first game was controversial at the time because it seemed to encourage criminal behaviour, attracting the ire of Senator Joseph Lieberman in the US. The producers? media savvy was demonstrated by the way they embraced the publicity this controversy generated. However, the later games begin to have more televisual or cinematic audiovisual qualities. GTA3 is the first fully three-dimensional game in the series. It contains more voice acting, a substantial soundtrack featuring radio stations from the fictional city, and a far more detailed world. GTA Vice City, released in 2002, gives the main character a name (Tommy Vercetti), a voice (played by Ray Liotta), and a historical location (the 1980s). It has even more developed visual and sonic complexity and depth to the point where the gaze, the glance and the glaze can be juxtaposed without any apology.

Three dimensions of the glaze

The audiovisual regime of the glaze is characterised by at least three distinctive characteristics: spectacular immersion, interactive agency and mimetic simulation. These are not sequential categories, but simultaneous layers, each corresponding with a different meaning of the term 'the glaze'. The first meaning of glaze refers to how a player's eyes become apparently 'glazed over' when they are absorbed in the sensations and actions of a game play space. The second meaning relates to stickiness, such as the glaze on a cake. Games are sticky because they incorporate strategies to hold players to them. By giving direct feedback; a sense of power; game teleology (missions), and narrative elements, the game generates affect and captures the player's desire to continue to play. The third meaning relates to the way that a glazed object (such as a glossy vase) presents a reflection of the world around it. Players recognise themselves, and a familiar world, in the game. But the images reflected in a curved vase are distorted according to the shape of the glazed object.

i. Eyes glaze over

Game players go into glaze space ? immersive spectacle ? when they progressively remove their attention from immediate material and cultural surroundings, and plunge into a world with different physical and ethical parameters. In this state, players seem to be in a vegetative condition ? their eyes glazed over. But this appearance is deceptive. They are actually intensively active, both consumed by, and actively consuming, the game. In glaze-space (which varies in intensity), players suspend their awareness of their day-to-day world to become cybernetically suspended within a virtualised sensorimotor space of the game world.

The ontological centre of any glaze-space is the game's virtual point of view (or multiple points of view). All games present a world from some perspective. Even the early text-based adventure games had points-of-view that became apparent in the descriptions they gave in text of what a player can 'see' and act upon. Early 2000s 3D console games like the later games in the GTA series offer spectacular views onto three-dimensional simulated worlds. The default glaze viewpoint hovers as an angelic virtual camera floating just above the shoulder of the protagonist character. When a player turns the avatar's body around and moves forward, the world literally revolves around him, and the camera follows obediently behind. This egocentric worldview immediately defines the stakes of the game: the interests of this character are central. Unlike cinema, which is edited, and therefore switches angles throughout, the glaze remains steady, unless players trigger a camera change themselves. Otherwise, the camera stares tirelessly in the same relation to the protagonist and his world. But like the dominant models of cinematic gaze, the glaze is undeniably a mode of vision that is gendered as masculine.

The viewpoints available within any game implicitly structure the challenges and pleasures experienced by its players. Other genres of games offer different points of view, some more detached and strategy-oriented, and others more subjective. The standard over-the-shoulder view of GTA3 contrasts with isometric points of view typical of real time strategy games such as Age of Empires and the Warcraft series, and god-games such as the Civilisation series, Sim City or the Sims. These viewpoints lack depth perspective: all distances are in the same scale, like a map that has been tilted. This literalised Cartesian view is better for arranging multiple avatars across a territory. The top-down view is also easier to compute, which was why it was characteristic of many early games like Pac-man or Space invaders. While GTA3 has a more subjective viewpoint than these map games, its over the shoulder view is more detached than the first person view in Quake, Doom, Half-life or Halo, which directly threatens the player's self. First person games generate a vertiginous sense of movement, and use sound and rapid movement within an enhanced depth-perspective space to give players a sense of visceral immersion. These are comparable with horror films or war movies that give audience members a direct sense of personal insecurity.

Figure 1 Isometric point of view in Age of Empires

Many games offer several choices of viewpoint, effectively supporting different experiences of playing the game. For example, in GTA3 it is possible to play around with point of view by switching between different camera modes. Three of the views offer alternative framings on the central character. In cinematic terms, one is a close-up, another a mid-shot (the default) and the other a long shot. Unlike cinema, the player controls the cuts between cameras, not the editor. When the character steals a car, the game switches viewpoints to something typical of a driving game, which also offers different levels of zoom and a first-person view. Another viewpoint (which is dropped in Vice City) recalls earlier incarnations of the game, and shows the character and surroundings from above. This view breaks the premise that the player experiences only what the character does, but it is useful for seeing what is on the other side of walls!

Points of view in games are almost always designed to provide possibilities for players to act within the game world. If a player wants to aim a gun, the interface will either offer direct line of sight down a barrel, or use an automatic targeting system. The potential victim must be on screen, and in clear spatial relation to the player avatar. This is quite different from the dominant shot-counter-shot editing structure in a movie shoot-out. In cinema, some sense of audience disorientation can be a directorial device. Some game viewpoints are the exceptions that prove the rule, by emphasising spectacular imagery over playability. When driving in the GTA games, it is possible to switch to a novelty viewpoint ? the ?cinematic camera?, which mimics a film car chase, with a series of shots from a range of perspectives. A sequence might start with a camera near the car wheels, then switch to a camera in a building above the car, and to another on the side of the road that pans as the car zooms past. This invasion of the gaze relationship into glaze space makes the game almost unplayable. At the same time it makes available to the player a new subjectivity ? of filmmaker rather than games junky.

The most significant shifts in point of view in GTA3 and Vice City, though, happen in cut scenes: cinematic sequences that typically appear at the beginning and end of missions. Like many other games, this convention advances the narrative and establishes the central character?s motivation for the next mission. During a cut scene, the screen switches to a wide-screen letterbox style, and the user has no control over the player for the duration. The protagonist character on screen is shown listening to the dialogue (more often monologues) of other characters, and there are cinematically significant camera movements. These scenes are important in establishing the scenario for the next sequence of action, and also help to develop a more textured mise-en-scene. They set the mood, and serve as a reward for successfully completing a mission.

Promotions of the later GTA games on television, print and the web tend to privilege the cinema-realistic imagery, and de-emphasise the default over-the-shoulder point-of-view. The television advertisement screened in Australia resembles a movie trailer. Many fan sites feature screen-shots: a still image capturing the contents of the screen during game play. These images tend to be taken at moments in which the game character, and the virtual camera, is in an extreme and unusual situation. During normal game play, the character is most often facing away from the virtual camera, so that the player can see what he is facing. In most screen shots, though, the player has chosen to turn the character towards the camera, borrowing from visual conventions of a personal snapshot, or a movie poster. None of these images give much of a sense of what really draws players to the game, nor what holds them to it.

ii. Sticky glaze and computing holding power

Games have a remarkable capacity to hold players? attention for long periods of time. Sherry Turkle talked in the 1980s about the computer game?s ?holding power? (1984:29). The glaze can also be imagined as a liquid adhesion that holds players' eyes to the screen. The handset and console are fixers in a two-part glue that connects players to the games console, physically and psychologically. The game interface offers affordances that let users in on the game by mapping buttons and joysticks to events within glazespace1. In this way users have agency ? to walk, run, jump, punch, shoot, duck and steal cars. Early in the game, players? powers are limited, and the challenges are relatively easy. The rudimentary narrative structure established in the cut scenes introduces missions that are pretexts for gradually introducing new capacities (weapons, cars, boats, routes to new areas etc.). To complete each mission players need to apply a combination of problem solving and learned skills.

Effective games sustain a cybernetic balance that dynamically maintains a hold over players. As a thermostat regulates temperature, ludostatic mechanisms regulate optimal playing conditions. If the glaze holds players too tightly, it will alienate them, and they will give up. But its grip cannot be too loose, either, or players drift off and lose interest. While some proponents of virtual reality dreamed of unlimited power and movement in virtual worlds, glazeplay works because it maintains limits to a player's capacities. The speeds and forces of actions are carefully modulated, and boundaries defined and policed. By playing more, players gather more power ? not only through faster cars and bigger guns within the game, but also because they themselves develop better skills and knowledge of the game. At the same time, the difficulty escalates as the missions become more complex and enemies become more aggressive. James Paul Gee argues that this balance found in computer games is an effective mode of learning:

Each level dances around the outer limits of the player's abilities, seeking at every point to be hard enough to be just doable. In cognitive science, this is referred to as the regime of competence principle, which results in a feeling of simultaneous pleasure and frustration… (2003)

Unlike the voyeurism of the cinematic gaze, the psychic relationship of the glaze is sadomasochistic. At its most basic level, the pleasure of play comes from inflicting and receiving pain within the ludostatic frame (Tulloch, 2002). Games engines tease players by tactically extending and constraining their powers according to the current state of play. The glaze relation is generally closer to what Laura Mulvey (1989) refers to as sadistic festishism rather scopophilia (the love of looking). However, there is also a strong masochistic dimension to the pleasures of glaze-play. It usually takes many failed attempts at completing difficult missions before the player succeeds. Failures are often more spectacular than successes. If you get ?wasted? (die), you lose control over your character?s movements, and the viewpoint drifts off into the sky. When you re-awake outside the hospital, you?ve lost all your weapons, and any mission you were on is failed. But you can restore your pre-death condition by returning to the most recent save. On the other hand, if you find health bonuses or armour (or controversially, visit a prostitute in the uncensored version), you increase your survival powers. Much of the pleasure of play is in facing and cheating death.

Games must always have something at stake ? an ethical universe with values worth fighting for. Like many games, GTA asks players to perform a constant dance with ethical problems. Players can replay sequences indefinitely, and experiment with different outcomes of the same circumstances. They can adopt a range of styles ? from being polite and careful to acting out murderous rampages. This capacity to explore a diversity of possible worlds contrasts with the tendency for television to impose a supposedly collectively agreed upon moral meaning for the events it represents. Television news advocates a moral order that centres on the family and domesticity, implying consensus and social cohesion.

However, games are not without their normative tendencies. In order to progress in a game, the player must play in a certain way. In the case of GTA this means acting according to what would usually be considered quite amoral standards. In Liberty City, extreme acts of violence are problematic not for their impact upon victims, but because they attract too much police attention. The game is self-consciously ironical, parodying the ethical relativism depicted in gangster films. The game logic itself is a form of social criticism or parody. Police take bribes, and are so dumb that they can?t recognise you if you have sprayed your car a different colour. Bystanders show self-interested indifference to violence nearby. You buy an ice cream factory that operates as a front for selling drugs, and a used car lot that gets its stolen cars delivered in the middle of the night.

Critics of the game (including the Australian Office of Film and Literature Classifications Board that initially banned GTA3) tend to overlook the complex ethical dynamics of games. Certainly gender and ethnicity are represented in this game in highly problematic ways, to a point of self-parody. The imagery of violence is gruesome and distasteful. But unlike many games in which the dead simply evaporate, in this game ambulances come to take away the bodies. Most games avoid dealing with ethnicity by marking others as space aliens and mutants. GTA?s cities are populated by a range of ethnically-marked criminal gangs: Italian Mafia, Japanese Yakuza, Columbians, Yardies, white collar corporate criminals, crooked cops and home-boys in colours. These glaze world stereotypes reflect recognisable images from a repertoire of Western urban cultural myths.

These self-conscious elements reflect a pervading dark sense of humour and social criticism that underlies these games. They are anti-authoritarian, with a cynicism towards legal boundaries as well as perceived boundaries of political correctness. The parody is most explicit in the radio soundtracks that feature self-obsessed announcers, vacuous talk-back callers and stupid advertisements. But more interesting is the warped ways in which the glaze world responds to the user?s actions. Many critics of game violence seem to evaluate games by applying a moral accounting that attempts to weigh up the social benefits and liabilities of a particular work to define whether of not it has redeeming value. The dynamic ethics of the game are not reducible to such a simplistic calculus. Much of what appeals about the game is in its exaggerated but evocative representations of a possible world that parallels our own without ever being it.

iii. Reflections of the self in glaze

The immersive and sticky glaze world operates on a third dimension by offering players a distorted image of themselves. A glazed ceramic surface such as a vase reflects the image of observers back to them, but distorts and colours these images out of their usual proportion. In a related way, game players see themselves present in the role that the game grants them.

While games are consumed domestically, they are not for the whole family. They are for atomic individuals. Where television addresses a collective viewing audience, games address each player as an individual. Of course there is still space in playing games for non-players to participate as spectators, tutors or collaborators. But it is the person with the active controller who is directly addressed by menus to set up the game and start play. They might even be asked to identify themselves by name, initials, or even an image (as in the 2003 PS2 game Eyetoy). This record documents their performance from that point forward. The status of each player (and their own world) is saved at the end of each session of play (and more often). There is a form of object permanency by which players? actions have ongoing effects. In multi-player games, the marking of individual identity supports competition and cooperation between distinct individuals.

Unlike television content, which is homogenised to suit family viewing, games are often addressed to include and exclude groups according to taste. Part of the pleasure of playing games that horrify your parents is that it asserts your own identity. The popular demonisation of games, characteristically on television, plays into this pleasure in non-conformity. Like much computer-based media, many games allow customisation and individualised role-play.

Ellis observed that cinema spectators tend to identify vicariously with several characters on the screen. Even if there is one hero, they can identify with almost any character at different times. By contrast, with game players there is a clear distinction between self and other. Players see their own decisions reflected in their avatars? movements. If they choose to move left or shoot, the avatar completes that movement for them. Unlike the full characters in a cinematic narrative, who seem to have independent existences, and psychological depths, the central character in GTA3, in particular, is empty. The central part-character is a heavily built white male with little else to distinguish him. As a designer from Rockstar Games said, there is a hole at the centre of the action, critical for identifying with the avatar:

The lead player is the only person who never speaks. He has no personality. Or rather he has your personality, the player has to have freedom to act as they want, not as they think the model they are controlling should want. Everyone else has a character, for the same reason — you are in their world, and they need to exist. (Douglass, 2001)

This lack of characterisation is no flaw. It is a necessary condition. Players fill that absence with their own motivations, strategies and reactions. The sequel, Vice City, gives the character some more depth by giving him a name and allowing him to speak, but retains an openness that allows users to partly recognise themselves within the character. Of course this openness is only limited, as the character is clearly marked as white, male and bad.

In the place of depth of characters in the GTA games is hyperbolic gender and ethnic stereotypes. The games ask players to participate in ridiculous patriarchal and phallocentric phantasies. In cut-scenes in GTA3, the character of Azuka plays dominatrix, tying up and whipping her victims, in a very literal portrayal of castration anxiety. What was apparently a romantic narrative is subverted in the soundtrack over the final credits, when the central character shoots the woman he has finally rescued because she starts complaining about her fingernails. Missions in Vice City include not only acts of direct violence, but also performances of masculinity including intimidating jury members, and smashing shop windows to extract protection money.

Glaze space is a familiar, but distorted image of the players? own world. While the cities depicted in the GTA games are recognisable as twentieth century urban landscapes, they are simulated composites of fragments of imagined American urban life. As the designers live in Scotland, these images are largely drawn not from personal experience, but from Hollywood cinema. The mansion in Vice City is borrowed from Scarface. Vans driven by gang members are just like those in Boyz in the Hood. The producers acknowledge that the world of GTA3 owes a lot to gangster films:

During the design phase of the game, the team would get together one night a week and watch films (however, this was actually an excuse to drink beer). We watched a lot of films including Bullitt, Casino, Scarface, Goodfellas, The Warriors, Debbie does Dallas... Basically anything with car chases, gangs and a lot of action. (Douglass, 2001)

Glaze space presents a possible world, not an actual world. In cinema, actors are recognisable as actual humans, and spectators are aware that an actual actor once performed this scene. Console games reference no world in particular (in the same way that a drawing, painting or animation can always be read as possibly pure invention). There are uncanny traces of human performances and physical places within the glazeplay, but these are not cinematic. The glaze world is a multimedia mosaic constantly being reconfigured. Fragments of actors' movements, voice performances, the shapes of buildings and images of surfaces are stored in a database as motion captured gestures, audio samples, 3D models and texture maps and triggered at appropriate moments. The world is composed on the fly according to the player?s whims. Glazeplay is invoked, not narrated2. Over hours of game play, the glaze spaces become a second home, almost as familiar as the streets in your own physical suburb.

Extending the glaze

Cinema, television and the console game each have a dominant regime of vision / agency: the gaze, the glance and the glaze respectively. However, these modes are by no means exclusive to each medium. Cut scenes in games are cinematic. Home theatre systems establish gaze relationships with spectators in domestic spaces. DVD menus bring a glaze interface to the gaze of cinematic texts. Experiments in interactive cinema attempt to take the glaze to the movies. Television creates glaze relationships in sport telecasts: viewers become closely identified with the sensorimotor world of the game, even without any direct control over the action. Channel surfing is a prototypical glaze activity. Other cultural forms that don't involve a cathode-ray tube or a projected image might also develop glaze relationships: gambling machines, pinball and (arguably) even driving a car. In each of these there are clear and narrow modes of action available, and limited parameters of affordance: a poker machine's lever, the flipper buttons on a pinball machine, the steering wheel of a car, and so on. Each of these does have a screen of a sort: a front cover, or windscreen.

The glazed over, sticky and identity-reflective conventions of games are ascendant as cultural forms, increasingly seen beyond traditional gaming subcultures. But the distinctiveness of the glaze relation is often not recognised. Games are sometimes read naturalistically ? as though they were like real life, and not medium-specific conventions. Screen theory and practice often overlook the specificity of glaze subjectivity, and read games according to the logics of gaze or glance. In producing glaze texts, some designers overemphasise immediate visual impact, while neglecting ludostatic engineering. Some theoretical work applies textual or visual analysis without attention to user subjectivity or to how game screens are experienced in very different ways. In aesthetic judgements of games, the lack of character development or narrative complexity in games is sometimes seen as an inadequacy. However, these mistakes are likely to become less common when, alongside the longing gaze and the distracted glance, the sticky glaze becomes an increasingly familiar human-screen relationship.

Note: an earlier version of this article was published as an essay in dLux Media Arts (2003) Plaything Exhibition Catalogue, Sydney: dLux Media Arts.

1 For more on affordances and user subjectivity see Chesher, Chris (2003) ?Layers of code, layers of subjectivity? in CultureMachine 5 (2003),

2 For more on invocational media see Chesher, Chris (2002) 'Why the digital computer is dead' in Ctheory electronic journal:


Aarseth, E. (1997) Cybertext. Perspectives on ergodic literature, Baltimore and London: John Hopkins
Ellis, J. (1982) Visible fictions, London, Boston, Melbourne & Henley: Routledge & Kegan Paul

Gee, J. P. (2003) ?High Score Education. Games, not school, are teaching kids to think? in Wired 11.05, online at accessed August 2003. See also Gee, James Paul (2003) What video games have to teach us about learning and literacy, London and New York: Palgrave Macmillan

Laurel, B. (1991) Computers as theater, Reading, MA: Addison-Wesley

Manovich, L. (2001) The language of new media, Cambridge, Mass.: MIT Press

Mulvey , L. (1989) Visual and other pleasures, Houndmills, Basingstoke, Hampshire: Macmillan

Perry, D. C. (2001) 'Rockstar's Sam Hauser mouths off' in IGN website,, September 10, 2001. Accessed August 2003

Perry, D. C. (2001) ?An Interview With DMA's Les Benzies? in IGN website,, October 3, 2001. Accessed August 2003

Rockstar Games (1998) Grand Theft Auto,; Rockstar Games (1999) Grand Theft Auto: London 1969; Rockstar Games (1999) Grand Theft Auto 2

Tulloch, R. (2002) ?The disciplined player: violence and the pedagogic process in computer games?, seminar presentation, School of Media and Communications, University of New South Wales, October 2002

Turkle, S. (1984) The second self: computers and the human spirit, New York: Simon and Schuster, p. 29

Wark, M. (1994) 'The Video Game as an emergent media form' Media Information Australia, 7:1 pp. 21-30