It is a disturbing quality of computer games, that they always dare you to yet another attempt at scoring more points, at reaching the next level. You oblige, but from a literary point of view, it is not at all obvious why you would want to. Computer games seem senseless pastimes, devoid of any point or reason. It then follows that they can hardly be described as art in any common sense of the word, since a meaningful phenomenon is assumed to contain something that can be read from the work. Computer games seem not to provide any feeling of having decoded the meaning of the game.
As a reaction to these meaningless games, interactive fiction is claimed to create games with meaning, games that tell a good story. The obvious example of the 1990's is the game Myst (Cyan 1993).
And it does sound like an obvious enterprise: To combine the two giant human activities of stories and games. Add to this that the computer games of today are largely catalogues of popular culture: fast cars, aliens, herds of monsters from hell. But this is possibly because the computer game for all practicality can not tell stories - the computer game is simply not a narrative medium. In actuality we are facing a conflict between game and narrative: They are two separate phenomena that in many cases rule each other out.
The main claim of this thesis is that the computer game and the narrative share some traits - both are temporal, for example - but apart from that are radically different: It may be reasonable to claim that the weight of the narrative comes from a sequence of past events, that have to follow, and that the end of every story gets is power from, if not destiny, then at least some causal logic and inevitability. Interactivity and games, on the other hand, are defined by that the reader/player can influence the events now. Additionally, the lack of a narrator in the computer game makes it impossible to use the novel's interesting devices in the tension between narrator and the narrated. Computer games are interesting for different reasons.
The idea of an interactive narrative has its problems. As the starting quote suggests, I am not the first person to draw this conclusion. I can not claim any originality in this; the value of this thesis is rather that I undertake a detailed examination of the relationship between game and narrative, an examination of how and why they are hard to combine. This also entails an examination of the computer game as a phenomenon on its own.
The techniques of the hypertexts and computer games, combining parts to create something new, have a long history, and do not presuppose computers. I will be discussing both the canonical precursors such as Jorge Luís Borges and I Ching as well as introducing works by the Danish author Svend Åge Madsen.
The term "interactive fiction" was first used in the magazine Byte in 1981. (Aarseth 1997, p.48). It is a loosely defined term, usually understood in the sense that "you play the title role", or that "you are inside the story", which again implies that there is some kind of story at all. The term is notorious of it's unclearness, but the same lack of clearness makes it worth studying as a utopian idea: A combination of the world of narratives with the world of games, where readers/players deeply concentrated participate in a story, continually unfolding and changing in ever more fascinating patterns. This thesis attempts to capture the rhetoric of interactive fiction by examining the paratexts of the computer game: The advertisements, the manual, the packaging. An old ad for the software company Infocom displays the main points of interactive fiction:
[...] as hard as we work at perfecting our stories, we always leave out one essential element - the main character. And that's where you enter in. Once you've got Infocom's interactive fiction in your computer, you experience something akin to waking up inside a novel. [...]Find out what it's like to get inside a story. (Infocom 1984b)
Interactive fiction is described as the addition of literary virtues to the computer game, but paradoxically this will at the very least break with the principles of a classical normative text, the Poetics of Aristotle:
A combination of games and narratives risk ruining both.
Computer games do not seem to be based on narratives: Classical action based games like Space Invaders (Taito 1977) or Donkey Kong (Nintendo 1981) do contain a framing narrative about, respectively, a space station attacked by aliens and a girl kidnapped by an evil ape. It is then the responsibility of the player to right this wrong - chase away the aliens, rescue the girl. This is the structure of a simple narrative: A good situation threatened, the hero has to restore order. But unlike narratives, where a part of the reader's incentive is the desire to know the ending, the ending of an action-based game is known from the start; it is the goal of the player to actualise this good, well-known ending. Additionally the narrative frames are not especially tied to the games; it takes only few graphical modifications to turn Space Invaders (space game) into Centipede (game where you fight spiders and centipede, Atari 1980). A traditional game like chess also has a similar narrative frame, one of two societies at war. But that is hardly the point of chess: Playing a game, one gradually ignores the story and graphics to focus exclusively on the structure of the game, i.e. what manoeuvres it takes to complete the game - no matter what the game "is about". Tetris (Pazhitnov 1985) with the falling squares that have to be fitted contains no frame story or any indication of what you are "really" doing: The squares on the screen seem to be nothing but squares on the screen: You can have a computer game without any narrative elements.
There seems to be a conflict between the temporalities of the game and the narrative: When something is interactive - like a game - the interactivity has to be now, when the player makes a choice. But the narrative has a basic trait of being about something past. Similarly, space is treated differently: Computer games always create space, where the player can move around, but narratives are very focused on skipping uninteresting spaces; a journey is only described when something actually happens. It is essential for the narrative that narration does not happen with constant speed, but that we shift between resume, cuts, and scene. The computer action game is based on real time, on the constant control of the player.
It is a constituting trait of the narrative as such, and of the novel in particular, that the time of the narrator and the time of narrated are distanced in time. And that any novel raises questions of the identity and knowledge of the narrator. This relation between the narrated and the narrator is an important device of the novel. But the computer game does not share this temporal split between the time of the narrated, of the narrator and of the reading: In the computer game, these three times are imploded to a single now. This means that the computer game does not allow for the interesting variations in the relation between narrator and narrated.
Narrativity and interactivity can not take place at the same time: Narration presupposes a jumping and compressed time, interactivity requires real time. Interactive fiction tries to work around this in different ways: Myst contains a story that the player gradually uncovers during the game. 99% of the game takes place after the event, outside the story. In this way Myst escapes some of the inherent conflicts in interactive narratives.
Interaction is a constant problem in interactive fiction: Where the text-based games sometime end in frustrated search of the right command, the graphical games foster irritation that you mechanically have to search the screen with the mouse for the exact spot you can click. In such cases the illusion of the game as a complete, different world breaks down because the interface gets in the way - it's all about mouse and keyboard.
There can be no doubt that computer games are very attractive to many people, but with theory from the narrative media, it is not all obvious why. In his book Moving Pictures (Grodal 1997), Torben Kragh Grodal argues that movies and other narratives need human or anthropomorphic entities for the viewer to retain interest. And this holds true for a classic computer games like Donkey Kong (Nintendo 1981), which has an active actor (Mario). Most computer games do share this trait, but not all games have such an actor. In the previously mentioned Tetris, the player controls the movement of some basic shapes, but you are not associated with a specific character or actor. In Lemmings (Psygnosis 1989), the player must guide a number of autonomous lemmings from an entrance to an exit - in this game you are not represented on screen either; you simply give orders to the lemmings. Accordingly, these games should be marginal and not very popular. But both games were big hits, which leads me to the argument that computer games can be much more abstract that narratives, because they will always involve an active subject - the player. So the fascination of the computer game is not necessarily connected to identifying with a character on screen, but related to the task you have undertaken as a real-life subject.
In Reading for the Plot (Brooks 1984), Peter Brooks defines narrative desire: Narratives tend to be both about desire and to stir the reader’s desire for knowing the ending. The continuing delaying of Scherazade’s execution in Thousand and one nights is a good example of this. In the computer game, on the other hand, the ending is often well known, but it is one you try to actualise by your playing.
The desire to understand and play the game is different from that of the narrative: The average player plays more games of Tetris than the average reader reads Ulysses or watches Gone with the Wind. This seems to be the connected to interactivity, which means that there is no fixed story that the reader/player must wait for the ending of. From this follows, that the more story you attempt to add to a game, the fewer times will the player play it.
In this thesis I will make the argument that the computer game should be seen as a combination of a formally defined level, the program, and a sign-based level, the material. In a simple game like the aforementioned Space Invaders, the graphics (spaceships & aliens) and the narrative frame (Earth attacked) can easily be exchanged with other graphics or another story - the program remains unchanged. The material obviously means a lot to the player’s experience, but it is the program that the player has to master. To examine a specific computer game is then best done by looking at the relation between program and material. For example to see if the material makes claims (such as possibilities of interaction) that are not implemented in the program.
I have to add that I am writing this partly from a position as practitioner. I have developed several computer games, of which some will be discussed here: The two related games Puls in Space and Euro-Space are briefly discussed, the larger project Blackout - where my role was primarily that of programming the Macintosh version - is also an object of analysis.
This has three implications: 1) Many of the arguments put forth here have their root in an actual aesthetic and technical complex; I am basically approving the utopia I am criticising. 2) I have a general technical and detailed knowledge of computer games that the normal player will not have. 3) I am attempting to replace the common sociological and even pathological question "Why do they play computer games?" with an aesthetic and phenomenological question; "What is a computer game?"; what are the limits and possibilities of the computer game?
The theoretical bias is in several directions. In general questions of non-linear texts, the central book is Espen Aarseth’s Cybertext (1997). But my project is primarily a combination of the computer game and the traditional narrative media: Gérard Genette’s Narrative Discourse (1980) is used for discussions in narratology, especially regarding questions of time, I use Peter Brook’s Reading for the plot (1984) to discuss the question of motivation, and finally Moving Pictures by Torben Kragh Grodal (1997) is used for questions of identification. These theories are applied to the computer game to distil general similarities and differences between the traditional medium and the computer game, and these results will be used in specific readings of games.
Space Invaders (Taito 1977) is used to represent the classical action game. As further developments, Tetris (Pazhitnov 1985) and Doom (ID Software 1993) are added. The discussion of interactive fiction is focused primarily at Witness (Infocom 1980), Myst (Cyan 1992), Blackout (Deadline Media 1997), and Last Express (Mechner 1997). With the possible exception of Blackout and Last Express, all of these games can be described as milestones in the history of the computer game. Last Express is used as example of a game that tries to work around some of the inherent contradictions of interactive fiction.
This thesis does not claim that computer games really are great literature - they are not literature at all. Neither do I claim that computer games are simply trashy popular culture - I consider games like Doom and Tetris to be quality. The merits of the computer game are simply different that highly praised literary virtues like character, existential themes, and virtuoso language.
This thesis is in five parts: