digital arts & culture 1998
short paper    

Jørgen Kirksæther

The Structure of Video Game Narration

In this short paper, I'd like to present some aspects of my ongoing doctorate research project, on video game aesthetics, narration and interactivity. Since the project is only a few months old, I'll just dance around a little, and show you my starting points and some ideas. Here, I'll mainly concentrate on games as non-linear storytellers and the underlying structures that control them. One major part of the project that I will not discuss is their actual appearance, in film-speak, their mise-en-scene. I have to have some things left for the next couple of years...


Consider the following passage from the instructions to the game "Silkworm":

(..) Earth's chances of survival hang on (sic) a thread, a thread so gossamer fine that it could be made of silk. Realizing this, the weapons scientists codenamed civilisation's last stand Operation Silkworm. Step forward hero, read the briefing and take the controls...
Now, if we for a minute can put aside the rather ridiculous explanation of the game's name, what happens here? I'd say three things: 1) You're being told the beginning of a story, 2) you're being invited to actively take part in it, and 3), it's quite obvious that the story isn't over.

This is of course old news to anyone who has ever played a videogame. Without #3 there would be no game at all, and without #2 there wouldn't be much point in playing, you'd only be invited to observe the action - that's what we call films... The interesting part, at least for now, is #1, the game designers have started telling you a story, or, in a sense, used the story as a sort of bait: if you play (and buy) this game, you will be told the rest. We then have, on one hand, a story which we are supposed to be told, and on the other are we supposed to physically take part in it, and you will be allowed to alter the story's discourse to some degree. Given this, I claim that some preconceptions about storytelling and narration are in danger. George Landow recognizes this in his "Hypertext (2.0)". Using the hypertext as an example of a user participatory medium, he quite boldly states: "Hypertext, which challenges narrative and all literary forms based on linearity, calls into question ideas of plot and story current since Aristotle." Landow leaves two possibilities: either it's impossible to create narrative in a non-linear environment, or Aristotle was wrong. He then cites several examples of hypertext-written fiction. Case closed.

Or maybe not. The picture is, as always, not as clearly defined as one could want. Landow is of course aware of this. The compromise, so to speak, between embracing and denouncing Aristotle, can be found in the notion of the reader-as-author. As Gunnar Liestøl in his essay "The Reader's Narrative in Hypertext" points out, "Nonlinearity in time is imaginary, it is a fundamental contradiction of terms and necessarily impossible. Time is linear." So, even though the medium in which the story is presented invites non-linear readings, we, the readers, experience the narrative in a temporally linear fashion. In other words: we use the available data and construct our own stories. (Since this personal narrative is the direct opposite of Lyotard's meta-narratives, we could perhaps call the sub-narratives?) Landow quotes J. David Bolter's "Writing Space": There is no single story of which each reading is a version, because each reading determines the story as it goes. We could say that there is no story at all; there are only readings." Again, these readings are what I call sub-narratives.

Computer games can easily be viewed as hypertexts, whether they are text-based adventure games or, as Silkworm, adrenalin-pumping blasters. Since they all consist of a series of projected images, they can also be viewed as films.

What one might call traditional film theory is concerned with a medium which displays its content in a fixed sequence, a sequence the spectator isn't supposed to control at all. This has led theorists like Seymour Chatman to see the spectator as one whose task is to decode, or interpret the film, to discover what is already in there. Others, like David Bordwell, is a lot closer to what the hypertext has made explicit; by also taking the viewer into account. In his "Making Meaning", Bordwell states: "Meanings are not found, but made."

During the last 20 years I have played literally thousands of electronic games, either as computer-/ tv-games or in arcades. As far as I can remember, they all have one thing in common: they give you a fictional framework for you as a player to place yourself within. Most games are brought home from the shop in a box with a short narrative on the back, and even arcade machines have these tiny novellas either printed on the cabinets or as text within the games itself. The only exception I can think of is pinball machines, where the only written instructions most of the time is something like "shoot left ramp for double bonus". Pinball games are obviously targeted at a very specialized audience, an audience that knows that all the different machines are based the simple idea of not letting the ball drop through the hole at the bottom. This, combined with the fact that half the fun with pinballs is the feeling of discovery every time you uncover yet another design detail, makes pinball instruction sheets rather sparse. "Insert coin and don't lose your balls."

And even so, the pinball designers still rely on fictional narrative as a "hooking device", whether they base their designs on popular films or science fiction TV-shows like Star Trek. You don't get in-game instructions like "shoot the ball into the thing on the left to get more balls on the board", the preferred version looks a lot more like "shoot ball into wormhole to enter warpspeed and release multiball." Even though I made the above examples up, you can find similar ones in all pinballs.

Videogames, whether in arcades or on computers or consoles, employ a slightly more elaborate strategy for easing the player into the game. Part of this of course derives from the fact that not all games are based on the same game-device. Although you in a lot of cases could restrict yourself to "press fire and don't get hit", this never happens. You are almost always told that you "are" someone and that "you" must do this or that. This beginning of the game is the same for everyone, and it is at least supposed to be "read" before you start playing.

When and how a game ends is a little more unclear, and varies from game to game and from genre to genre. But, whether you get the proverbial "Game Over" or if you simply decide that you've had enough and shut down the machine, you can not escape that the game does end, or rather, the reading of the game ends.

Then, on the surface, this leaves us with good old Aristotle again, with a beginning, an end, and consequently; a middle. This middle is the really interesting part.

Games are constructed, and their constructors are called designers (who may also be programmers) (in addition to graphic artists and musicians). The designer's primary task is to think up an objective for the game, and obstacles that will make it difficult for the player to reach it. In other words: to give you a mission and make it near but not completely impossible to complete it. Very few films, at least until recent years, have been designed to the same degree as games. True, you have costume- and set-designers, but they do nothing but tweak the way the film looks. The film equivalent of the game designer is a cross between the script writer and the director. Their task has traditionally been to devise a way to tell a story in a straight temporal line. Given the cyclic structure of games, the games designer faces other challenges.

If you come home with a new game and finish it in a couple of hours, you feel cheated. Since a computer game, at least here in Norway, costs about ten times as much as a movie ticket, you expect it to last a little longer. It doesn't matter if you had a fabulous time those two hours, it just isn't value for money. So, the game designer puts obstacles into the game. They force you to replay game segments until you master them, which in turn moves the game "story" forward. In a shootemup this would be like progressing to a new level, in an adventure like gaining access to new areas. This underlying structure results in something interesting: games are almost never played in straight line, plot-wise. Instead, you move through them in circles.

This elliptical structure appears at several levels of the game reading. This varies from game type to game type, but is, in my opinion, central to any type of game. Reading in such a way is not only different from reading film, but also from hypertext fiction. The "standard" hypertext isn't as much designed for user input as it is for user control. Some hypertexts may be written/designed so that the reader can only visit certain nodes after having visited certain others, but this isn't a central issue in hypertexts. I can not, however, remember playing one single game, (with the exception of computerizations of board games like Chess and Othello, but I see these as hybrids and therefore not really relevant to my discussion) where all locations/levels/boards/whatever were accessible when the game started. This leads me to believe that the two reading experiences in some ways (but certainly not all) are quite different. In a hypertext fiction you are more or less on for the ride, you click along to see where the road ahead may lead. When playing a game, you have a set goal, the gaming experience comes in the back door so to speak, as a bonus when trying to reach it. It can be likened to reading a novel where your task is to get to the last chapter, but where skipping sections is not allowed.

When all the logic structures are revealed, one would think that the game would be discarded. For some games, though, this is not true. A completed adventure is seldom booted up again, but another round of Pac-man, even after you have figured out Inky and Blinky's every move, is still a blast. Obviously there must be different rewards in playing different types of games. In Pac-Man's case, one could say that there are two set goals; the low-level one of clearing the board of dots, and the high-level one of clearing as many as possible. How many times can you do this before you've had enough?

Game designers use different strategies in trying to extend the playing. The immediate one is to make the game "larger", for example by increasing the number of locations in an adventure. This in itself will not be enough, an expansion of the underlying plot is also needed, otherwise these new locations will never be reached. The equivalent in a shooting game is to add new types of enemies and new playing fields.

A second option is to introduce elements of randomness. This can make even the most convoluted game interesting. A game like Space Invaders quickly gets boring, simply because it's nothing more than an exercise in manual dexterity. If you introduce more invader movement patterns and some randomness to them, you not only have a different game, but also a game it would be fun to play for a while.

If computer generated randomness isn't enough, you can always throw in that super-unpredictable factor: another human. Socialising factors do play a part in the popularity of multiplayer games like MUDs, but one can not overlook the human random factor when viewing them as "pure games". I have in this project, for several reasons, decided not to include multi-player games, but I feel that it was right to at least mention them here.

Playing a game involves manipulating a graphic interface between the player and the game logic. The number of control options is usually not particularly high, simply because it would otherwise take too much time to learn them before you could start playing. I'm not sure if you really could train a monkey to play Space Invaders, but it probably isn't far from the truth. The appeal of games isn't in mastering a complicated set of controls, but rather in submitting to a set of rules and trying to accomplish something under these rules' restrictions. When you play a game, it may seem like you are "playing the interface", but what's really happening, is that you're trying to figure out the game's logic. Yet, it's the interface just as much as the inner workings of the game that make up the total gaming experience. I think I should emphasize that I do not think of games as a three-layered system where the layers are self-contained. I do not believe that it is possible to separate the graphic interface from the game logic and structure, just as I don't want to look at the interface without a player. I find Donna Haraway's cyborg concept appealing when looking at this side of gaming. Although Haraway's cyborg is a political construct and not an analytical tool, I see it (him? her?) as useful when looking at videogames. What are we when we play, if not man-machine hybrids?

As I started by saying, this is just a presentation of a rather fresh project, and its starting points and ideas. Yet, it has given me the opportunity to try and condense some thoughts that needed condensing. I hope that at least some of you have been given some new ideas.

Jørgen Kirksæther is a doctoral student at the Department of Art and Media Studies at the Norwegian University of Science and Technology in Trondheim.





about Bergen

Send us an e-mail if you have any questions.
Last update: 26th of October 1998.