A Clash between Game and Narrative

Jesper Juul


Jesper Juul: "A Clash between Game and Narrative". Paper presented at the Digital Arts and Culture conference, Bergen, Norway, November 1998.



There's a conflict between interactivity and storytelling: Most people imagine there's a spectrum between conventional written stories on one side and total interactivity on the other. But I believe that what you really have are two safe havens separated by a pit of hell that can absorb endless amounts of time, skill, and resources.

-Walter Freitag, game designer.

... the fundamental qualities that make a good game have remained unchanged and elusive. Consumers still flock to buy original, addictive, and fun games, leaving many flashy products with million-dollar budgets languishing in the $9.99 bin. These costly failures demonstrate that the consumer does not desire a cinematic experience, but rather a quality gaming experience.

-Sid Meier, game designer.

To briefly introduce myself: My name is Jesper Juul, I'm currently finishing my master’s dissertation in Danish literature on the subject of interactive fiction, which is incidentally also the subject of this presentation. I've also designed and programmed several computer games, some on CD-ROM, including one "interactive fiction", some on the web, I’ve done chat, and I've worked with computer-based art.

In this paper presentation I'll be making a simple point. That computer games and narratives are very different phenomena and, as a consequence, any combination of the two, like in "interactive fiction", or "interactive storytelling" faces enormous problems.

I'm not the first person to make that point. The merit of this presentation is hopefully the detail with which this point is made. But it is slightly strange to be saying this. On one hand, it seems that the idea of an "interactive narrative" died commercially around 1993-94. On the other hand, much work and effort is being put into claims that game and narrative can be mixed- witness Janet Murray. And the dominant theoretical way of dealing with computer games still seems to be claiming that they are in some way narratives.

But computer games are not narratives. Obviously many computer games do include narration or narrative elements in some form. But first of all, the narrative part is not what makes them computer games, rather the narrative tends be isolated from or even work against the computer-game-ness of the game. I'll briefly try to isolate that gameness, and to sketch a way of saying something meaningful about a computer game.

The main point of this paper does clash with several to be presented tomorrow. Since fighting over words tends to be unfruitful, I'll mainly be pointing to characteristics of the traditional narrative media and compare them to the computer game. But I do think that the term narrative doesn't fit the computer game very well.

This happens in four parts.

  1. A brief examination of the rhetoric of interactive fiction.
  2. A theoretical discussion of the differences between computer games and narratives.
  3. A discussion of the narrative frames that most computer games do have.
  4. Conclusions.

1. The rhetoric of interactive fiction

IF as a concept is not very well defined. In Cybertext, Espen Aarseth has plainly rejected the IF term for its unclearness, but on the other hand it is being used to convey some specific connotations.

It is a word that belongs to the marketing people. But it is has its problem, the first problem being the second word: fiction. In English this is often taken to mean novels, as in the bookstore. Or a magazine like Modern Fiction Studies. IF is basically understood as interactive "narrative" or "story".

"Interactive fiction" is then an attempt at combining games and narratives. This combination sounds extremely attractive, and it is usually described as the best of both worlds, where the reader/player deeply concentrated can participate in a story that unfolds in new and ever more interesting patterns. To exemplify this, we can look at a 1983-ad from the company Infocom:

[...] 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 1983)

You are "inside the story". But the major part of IF rhetoric always seems to very much be a revolt against things, IF doesn’t want to be.

  1. For one thing, IF doesn’t want to be an action game with all its violence, facile themes and focus on quick responses and joystick mastery:

    Myst is real. And like real life, you don't die every five minutes. In fact you probably won't die at all. [...] The key to Myst is to lose yourself in this fantastic virtual exploration and act and react as if you were really there. (Myst manual, Cyan 1993.)

  2. The subculture of textual IF doesn’t want to be based on graphics:

  3. We unleash the world's most powerful graphics technology. You'll never see Infocom's graphics on any computer screen. [...] We draw our graphics from the limitless imagery of your imagination - a technology so powerful, it makes any picture that's ever come out of a screen look like graffiti by comparison. [...] Through our prose, your imagination makes you part of our stories, in control of what you do and where you go - yet unable to predict or control the course of events. (Infocom 1984.)

    This clearly tries to appeal to the intellectual user who prefers novels over movies, worships the imagination and so on.

    Later events, however, led to the decline of the IF genre. As the educational level of the average computer user decreased and the features and capabilities of the average computer increased, the trend in computer games went to 'arcade' games instead of text. (rec.arts.int-fiction FAQ)

    This is the nostalgic view of IF - the golden age we’ve lost.

  4. Graphical IF doesn’t want to be textual, because that is just too nerdy.
  5. IF doesn’t want to be based on puzzles, rather IF want to be about character and human relationships.
We're trying to make a game that is based on plot and characterization, not puzzles - the way a book is. If you read Fahrenheit 451, you don't get stuck on page 50. And if you play the game, you don't get stuck on frame 50, because the whole idea is that you're interested in the game because of the characters and the plot and what's happening. (Interview with Trillium software, Darling 1984.)

All of this has been constant almost for the entire duration of the genre. Most "interactive fictions" will claim to match one or more of the above points. Interactive fiction is then in reality largely the rhetoric for a Utopia, a promise of a new and more intellectual/cultural kind of computer game. One that is basically higher culture than the low-culture action computer game.

But IF has its problems.

2. Game vs. narrative


The most popular comment on computer games is to say that they are non-linear / multicursal, meaning that they differ from narratives because they can be different sequences. But it does seem reasonable to claim that narratives are sequences evoking a sense of destiny, of events that have to lead to each other. Roland Barthes says that narrative is the language of destiny. "the mainspring of narrative is ... what comes after being read in narrative as what is caused by."

I think this is a very precise statement. Sequence matters in narratives, and the famed translatability of narratives between different media does presuppose fixed sequences. While you can recognize, say Sherlock Holmes or Knut Hamsun's Hunger between novels and movies, you clearly can't deduct the story of Star Wars from Star Wars the game.

Unlike the fixed sequence of the narrative, Games seem to be based upon the relative freedom of the player, on the players' possibility of influencing the course of action.


There are some very interesting temporal differences between games and narratives. (I'm mainly drawing upon narratology as in Gerard Genette's book Narrative Discourse:) Narrative is a ... double temporal sequence ... : There is the time of the thing told and the time of the narrative (the time of the signified and the time of the signifier). This duality not only renders possible all the temporal distortions that are commonplace in narratives (three years of the hero's life summed up in two sentences of a novel or in a few shots of a "frequentative" montage in film, etc.). More basically, it invites us to consider that one of the functions of narrative is to invent one time scheme in terms of another time scheme. (Christian Metz, quoted from Genette 1980, p.33)

Some of you might recognise this quote, as it's been quoted by Gunnar Liestøl in Hyper/text/theory. It tells us something interesting about the narrative - it presupposes two different times, interacting. In the traditional view of the narrative, you divide between the fabula and the sjuzet, or if you will, the story and the discourse. Reading a novel, you construct a storyline from the discourse presented to you, mostly in non-chronological order. Part of the traditional novel-format is a narrator recounting previously happened events. So we get time of the narrated, time of the narration and time of the reading.

But if we play a game like Doom, these temporal distances are clearly not present. You can press the control key, a gun will be fired, and this will affect what's happening on the screen. What you see on the screen can't be past or future, but must be present, since we can influence it. So the three times, the time of the narrated, time of the narration and time of the reading implode in a game, and every time you have interactivity.

Equivalently, and as a consequence of the interactivity, games do not use the temporal possibilities of the story/discourse pair. You don't get flashbacks or flash forwards while playing Doom, because such variations would preclude the interactivity: In a game, you are not able to first play a scene in the present, and then jump to an earlier point on the time line and have interactivity there. Because the first scene would then be determined by whatever the player does earlier on the time line. This would be to a classic time travel paradox.

The story/discourse pair is in other words meaningless in the computer game. The computer game simply doesn't have an active dualism like that.

If we go back to the Christian Metz quote, it seems quite clear that the computer game doesn't "invent one time scheme" from another time scheme. Which could indicate that it is not really narrative.

3. Narrative frames

On the other hand, most computer games do feature some kind of narrative framing. Take Space Invaders.

When we are urged to "play Space Invaders", it does imply a minimal story. The concept of invasion presupposes a time before the invasion, and from the 1950's science fiction it draws upon, we just know that these aliens are evil and should be disposed off. So there is a story, and from the title screen we know all of it: Earth attacked, Earth freed from the alien menace. This is the basic mode of the classical action game: A clicheed story with a well-known ending, and a game that actually never reaches that ending, it just gets harder and harder.

I wrote two simple computer games recently. The first is called Puls in Space and was made for a youth-oriented Danish television program called Puls. It's the kind of thing where you have a variety of different hosts and a lot of young people are very actively engaged in evaluating them. In this game you naturally control a spaceship and get attacked by the heads of the hosts.

Another game I did is called Euro-Space. This was done for a Danish EU-sceptic organisation, and you obviously fight various symbols of the Danish EU-debate.

The program and the gameplay is identical for the two games. It's just the graphics that are different. It would be equally easy to do a pro-EU game, just replace the symbols.

As you can see, the symbolical or metaphorical meaning of the game is not connected to the program or the gameplay. The relationship is, in a word, arbitrary.

When I was 14 years old (this was in 1984) I got the computer game Neoclyps for Christmas. The box of Neoclyps says: You are the good guy. Your mission is to destroy the bad guys.

This means that the narrative frame, especially in the action game, has always been considered unimportant. It's something used for selling the games, for having a way to refer to them.


How does this then work in a supposedly storytelling game like Myst? Due to the various differences between games and narratives, there is no simple way to do an interactive narrative, and Myst works around this problem quite cleverly. Except for the very end of Myst, you are not really interacting with a narrative. You're certainly not the major character either, just a minor one who's happened to pop by long after the event. The way Myst works is that you pick up various artifacts: A diary, you see a video clip you can't interact with. A story is told by these artifacts within the real-time of the game, but it is a story of something that happened before you entered the world of Myst. Myst is then actually not an interactive narrative, but it shows a fairly good way of adding narrative material to a game without killing the game. It is the detective-format, where the detective works to uncover earlier events from traces he/she finds.

Doom, Quake

Two of the most popular games in recent years, Doom and Quake are justly famous for their lack of storylines. Legend has it that ID Software only refer to the concept "story" as "the s-word". In a recent interview, ID head John Carmack says of the upcoming Quake III that it will have "The best graphics, the best networking, the best gameplay - but no plot."

While it's perfectly possible to find best-sellers that try to incorporate narrative elements - Myst is one example - it's also perfectly clear that the most played games are low on plot but high on gameplay. In the PC Cafes or in the competitions, they're not playing Myst, they're playing Quake, Unreal, Starcraft. Real-time games with little or no storyline to speak of.


Another question is the question of repeatability. In literature there is an idea of the endless work, of books you can read and read, and never tire of. This can both be a religious work like the bible, or a modernist work like Ulysses or The Wasteland. Contrast this with the term trash novel, implying that a book is disposable once read. It does seem that repeatability is perceived as connected with high culture, the reverse with low culture. The surprising part is that the notoriously "low" computer game lives up to this much more than novels tend to. The dominant mode of receptions of narratives is one-shot, but games are inherently something you play again, something you can get better at.

It then appears that trying to add a significant story to a computer game invariably reduces the number of times you're likely to play the game. Literary qualities, usually associated with depth and contemplation, actually makes computer games less repeatable, and more "trashy" in the sense that you won't play Myst again once you've completed it. There's no point.

This doesn't mean that Tetris is an endless work that can always be reread and always sheds new light on the world - for we usually do not see computer games as statements about something else. But it seems paradoxical that introducing a narrative reduces the number of times you play the game.

On closer examination this is not that surprising. A lot of the reason for reading a novel, or enjoying a narrative is the desire to know the ending. Peter Brooks has described this quite well.

But the reasons for playing are completely different. The frame stories that most computer games have are ridiculously shallow and clicheed. It does mean that there is often no kind of narrative desire to reach the ending of the game (and it can't even be reached in many action games - they just go on and on). The computer game is rather based on two different types of desire. The first is the desire for a structural understanding of the game; to understand how the game works. In Doom this would be knowing how the monsters move, which buttons to push to open a door. The second is the desire for having the performative skills needed to actualize this knowledge.

In an action game, the desire for understanding the game world and for the skills to actualise this knowledge is something independent of whatever narrative frame there might be.

4. A model of the computer game

I am basically proposing that computer games should be viewed as a dualism of two layers. The program and the material.

These two layers belong to distinctly different traditions. The material; the graphics, sound, narrative frame belongs to the traditional media. This is what the aesthetic faculties of universities know how to handle. The other level is the program, which is the new thing in this context. The program is perfectly formal, it works on a purely electrical level.

To present such a dualism between an underlying formal layer and its accidental appearance is probably tragically unhip these days. Present-day literary theory would love to spend energy trying to point out that, say, the supposed underlying formal layer is actually something constructed from the interpretative layer. While I think such an objection is perfectly true for literature, for narratives, the computer game differs radically in that the two layers can be taken apart.

I think that the program/material relationship is the most interesting thing to study in a computer game. If we look at Myst, there is a very clear incompatibility between the claims of the packaging and the graphics vs. what the player actually can do. According to the packaging, you have been sucked into a book, you are in a new world which you are free to explore. The graphics suggest that you are able to interact with everything.


For some reason, you can manipulate the switch, but you can't touch the small ship. Why? Well, the program is not capable of doing everything the material suggests is possible.

Simple games like Space Invaders are characterized by that the material doesn't promise something the program can't keep, the same goes for an action game like Doom.

The material can then more or less successfully fit the program. Myst partly fails on this issue - you can't do what the material promises.

Conclusions and perspectives

The most widely accepted games are clearly those exhibiting the most traditional aesthetics. They also tend to be considered actual culture and be reviewed by traditional-media reviewers. And a game like Myst has the quality of being representable in a traditional medium like the newspaper. You can see the pictures, the narrative frame can be summarized. The accepted games are clearly those carrying the most luggage from traditional media and aesthetics. But they're usually the worst games.

A game like Tetris, on the other hand, is a very popular game. It surely is one of the most implemented games ever. There are countless versions of it everywhere. On every computer and in mobile phones and digital watches. But Tetris looks dull in the paper, it has no story. And imagine a narrative as abstract as Tetris. This would be out of the question. Stories need human or anthropomorphic characters. Games don't.

To sum the theoretical part, an Interactive fiction, such as a Choose Your own Adventure book, an interactive movie like Urban Runner or Wing Commander III, or any kind of "interactive story" works by switching between two temporal modes, the narrative mode and game mode.

Anyone who's played a few modern "cinematic" interactive fictions will testify that they are really awful games and stories. You are trapped by unmotivated shifts between the narrative mode and the game mode, the story gets destroyed by the interactivity, the interactivity gets destroyed by the story.

This leads us to a final comparison of the relationship between narratives and computer games:


Computer games

Fixed sequence

Flexible sequence

Variable speed (usually compressed)

Fixed speed





Needs human or anthropomorphic actors

Can be abstract

Narrative desire

Desire for understanding and performance

Consume once

Play many times

  • Narratives are fixed sequences, games are flexible sequences.
  • Narratives vary in the speed with which they are told; uninteresting periods of time are skipped; the movie 48 hours doesn't last 48 hours. Computer games, especially the action game, are fixed speed, real time.
  • A narrative has a dualism between the story and the discourse, the computer game is divided between the formal program and the material.
  • A narrative is basically something past, a computer game something present.
  • A narrative needs human or anthropomorphic actors, a game can be abstract. You can’t imagine a narrative as abstract as Tetris.
  • In a narrative, the reader desires to know the ending. In a game, the player wants to understand the structure of the game and to acquire the skills to use this knowledge.
  • A narrative is something you consume once, a game is something you play many times.

To sum it up. Computer games and narratives are very different phenomena. Two phenomena that fight each other. Two phenomena that you basically cannot have at the same time. Any interactive narrative or attempt at interactive storytelling is a zigzag between these two columns.



This paper was presented in an interesting situation, where at least two other papers were based on premises opposed to this papers conclusions. I should make clear that I do agree that many computer games contain narrative elements, and that in many cases the player may play to reach a narrative sequence. My point is that all such computer games are a conflict between the now of the interaction and the past of the narrative. You can't have narration and interactivity at the same time.

I'm not saying this to suggest that games and narratives can't or shouldn't be combined. It's very dangerous to claim any problem or conflict to be of no aesthetic value. It might happen, somebody may find a way to use this clash for artistic or entertainment purposes. I'm just saying this to point to a list of standard problems inherent in the idea of interactive stories. There's is no such thing as a continuously interactive story.

Briefly thinking about the future of this field, I do see progress. In literary theory, it has always been presupposed that one has read perhaps 1000 books and seen a 1000 movies. But when studying computer games, it has been acceptable to play four computer games and then write articles about it. Fortunately, we seem to be slowly moving from seeing the computer game as a sociological or even pathological phenomena towards seeing it as an aesthetic object to examine.

Hypertext, and the computer as such, has generally been linked with the postmodern (or even the poststructural). But is the computer game postmodern? Literary theorist Brian McHale has suggested that the difference between modernism and postmodernism is that modernism is epistemological; oriented towards knowledge and the conditions for our knowing the world, whereas postmodernism is ontological; oriented towards creating fictive worlds. The computer game is kind of hard to place. The player clearly tries to discover how the game is structured - which is epistemological. But creating a game is clearly creating a world, and one which is usually without special reference to anything. So computer games do not fit neatly into one category or the other.



Roland Barthes: Image, Music, Text. Fontana, 1977.

Peter Brooks: Reading for the Plot. Knopf, New York, 1984. Harvard University Paperback Edition, 1992.

Cyan: Myst. Brøderbund, 1993.

Sharon Darling: "Byron Preiss and Ronald Martinez. Trillium Software Designers". In Compute's Gazette. December 1984.

Gerard Genette: Narrative Discourse. Cornell University Press, 1980.

ID Software:

  • Doom. GT Interactive, 1993.
  • Doom II. GT Interactive, 1994.
  • Quake. GT Interactive, 1997.


  • "We unleash the world's most powerful graphics technology". Ad in Creative Computing. September 1983 p.112-113.
  • "And now for something incompletely different!". Ad in Compute's Gazette. December 1984, p.14-15.

Jesper Juul:

George P. Landow: Hypertext 2.0: The Convergence of Contemporary Critical Theory and Technology. Johns Hopkins University Press, 1997.

Sid Meier: A Revolution. In Game Developer. April-May 1997 p.72.

Alexey Pazhitnov: Tetris. Spectrum Holobyte, 1985.

Charles Platt: "Interactive Entertainment". In Wired 3.09, 1995

Taito: Space Invaders. 1977.