Games as Artistic Media (Or Not) - Game Studies (Part 2)

Previous Chapter >>

Contrary to the views of the writers of Computer Games, Aarseth is happy to describe games as a new form of literary text in his book Cybertext: Perspectives on Ergodic Literature. Games are here described as a specific kind of cybertext, and cybertext as a particular kind of what Aarseth terms ergodic literature, that is, literature where non-trivial extranoematic effort is required by the user in order to traverse the text. Extranoematic means, simply, a process which occurs outside of the confines of human thought. That is to say, in this case, that the user is working in or with the text at a level which is more involved than even in reader-response theory, and in a way which is outside the normal concepts of 'reading'. For this reason, we infer, it is more proper to call a person experiencing ergodic literature a 'user' rather than a 'reader'.

Examples of ergodic literature include the I Ching, in which the user uses either three coins or forty-nine yarrow sticks to randomly generate two hexagrams out of a possible 64, which are the binary combinations of six lines (whole or broken). The text associated with the two hexagrams generated holds the answer to a question the user wrote down beforehand. Other examples citied are Guillaume Apollinare's “calligrammes,” which are poems presented in such a way that the words form an image on the page, with no clear order in which to be read and Ayn Rand's play Night of January 16^th which features a trial in which members of the audience are selected to form a jury. The play has two different endings, depending on the verdict of this jury.

Cybertext seems to be a more specific type of ergodic literature, although Aarseth sometimes seems to use the terms interchangeably. The key feature of cybertext as a literary structure is it's non-linearity. The word non-linear is not here used to mean a text of ambiguous meaning, or one which lacks a traditional narrative structure, but rather that the text is capable of producing a variety of expression dependant on the input of a user.

What this means is that, as Aarseth explains later in the chapter, in a text which enables the user to make actual choices which effect the text (as opposed to the choice of different interpretation), the player will necessarily experience some things and not others – a player may have many options for different paths, but once one has been chosen, the others can't be experienced without starting the game again or returning to an earlier point. This means that the experience of a game is non-linear, and variable in the sense that it is different depending on the choices made. The fact that the user has had a different experience means that the impression they receive or the effect the text has on them differs. Hence, “variable expression” is allowed for – the creator, by giving multiple possible pathways (debatably, these are infinite in any game) can express (and therefore impress upon the user) something different for each play-through, as the result of a variance in paths chosen, and something different for each user not because they interpret the text differently, but because they are actually presented with a different text.

This is possible because in cybertext, unlike in traditional narrative, the reader is part of the story. In a traditional narrative, the reader takes the role of a voyeur – observing events but not part of them, and not at any risk themselves. In cybertext, the user is a player in the sense that although they cannot alter the labyrinth that the text makes up, they can choose which paths to take. They are also at risk in the sense that depending on their choices they may experience, according to Aarseth, intimacy or failure. A player who carefully explores a game and is able to assemble the different pieces of the story into a complete picture might be said to have achieved this intimacy with the text, where as a failing player may find the text confusing or unavigable. This is the fundamental difference between a traditional narrative and a cybertext, and what allows them to be used for variable expression – the difference of the structure of the text means that it demands a user or player who is actively involved in shaping the text as opposed to a reader who is only involved in shaping their own interpretation of the text.

Aarseth also provides a description of the components of a “Generalized, Role-Playing Cybertext” in his book. This description can easily be applied to games in general, and is illustrated in the diagram below¹

These components are, from left to right, the database, which contains the details of the 'game-world' and data on the characters and objects in it (data which can be static – unchangeable – or dynamic – changeable by player actions or other events in the game-world). Next is the simulation/representation layer. The simulation engine is what analyses player input and determines the results of their actions, and the representation engine feeds back these results to the player. The interface level is concerned with, in the case of analysis, interpreting player input into a language the simulation can understand, and, in the case of synthesis, in expressing the information from the representation engine in a form understood by the player (in a video game, through graphics, audio cues, etc.). The user is who ever is playing the game – giving out input and receiving the output. This fundamental view of what a game consists is useful in that it allows one to draw similarities between video games and other kinds of games which share this same fundamental structure, such as pen and paper role-playing games. However it is not detailed enough to determine whether the games are artistic or not.

Next Chapter >> 
 

1 Diagram from page 104 of Cybertext.