Towards Computer Game Studies

Towards Computer Game Studies

2004-05-22

Literature scholars eager to understand gaming have made early inroads. Markku Eskelinen sets up serious checkpoints.

Introduction: LudologyA concept introduced to computer game studies by Gonzalo Frasca in 1998.and Narratology

It is relatively stress-free to write about computer games, as nothing too much has been said yet, and almost anything goes. The situation is pretty much the same in what comes to writing about games and gaming in general. The sad fact, with alarming cumulative consequences, is that they are under-theorized; there are Huizinga (1950), Caillois (1979), Ehrmann (1969), and Sutton-Smith (1997, Avedon and Sutton-Smith 1971) of course, and libraries full of board-game studies, in addition to game theory and bits and pieces of philosophy – most notably Wittgenstein’s – but they won’t get us very far with computer games. So if there already is or soon will be a legitimate field for computer game studies, this field is also very open to intrusions and colonisations from the already organized scholarly tribes. Resisting and beating them is the goal of our first survival game in this paper, as what these emerging studies need is independence, or at least relative independence.

It should be self-evident that we can’t apply print narratology, hypertext theory, film or theater and drama studies directly to computer games, but it isn’t. Therefore the majority of the random notes and power-ups that follow will be spent modifying the presuppositions firmly based on the academic denial of helplessness. Obviously I need a strategy, and fortunately I have one: to use the theories of those would-be-colonizers against themselves. For example, as we shall soon see, if you actually know your narrative theory Those who see and wish to see narratives everywhere (to me, a serious disorder in aesthetic pattern recognition) should at least know their narratology, which is usually not the case. Narrative is a contested concept for sure, but it still doesn’t make sense that comparisons between narratives and games, as well as those between print and hypertext narratives, are and were based on seriously outdated and unsophisticated theories of narrative. In order to make any reliable claims for novelties or similarities between modes and media, one should (at least) first gather the most sophisticated knowledge there is; let’s say combining formal narrratology (Genette, Prince) with the narrative tricks and treats of postmodernist fiction that once again reconfigured the relations between narrative and textual designs (see McHale 1987, 1992), and the tradition of procedural writing (especially various poetics of the OuLiPo; see Bénabou 1998) – and then transform that knowledge into the digital realm, perhaps through Aarseth’s cybertext theory (Aarseth 1997) and its functional and heuristic map of the textual medium (a seriously understudied dimension of traditional literary studies). It’s painfully obvious this is not the case, and narrative is just another marketing tool used to sell us everything else except narratives. To complete the irony, it could be observed that various poetries and poetic practices (such as John Cayley’s programmatology, Eduardo Kac’s holopoetry, and Loss Pequeno Glazier’s kinetic works) which give their strings of signs different durational values are much “closer” to games than print and classic hypertext narratives with their static (permanent) scriptons and intransient time.(instead of resorting to outdated notions of Aristotle, Propp, or Victorian novels) you won’t argue that games are (interactive or procedural) narratives or anything even remotely similar. Luckily, outside theory, people are usually excellent at distinguishing between narrative situations and gaming situations: if I throw a ball at you, I don’t expect you to drop it and wait until it starts telling stories.

It’s good we don’t have to start from scratch, as there have been attempts to locate, describe, and analyze the basic components and aspects of the gaming situation, which are essentially different from the basic constituents of narrative and dramatic situations. I’m thinking here of Chris Crawford’s early classic The Art of Computer Game Design (1982) (and its comparison of games and computer games in particular), Gonzalo Frasca’s (Frasca 1998, 2001) and Jesper Juul’s (Juul 1999, 2000) papers and theses on ludology, and most of all Espen Aarseth’s articles on computer games and cybertext theory (Aarseth 1994, 1997, 1998a, 1998b, 2001a, 2001b).

To begin, I’d like to demonstrate or test a safe and painless passage from narratives to games by trying to “exhaust” classic narratology (Chatman 1978, 1990; Genette 1980, 1988; Prince 1982, 1987; Bordwell 1984). Most naïve comparisons between narratives and games usually result from too narrow, broad or feeble definitions of the former: usually it comes down to discovering “plots” and “characters” in both modes – games and narratives. However, we should know that’s not good enough, as we can find those events and existents in drama as well, which is clearly its own mode. The minimal definition of narrative derived from Gerald Prince and Gérard Genette states basically that there must be two things or components to constitute a narrative: a temporal sequence of events (a plot, if you want to water down the concept) and a narrative situation (with both narrators and narratees for starters). I think we can safely say we can’t find narrative situations within games. (Or if we sometimes do, most probably in Myst or The Last Express, the narrative components are then at the service of an ergodic dominant).

In short: a story, a backstory or a plot is not enough. A sequence of events enacted constitutes a drama, a sequence of events taking place a performance, a sequence of events recounted a narrative, and perhaps a sequence of events produced by manipulating equipment and following formal rules constitutes a game. This is really very trivial but crucial; there are series and sequences of events that do not become or form stories (in Tetris, for example). The reason for this is equally simple. There are plenty of reasons, of course. The main thing is that any element can be turned into a game element, and a single element is enough to constitute a game if it allows manipulation, and this fact alone allows combinations not witnessed in narratives or drama. Consequently, both the number of game elements and the relations between them can be different in specific ways that are typical of (computer) games and only of them. In games, the dominant temporal relation is the one between user time and event time and not the narrative one between story time and discourse time.

Regarding the fallacy of recognizing similar characters or existents in games, drama, and narratives, the situation is similar. In computer games you can operate your character, if there is one in the first place, This is crucial too, as from chess and soccer to Tetris, games have managed quite well without characters. perhaps also discuss with other characters or voices; and the characters can be dynamic and developing (not only in an interpretative sense), such as by changing themselves with level points and power-ups. Such “characters” are entirely functional and combinatorial (a means to an end); instead of any intrinsic values, they have only use and exchange values to them. These entities are definitely not acting or behaving like traditional narrators, characters, directors, and actors, their supposed counterparts in literature, film, and the stage.

To summarize: different existents, different event structures, and different situations. On the other hand, narratology is not completely useless, if its key concepts and distinctions are not taken for granted but traced back to their roots. In the following pages that is exactly what we try to do. We’ll discuss the gaming situation and game time in separate sections; this division mimics Genette’s presentation of tense, mood and voice in Narrative Discourse (1980). The elementary categories of classic narratology are transformed into an open series of ludological components, if for no any other reason than to further specify the features inherent to games.

Before going into the finer points of ludology, the more or less peaceful coexistence of local traditions and global technologies should also be acknowledged. There’s no guarantee whatsoever that the aesthetic traditions of the West are relevant to game studies in general and computer game studies in particular. It’s tempting to assume that one reason for the never-ending series of unsuccessful game definitions and disciplines is the need or urge to make clear-cut distinctions and compartmentalize aesthetics. To take an obvious counterexample: according to the Natyasastra every art contains parts of other arts. This Sanskrit classic is then about remediation some 1,500 years before Bolter and Grusin (1999). There are important differences of course. The fact that dance theatre contains elements of music doesn’t turn the latter to the former in the Natyasastra. But for Bolter and Grusin, computer games are audiovisual narratives, because they seem to contain cinematic components. It would be almost equally sensible to speculate on Japanese aesthetics (after Keene 1995) and claim that a tradition that emphasizes the values of perishability, suggestion, irregularity, incompleteness, and simplicity, is perhaps better suited to approach computer games than its Western counterpart. Of course this is a broad generalisation, but an educated one.

The Gaming Situation This section is kind of a footnote to Eskelinen (2001b), where the gaming situation and its spatial, causal, functional, and temporal parameters are studied and articulated more fully.

According to David Parlett, formal games are systems of ends and means (Parlett 1999, 3). The latter part consists of specific procedural rules of how to manipulate the equipment (pieces or tokens or whatever). In computer games there are events and existents, the relations and properties of which the player has to manipulate or configure in order to progress in the game or just to be able to continue it. Events, existents, and the relations between them can be described at least in spatial, temporal, causal, and functional terms. It’s equally self-evident that the importance of these dimensions varies from game to game and sometimes also within the phases and levels of an individual game.

A quick look at Espen Aarseth’s typology of cybertexts (Aarseth 1997, 62-65) should make us see that the dominant user function in literature, theater and film is interpretative, but in games it is the configurative one. To generalize: in art we might have to configure in order to be able to interpret, whereas in games we have to interpret in order to be able to configure, and proceed from the beginning to the winning or some other situation. Consequently, gaming is seen here as configurative practice, and the gaming situation as a combination of ends, means, rules, equipment, and manipulative action.

Jacques Ehrmann understood games as economy, articulation, and communication, and the player as both the subject and the object of the game (Ehrmann 1969, 55-57). The levels of articulation as specified by Warren Motte – the relations of player-to-game, player-to-player and game-to-world (Motte 1995, 25) – give important clues concerning the elementary differences between games and narratives. To take only one example: in multiplayer games the positions of players constantly affect each other. Such an arrangement would be very unusual but not impossible to execute in narrative fiction. The way I read The Idiot (Dostoyevsky 1955) would then change other people’s Idiots, or their readers’ possibilities when reading them, and vice versa. That wouldn’t make much sense, but in games such a practice has always already been in existence, and the current massive multiplayer games may very well be the most important change in audience structure since the invention of the choir, as Espen Aarseth (2001b) recently suggested.

Accordingly, we can distinguish the static user positions of literature, film, and average drama from the dynamic ones of games and certain installations and performances. We should also mention mobile positions in the wake of mobile gaming and games such as the recent Nokia Game Nokia Game http://www.nokiagame.com is interesting in how it makes use of the immediate media environment of the player, as the following excerpt from its rules makes clear: The player must complete various kinds of challenges and puzzles based on the given clues in order to proceed to the next stage of the Game. A time period for completing a task in question may be limited for some tasks (e.g. for couple of hours or the clue might be given at an exact time). This time limit will be informed to the player with the task or clue in question. The player may find clues via received short messages to his or her mobile phone or via other various kind of media, such as e-mail, Internet, TV, radio, magazines or newspapers. At most stages of the Game the player has only one chance to complete the task in question. At some stage of the Game some players will be excluded from the Game based on a wrong answer or action, or based on not being among the announced number of best players that performed the task in question. The game continues for a month for the winner, and a little less for the other players.that contacts the player through multiple channels (text messages, television, the web etc.) and demands action.

As we already stated, games have other than mere interpretative goals. These goals can be reached by traversing, negotiating, or otherwise overcoming a series of obstacles and gaps. When studying narratives as systems of gaps, Meir Sternberg (1978) made three heuristic distinctions: gaps are either permanent or temporary, focused or diffused, and either flaunted or suppressed. Sternberg’s gaps are not to be confused with the inevitable overdetermination and ambiguity of meaning in Wolfgang Iser’s phenomenology of reading (Iser 1978) and literary anthropology (Iser 1989). Instead, they are regulating the flow of information, and what readers can and cannot know. So even though we might guess it was the butler who did it, the appropriate information will be released at a specific point in the text.I think computer games can also be described that way, with the all-important exception that these gaps are not static and interpretative but ergodic (Aarseth 1997, 1) and dynamic: they need action to be encountered, closed, and dealt with. Aarseth’s four user functions (interpretative, explorative, configurative, and textonic (Aarseth 1997, 60-62)) are useful in specifying what kind of action is required from the player. In practical terms this means options such as finding paths, completing prefabricated relations, or adding new game elements for the other players to struggle with. The resulting typology of 32 possibilities could then be used to map out both qualitative and quantitative differences in the information given to the player in different stages and phases and levels of the game.

Focalization is one of the key elements of the narrative situation in classic narratology. In its most abstract sense, it’s a channel for narrative information and is ultimately based on the assumption of the uneven distribution of knowledge. Focalization is accompanied by the category of distance that regulates the amount (too much or too little) of information distributed through the channel, or two channels (audio and visual), as in film. This is exactly the level where I would like to draw a few parallels between this ludology-in-progress and narratology. One could argue that information is distributed and regulated very differently in games than in narratives, as in the former it’s also invested in formal rules. In some cases the knowledge of these rules is all that’s needed to succeed in the game (in Tetris for example). It’s important to understand that rules are not conventions. One can by all means change between conventions while reading a narrative, but one cannot change the rules of the game while playing. Or if one does, then it is another game. Conventions usually change over time but rules don’t (or not necessarily). This means games can be played by their original rules (if they are known) whereas writing is always already an orphan that can’t be reduced to its original context (and conventions long gone. The situation is more complex however, since it is common that the player has all the information needed but lacks skills.

In Genette’s (1980, 215) narratology there are three main categories – narrative level, person, and time of the narrating – that specify the narrator’s position or the coordinates of narrative acts. Parallels are pretty obvious, or at least easy to draw, as it would be only sensible to note the arrangement of levels in a game, and whether or not the player is represented by a character in a game as well as the player’s possibilities to time her action.

Aspects of Time in Computer Games

According to the famous statement of Christian Metz, “one of the functions of narrative is to invent one time scheme in terms of another time scheme” (Metz 1974, 18). Contrary to this, in games there’s only one necessary time scheme, the one already noted: the movement from the beginning to the winning or some other situation. In cases where another time scheme is invented, it is not as important as the first one.

Still, we could split this progression into two interplaying registers and argue that the dominant temporal relation in (computer) games is the one between user time (the actions of the player) and event time (the happenings of the game), whereas in narratives it is situated between story time (the time of the events told) and discourse time (the time of the telling). The key concept here is the dominant. Dominant, or to put the same idea in a politically correct way: “textual service” (see Chatman 1990, 10). Throughout this essay I invoke the heritage of Russian formalism on purpose, as it may well be that computer game studies need to go through formalistic phases similar to the ones that film and theater studies went through in the first half of the twentieth century, to gain their relative independence. As we all know, narratives such as Stuart Moulthrop’s Hegirascope (1995) and Reagan Library (1999) Hegirascope, a web fiction by Stuart Moulthrop, limits the reaction time of its readers to 30 seconds per node. Within that period of time the reader must decide which narrative thread to follow and choose a link; otherwise the program makes that decision for the player. In Reagan Library, also by Moulthrop, the content of the nodes change when they are revisited for the first three times (there’s more text available for the persistent reader). This affects or at least has the capacity to affect and alter the temporal relations between story time and discourse time (see Eskelinen 2001a). can utilize both user and event times for narrative purposes, and games like The Last Express The Last Express is an adventure game (a murder mystery) happening in the real-time of the game world. The player must find the culprit in time; that is, he may run out of time to solve the crime, as there’s a temporal limit to the duration of the exploration. In other words, the wasted time also counts, and the player has to manipulate “discourse time” and condense it to contain the relevant story events. can use story and discourse times for gaming purposes (see sidebar). Despite these possible hybrids, the underlying restriction remains the same: there’s no narrative without story and discourse times and no game without user and event times - everything else is optional.

In the course of a game the player encounters temporal phenomena or events with different durations, speeds, orders, and frequencies – and some of these must be manipulated or configured to move from the beginning to the winning or some other situation. Even though game time doesn’t have much in common with narrative time, this does not prevent us from observing similar temporal categories in both modes, as order, repetition or speed are not narrative or gamelike in themselves.

In formal narratology, Genette distinguishes between formal and thematic narratology (Genette 1988, 16). The latter is content-oriented, and interested in stories and themes (i.e., things like plot configuration and characters, in general, the “narrated”), and the former focuses on the specifics of narrative as a mode. To the detriment of thematic narratology (best exemplified by Marie-Laure Ryan’s approach to games (Ryan 2001)), there are no specific “narrative contents.” events are divided into actions and happenings based on their agency, and into kernels and satellites based on their relative importance. There’s also a difference between punctual acts and more durational actions (Chatman 1978, 32-56). Events can, of course, be more or less separate or connected, and we can borrow the three elementary possibilities of combination from Claude Bremond: embedding, enchaining, and joining (Bremond 1980). In our case, games can be differentiated from each other on the basis of which events can or cannot be manipulated, which parts and dimensions of events can be manipulated, and for how long and how deeply. An almost ready-made set of temporal relations can be derived from print and film narratologies – this act gives us six categories to study: order, speed, duration, frequency, simultaneity, and the time of action. It’s very probable there exist other noteworthy temporal relations, but I’ll begin with these. For possible “new” categories, see Eskelinen and Koskimaa (2001).

Let me note in passing that the manipulation or completion of multiple relations takes place in time – a kind of general economy of games – but here we are dealing only with the restricted economy of manipulating temporal relations. The importance of mutable temporalities varies from game-to-game, and there are games that are more dependent on other kinds of variables. For example, turn-based strategy games such as Civilization seem to favor causal relations over temporal ones to create event structures that have remarkable similarities to complex board games. We are talking here about quantitative differences: at one extreme there are multiple and highly interdependent chains of events with a complex tactical and strategic calculus, and at the other end looser chains of completed action episodes or stimulus-response cycles with no or minimal cumulative consequences. Taking into account the demands of gameplay (a well-balanced combination of tempo and cognitive tasks) it makes sense that the former games utilize intransient time and the latter transient time.

Order. In computer games this is the relation between user events and system events, or the actions of the player and their interaction with the event structure (happenings) of the game. In some cases there’s only one sequence of events, and the player has to act accordingly, in the sense of keeping up with it for as long as is humanly possible. Tetris (like many of its arcade relatives) best exemplifies this type of game. In other cases, commonly in exploration games such as Doom, order is a tripartite combination of events, negotiation, and progression (Aarseth 1997, 97-128); in these cases the player must find and test possible event sequences until the right one is found and the game can continue. So you either follow the order or spend your time finding it. In cases where the player cannot affect the order of events there is still the difference between variable and invariable sequences of events. In Tetris, where those objects just keep falling, the player can’t know in what exact order they’ll follow each other. This is also one of the simplest ways to limit or prevent anticipation.

Frequency. This factor concerns the repetitive capacities of the game. Basically, both events and actions (or to be precise, the player’s chances for taking action) may happen only once or an unlimited number of times. There may also be a limit to these recurrences, a kind of a middle ground between those two extremes. In some computer games, especially in role-playing games such as Ultima Online, at least some actions are irreversible and one cannot go back to a previous situation and undo the changes. In other kinds of games, the player can by all means keep banging his head against the wall until a break occurs somewhere. Sometimes it is even advisable. We could also describe recurring events in terms of their determination (the span of time in which an event or set of events recurs), specification (the rhythm of recurrence of the event or set of events), and extension (the duration of the recurring event or set of events) (Genette 1980, 127-140).

Speed. This aspect concerns pace. As we know, one of the great gifts computers brought to gaming is their superb ability to keep pace. To once again borrow a concept or two from Espen Aarseth, we can say that the main difference here is between transient and intransient games. In the former, the computer controls the pace and in the latter the player controls the pace. On the other hand, this concerns only the agent of speed. There are at least two other relevant dimensions of speed: its steadiness (for some reason the obvious alternative to this is almost always acceleration, not deceleration), and its importance as a goal in itself (as in some sports games).

Duration. This variable contains at least three aspects. First, Richard Schechner distinguishes between event time and set time (Schechner 1988, 6-7). In the former case, the game is over after all the events are properly traversed, and in the latter there’s a temporal limit to all this and the winner is the one who’s in the better position when the set time is up. Second, temporal limitations can either affect the whole game in its entirety, or only some parts of it that should be traversed within the set time. The Last Express is an intriguing combination of these possibilities. In games such as Doom the players should usually try to reduce the time span or duration allotted to any odd monster. If such an entity is allowed to live life to its full extent, the game is over. Third, the reverse options may be equally valid depending on the situation – to reduce the duration of an event by cheating or getting out of the situation, or to prolong the duration of an event (letting it happen) by avoiding any confrontation, as in Thief.

The time of action concerns the player’s possibilities for action. Basically, the player can act before, after, during (or in between) events. Not all games allow all these possibilities, and not all of these possibilities are equally important in any one game or in any one situation in a game. This is just one aspect of the type or the modality of action. It also corresponds in some degree to the difference between turn-based and real-time strategy games.

Simultaneity. The player may have to increase or decrease the number of simultaneous or parallel events, generate, or initiate such events. A typical example would be Command and Conquer and its multiple pieces. Events may have to be alternated, embedded, or linked to each other, or such prefabricated connections and arrangements may have to be reversed and dismantled.

We could easily go into greater detail here by introducing various subdivisions to the temporal categories discussed previously; or by taking more rigorously into account temporal requirements (in terms of speed, order, duration, etc.) set for the player’s possible and necessary actions, and mapping them onto the temporal dimensions of game events. So, after all, there’s still much nontrivial work to be done, as ludology, like the games it studies, is not about story and discourse at all but about actions and events, the relations of which are not completely fixed. Here’s a preliminary example of how to apply some of the key concepts utilized in this paper to Tetris, probably the most successful abstract computer game ever: Discourse time in narratology is somewhat similar to event time in ludology. The former could be seen as a series or a combination of individual event times, either fixed (or semifixed) as in print or hypertext narratives, or variable as in games. Still, as the time needed to complete a game usually varies considerably from player to player, I prefer event time to discourse time. One should also note that in computer games there’s always a conceptual difference between events as they exist in the game and as they are presented to or generated for the player (very much like textons and scriptons in cybertext theory; see Aarseth 1997, 62). In less abstract games there might be a fictive timeline into which the events are situated (in Civilization, it is the continuum between 3000 B.C. and 2020); it could be called content time (because we are not dealing with stories here).

 

Sidebar

Sidebar images

Responses

J. Yellowlees Douglas responds

Richard Schechner responds

Markku Eskelinen responds

Julian Raul Kucklich responds

References

Aarseth, Espen (1994). “Nonlinearity and Literary Theory.” In Hyper/text/theory, edited by George P. Landow. Baltimore: Johns Hopkins University Press.

—. (1997). Cybertext: Perspectives on Ergodic Literature. Baltimore: Johns Hopkins University Press.

—. (1998a). “Dataspillets Diskurs.” In Digitalkultur og Nettverkskommunikasjon, edited by Espen Aarseth. Bergen: Espen Aarseth.

—. (1998b). “Aporia and Epiphany in Doom and The Speaking Clock: Temporality in Ergodic Art.” In Cyberspace Textuality, edited by Marie-Laure Ryan. Bloomington: University of Indiana Press.

—. (2001a). “Allegories of Space.” In Cybertext Yearbook 2000, edited by Markku Eskelinen and Raine Koskimaa. Saarijärvi: Publications of the Research Centre for Contemporary Culture, University of Jyväskylä.

—. (2001b). “Computer Game Studies, Year One.” Game Studies 1, no.1 (July 2001). http://www.gamestudies.org/0101/editorial.html.

Avedon, Elliott M., and Brian Sutton-Smith (1971). The Study of Games. New York: Wiley.

Bénabou, Marcel (1998). “Rule and Constraint.” In Oulipo: A Primer of Potential Literature, edited by Warren J. Motte, Jr. Normal, IL: Dalkey Archive Press.

Bolter, Jay David, and Richard Grusin (1999). Remediation: Understanding New Media. Cambridge, MA: MIT Press.

Bordwell, David (1984). Narration in the Fiction Film. Madison: University of Wisconsin Press.

Bremond, Claude (1980). “The Logic of Narrative Possibilities.” New Literary History 11 (1980): 398-411.

Caillois, Roger (translated by Meyer Barash) (1979). Man, Play, Games. New York: Schocken Books.

Chatman, Seymour (1978). Story and Discourse. Ithaca: Cornell University Press.

—. (1990). Coming to Terms. Ithaca: Cornell University Press.

Crawford, Chris (1982). The Art of Computer Game Design. http://www.vancouver.wsu.edu/fac/peabody/game-book/Coverpage.html.

Dostoyevsky, Fyodor (translated by David Magarshack) (1955). The Idiot. Penguin: London.

Jacques Ehrmann, “Homo Ludens Revisited.” Yale French Studies 41, 38-57 (1969).

Eskelinen, Markku (2001a). “Introduction to Cybertext Narratology.” In Cybertext Yearbook 2000, edited by Markku Eskelinen and Raine Koskimaa. Saarijärvi: Publications of the Research Centre for Contemporary Culture, University of Jyväskylä.

—. (2001b). “The Gaming Situation.” Game Studies 1, no.1 (July 2001). http://www.gamestudies.org/0101/eskelinen/.

—., and Raine Koskimaa (2001). “Discourse Timer: Towards Temporally Dynamic Texts.” Dichtung Digital 17 (2001). http://www.dichtung-digital.de/2001/05/29-Esk-Kosk.

Frasca, Gonzalo (1998). “Ludology Meets Narratology.” http://www.ludology.org/articles/ludology.htm.

—. (2001). “Videogames of the Oppressed.” M.A. Thesis: School of Literature, Communication and Culture, Georgia Institute of Technology, Atlanta, Georgia. http://www.ludology.org/articles/thesis/.

Genette, Gerard (1980). Narrative Discourse. Ithaca: Cornell University Press.

—. (1988). Narrative Discourse Revisited. Ithaca: Cornell University Press.

Huizinga, Johan (1950). Homo Ludens: A Study of the Play Element in Culture. New York: Roy Publishers.

Iser, Wolfgang (1978). The Act of Reading: A Theory of Aesthetic Response. Baltimore: Johns Hopkins University Press.

—. (1989). Prospecting: From Reader Response to Literary Anthropology. Baltimore: Johns Hopkins University Press.

Juul, Jesper (1999). “A Clash Between Game and Narrative.” M.A. Thesis, University of Copenhagen, Denmark. http://www.jesperjuul.dk/thesis.

—. (2000). “What Computer Games Can and Can’t Do.” Paper presented at the Digital Arts and Culture, Bergen, August 2-4, 2000. http://www.jesperjuul.dk/text/WCGCACD.html.

Keene, Donald (1995). “Japanese Aesthetics.” In Japanese Aesthetics and Culture, edited by Nancy G. Hume. Albany: State University of New York.

McHale, Brian (1987). Postmodernist Fiction. New York: Methuen.

—. (1992). Constructing Postmodernism. London: Routledge.

Metz, Christian (1974). Film Language: A Semiotics of Cinema. New York: Oxford University Press.

Motte, Warren (1995). Playtexts. Lincoln: University of Nebraska Press.

Moulthrop, Stuart (1995). Hegirascope. http://iat.ubalt.edu/moulthrop/hypertexts/hgs/.

—. (1990). Reagan Library. http://iat.ubalt.edu/moulthrop/hypertexts/rl/.

Parlett, David (1999). The Oxford History of Board Games. Oxford: Oxford University Press.

Prince, Gerald (1982). Narratology. Berlin: Walter de Gruyter.

—. (1987). The Dictionary of Narratology. Lincoln: University of Nebraska Press.

Ryan, Marie-Laure (2001). “Beyond Myth and Metaphor: The Case of Narrative in Digital Media.” Game Studies 1, no.1 (July 2001). http://www.gamestudies.org/0101/ryan/.

Schechner, Richard (1988). Performance Theory. London: Routledge.

Sternberg, Meir (1978). Expositional Modes and Temporal Ordering in Fiction. Baltimore: Johns Hopkins University Press.

Sutton-Smith, Brian (1997). The Ambiguity of Play. Cambridge: Harvard University Press.

Vatsyayan, Kapila (1996). Bharata: The Natyasastra. Delhi: Sahitya Akademi.