My Master's Thesis:
Theorising Video Game Narrative

Home Page

About Me
My Wedding Page

My Career
My Games


My Master's Thesis


Polish Games Industry
Industry Overview
Games Industry List


Games Modding



[previous section]                                    [table of contents]                                     [next section]

1. Introduction


Ever since the study of video games (1) has expanded beyond the basic debate of video game violence, and turned to analysing the content of the games themselves, one of the most frequently debated questions has been that of game narrative.

Noted game scholar Espen Aarseth once asked whether games are or can be stories (2). The answer to such a question is, naturally, no. But books and films aren’t stories either – they merely contain them, so Aarseth’s question couldn’t produce an affirmative answer for any medium. A potentially more effective question would be whether games can contain stories. However, given the number of examples out there of games that purport to contain stories, a negative answer would be almost certainly untenable, and thus answering this question doesn’t really help us to better understand any issues of the interaction between gameplay and narrative.

The real question, then, is two-fold:

a) Do game narratives have any special traits of their own, or is it merely a case of traditional, linear narratives being superimposed onto a non-linear game environment?

b) Should games even have narrative, given the fact that they are games?

Of these two questions, the latter seems to have all but exhausted its usefulness. It is becoming clear that computer games are not a single form – indeed, it is difficult to decide whether they even constitute a single medium, even if it is more convenient to keep them grouped together as a single academic field of study than to separate them into different fields. Thus, narrative is simply appropriate for some games, and inappropriate for others – not all criteria that can be used to study games will apply in all cases.

It is that first question, then, that needs to be examined in detail. Indeed, the need to examine the characteristics of video game narrative is especially urgent precisely because we need to understand what constitutes video game narrative in order to be able to determine what the differences are between the narrative forms of video games and the non-narrative forms. A few words, however, must be first said about the idea of narrative. Film theorists David Bordwell and Kristin Thompson state that narrative is “a chain of events in cause-effect relationship occurring in time and space” (3). Thus, a narrative is a story consisting of a number of events, and narration is the act of relaying this story to an audience. Not all the events and causal links in a story need to be explained by the narrator, leading to the distinction between the plot (everything that is explicitly presented to the audience) and the story (the entire narrative) (4), but in order to form an understandable narrative, the plot needs to include a sufficient amount of information for the audience to perceive the existence of a chain of events and their causal links (5).

The basic distinction between narrative and non-narrative forms of media is thus generally a simple matter – the narrator in narrative forms seeks to relay enough of the plot to the audience to allow them to construct the story, whereas in non-narrative media, even if there is a narrator, she does not attempt to do this. This is not always a clear-cut divide, however – although, as Markku Eskelinen notes, if another person throws a ball in your direction, you do not see this as a narrative situation (6), other similar situations might be closer to narrative – a board game like Monopoly can be considered as a narrative of an individual person’s rise to wealth within the capitalist system.

In video games particularly, the separation of narrative and non-narrative forms can be problematic – Tetris (1985) and Space Invaders (1977) are equally abstract situations, but whereas the former is clearly non-narrative because it makes no attempt to tell a story, the latter is a more complicated case. Its title suggests that an invasion has taken place, and a causal link between the invasion and the player’s fight against alien creatures is implied, thus generating a rather rudimentary narrative. In general, however, even if a game cannot be identified as completely non-narrative or completely narrative, either one of these aspects will be emphasised by the game as being more important than the other. One way in which this can be seen in games is the degree of abstraction of the player’s reward for playing the game. Games where the non-narrative element is the dominant one will reward the player with points, as is the case in both Tetris and Space Invaders. On the other hand, games that emphasize narrative may reward the player with nothing more than the conclusion of the story, as is the case in Wing Commander (1990). Another distinction that could be made between the two is that non-narrative games will emphasize the experience of playing the game, with the player’s enjoyment being derived from the gameplay (7). On the other hand, playing the narrative game, the game’s story is emphasized – which is not to say, of course, that the player of a narrative game plays it solely for the story; the gameplay presumably needs to remain enjoyable, because otherwise negotiating through the game’s story would be a tedious process. In this thesis, at any rate, we will be mainly examining those games where a causally-linked chain of events – a narrative – can be perceived.

This thesis will begin by reviewing some of the most notable arguments for and against video game narrative. Although, as I’ve explained above, question of whether games should or shouldn’t have narrative is not an especially constructive way of approaching the subject, it still needs to be examined here to demonstrate why this is the case. This will be done by first examining the most extreme arguments on both sides. We will begin by discussing Markku Eskelinen’s claims that video game narrative is nothing more than a marketing tool undeserving of study, and that it is gameplay that should in fact be studied. Another author that will be discussed is Janet Murray, who claims that the future of interactive games, if they are to be considered as a mature art form, is inextricably bound with narrative. These two authors represent the outer edges of the debate, rejecting the value of either narrative or non-narrative games. We then will proceed to explore more moderate points of view, and in particular the work of Jesper Juul, who has suggested that narrative and gameplay are two separate things in conflict with each other, but that hybrids combining narrative and gameplay can exist, and can be interesting. Juul’s arguments are useful in defining some of the basic limitations that are imposed on narrative in games. In terms of narrative techniques, Juul argues, games are limited in the range of temporal possibilities available. He argues that within the video game, story, narrative and reading (playing) times are conflated to the point where they appear to be a single temporality. This argument does not seem entirely accurate – the single temporality is an illusion, since a game’s story may take place in the future or the past, regardless of when the player is playing it; however, it is true that the kinds of temporal manipulation that are commonplace in non-interactive narrative are difficult to implement in video games. This, together with the other problems that interactivity imposes on narrative, leads to the separation of the narrative video game into gameplay (interactive, action-centred) and cutscenes (non-interactive, narrative-centred) sequences. Juul’s work helps to illustrate these problems, exploring the relationship between gameplay and cutscene sequences in video games.

Juul’s work will serve as a transition of sorts, shifting our focus away from the question of whether video games should have narrative, onto the actual nature of narrative in games. Following this transition, then, I will devote the rest of this thesis to the construction of a more comprehensive analysis of video game narrative – or indeed, the dramatic experience of video games, since some academics have argued that many games are more like theatrical drama than like narrative books (8). Theatre, being more akin to an event unfolding before the audience’s eyes than to an event being narrated to an audience, arguably does not constitute narrative as such (9). This will be one of the first points to be examined in the second part of this thesis – although performance studies is not synonymous with narrative studies, some space needs to be devoted to examining the dramatic performance aspect of video games in order to better understand the player’s relationship with the game. The comparison of video games to theatrical drama seems particularly noteworthy in regards to Jesper Juul’s arguments that games, compared to traditional narrative media, offer a highly truncated range of temporal possibilities, with most games unfolding entirely within the present tense (10). Thus, it may well be that the experience of playing a computer game has as much in common with watching a film as it does with an actor’s experience of performing a role in a play.

Apart from offering the benefit of looking at game storytelling from an alternative point of view, I believe that examining the performance aspect of video games will also help shed some light on issues of subjectivity in games. This is an aspect in which video games differ vastly from other forms of narrative media – while films where all or even most of the scenes are presented in first-person view, as though the audience and the lead character were one and the same, are an extremely rare occurrence, there is an entire category (or genre, or form – such divisions remain questionable in video games) of computer games where the player spends most of the game looking through the eyes of her character.

The next aspect of game narrative that I examine is the basic narrative structure in games. This will focus on the narrative structures present in all stories. Robert McKee identifies three basic narrative structures in linear texts – the classical structure, the minimalist structure, and the anti-structure (11). The first of these is essentially the so-called hero’s journey structure. This type of structure, first theorised about by Vladimir Propp and Joseph Campbell is explored in the works of Christopher Vogler, and Robert McKee, among others, and dominates not merely Hollywood cinema, but indeed most storytelling traditions (12). The remaining two structures are variations, and in some ways reactions to that first, most standard structure. All three structures will be described here, and then the ways in which they are affected by the interactive nature of the video game will be examined.

Which of these three linear narrative structures best reflects the experience of playing a computer game? It seems as though most narrative games utilise the classical structure to some degree, This is usually possible because most such games restrict the player’s possibilities of influencing the narrative, in such a manner that the story’s overall structure can remain fairly linear. However, there are some narrative games that are more experimental with their narrative, offering the player the ability to select the outcome of the story, and others that allow the selection not of the outcome, but of the story itself; in such cases, the experience of playing the game becomes less like a classically structured narrative and more like a minimalist structure or anti-structure.

Meanwhile, even games that let the player choose the outcome of the story are not exactly what the utopian ideal of ‘interactive narrative’ would seem to have in mind – as Steven Poole writes, such narratives are far from true interactivity, because they offer only a limited number of opportunities for choices within the narrative, and even then, only a limited number of options in each such situation. Indeed, Poole notes, such limitations are necessary for a storyline to have strong emotional resonance (13). This is confirmed by a recent quantitative and qualitative study of video games, which found that the stronger the storyline gets, the less the player can influence it (14). The biggest scope of influence is present in games that quite simply have no narrative – however, studying such games from the perspective of narrative studies would offer little insight.

Within this thesis, I examine what can be described as the three basic models for narrative structure in video games. A fourth model, for games with unstructured narrative that is generated during gameplay, can also be identified, but does not entirely fit in within this work. Those first three models include the ‘string of pearls’ approach, where the player goes through a series of pre-set events. In between these major events, the player has more freedom, but ultimately can only go on to the next event (15). The second model is the ‘branching storyline’ approach, where the player may affect the story by making choices at strategic moments during the story. Finally, there is what may be called the ‘amusement park’ approach. This approach is similar to the beaded necklace approach, but the exploration is emphasized over the central storyline, and the narrative is fragmented into sub-plots. The player has a certain degree of freedom to move from one sub-plot to another, and advances made in one sub-plot may affect other sub-plots. There is likely to be an overarching plot that needs to be resolved in order to finish the game. The stronger the overarching plot, the more the game will resemble the beaded necklace approach.

The fourth model, for games with unstructured narrative, will also be discussed. This model, which we will here call the ‘building blocks’ model, is visible in games such as Civilization (1991), and stands on the border between narrative and non-narrative. Civilization is a game where the player guides a civilisation throughout history, striving to expand its influence in the world, and its importance compared to other civilisations. In other games of a similar type, the player may be guiding a city or a nation. These games do not so much tell a story as they create it in collaboration with the player. Nor is it a typical story – Ted Friedman argues that the primary narrative agent in Civilization and other games like it is actually geography. Thus, Civilization tells the story of a map changing over time (16). Although one might disagree with this conclusion, arguing that Civilization is rather the story of the player’s civilisation over time, the point remains that it is not the story of a particular character, and it has no pre-designed structure – unlike in the other three models, the designers here can only create the initial situation. What happens next is entirely in the hands of the player.

In this thesis I focus on pre-designed narrative, and thus most of my attention will be on the first three models. These three models are not entirely independent of each other – nor, indeed, of the building blocks approach – so-called massively multiplayer online role-playing games (MMORPGs) would fall in the grey area between the building blocks and the amusement park models.

In the course of the discussion on game narrative structures, each of the three pre-designed narrative models is discussed in detail, with various games being dissected to provide examples of these narrative models. I will compare them to each other, and to the more traditional structure found in linear narratives. I will also discuss which types of games are more likely to use which approach. As for the building blocks model, a detailed discussion of this model would be outside of the scope of this thesis. However, the model will nonetheless be briefly discussed in order to explore the idea of narrative as the player’s post-gameplay reconstruction of the story generated by the game during the gameplay.

Before we begin exploring all these subjects in depth, a terminological side-note needs to be made here. Over the past few years, the concept of interactivity has been increasingly questioned. Aarseth has in fact rejected its use altogether, utilising instead ‘ergodicity’ – a term derived from the Greek words ‘ergon’ and ‘hodos’, meaning ‘work’ and ‘path’ respectively. Thus, ergodic texts (a category in which computer games may be placed) are texts where ‘nontrivial effort is required to allow the reader to traverse the text’ (17). By contrast, Aarseth argues, the term ‘interactivity’ carries with it ideological implications of an equal partnership between the user and the interactive text, while simultaneously remaining rather vaguely defined (18). In this thesis, however, I will generally continue to use the term ‘interactive’ rather than ‘ergodic’, except where the latter seems to be more appropriate.

The reason for this decision on my part is that while interactivity may indeed be interpreted as overplaying the games’ interaction with the player, it seems as though the notion of ergodicity may be doing the exact opposite – reducing the player-game interaction to a one-sided effort on the part of the player to traverse the text. Thus, ergodicity is a highly limiting notion as far as the study of computer games go. For example, when this notion is applied to a non-narrative game, like a computerised chess game (where one player is a computer), the idea of a reader engaged in a non-trivial effort to traverse the text entirely misses the point. The notion of chess as a one-sided effort rather than an interaction between two players simply doesn’t work. Ergodicity makes more sense when applied to narrative games where there is a limited number of possible outcomes and the player is indeed striving to attain a particular outcome. Even for narrative games, however, it does not apply to every case – especially in games where the narrative fits the amusement park model, the player’s experience may be more focussed on interacting with the game world than reaching the end of the text.

Having listed the topics that will appear within this thesis, we can now begin an in-depth exploration of the first of these topics, the validity of the question of whether computer games should actually have narrative.

Footnotes   [back to top]

1. Throughout this thesis, the terms’ video game’, ‘computer game’, and ‘game’, will be used interchangeably. Computer games and video games are not particularly different in regards to narrative, and using the terms interchangeably helps make this point. Needless to say, whichever term is used, I am only referring to digital games and not their analogue counter-parts. [back to text]
2. Espen J. Aarseth, ‘From games to game studies: how to build your own department’, paper presented at the From Space War! to the Ivory Tower conference, May 26, 2003. [back to text]
3. David Bordwell and Kristin Thompson, Film art: an introduction, New York, McGraw-Hill, 2001, p. 60. [back to text]
4. Ibid., pp. 61-62. [back to text]
5. Ibid., pp. 60-61. [back to text]
6. Markku Eskelinen, ‘The gaming situation’, Game studies,, vol. 1 no. 1, 2001, [back to text]
7. See, for example, Steven Poole, Trigger happy: videogames and the entertainment revolution, New York, Arcade Publishing, 2000, pp. 168-171. [back to text]
8. Brenda Laurel, Computers as theatre, Reading, Addison-Wesley Publishing Company, 1998, pp. 93-97. [back to text]
9. However, it can also be argued that theatre is a type of narrative, with the dramatic performance being the interface. See for example Espen J. Aarseth, Cybertext: perspectives on ergodic literature, Baltimore, The Johns Hopkins University Press, 1997, pp. 136-138. [back to text]
10. Jesper Juul, Time to play – an examination of game temporality,, 2002. [back to text]
11. Robert McKee, Story, New York, ReganBooks, 1997, pp. 44-47. [back to text]
12. Christopher Vogler, The writer’s journey, London, Pan Books, 1999, pp. 1-7. [back to text]
13. Steven Poole, op. cit., pp. 96-99. [back to text]
14. Jeff Brand, Scott Knight, and Jakub Majewski, The diverse worlds project, unpublished. [back to text]
15. Jane Jensen, in Mark Saltzman (ed.), Game design: secrets of the sages, Indianapolis, Macmillan Publishing, 2000, pp. 96-99. [back to text]
16. Ted Friedman, ‘Civilization and its discontents’ in Greg M. Smith (ed.), On a silver platter: CD-ROMs and the promise of a new technology, New York, New York University Press, 1999, pp. 138-143. [back to text]
17. Espen J. Aarseth, Cybertext: perspectives on ergodic literature, op. cit., pp. 8-9. [back to text]
18. Ibid., pp. 47-51. [back to text]

[previous section]                                    [table of contents]                                     [next section]

Copyright 2003 Jakub Majewski