Ab(Sense) of An Ending: Telos and Time in Digital Game Narratives

Souvik Mukherjee


The potential of digital games as a storytelling medium has started to be recognized over the last two decades. However, their status as a narrative medium is still the subject of much controversy. Unlike the apparently linear plots of earlier narrative media, the story-space in digital games is seemingly endless or multi-telic. For some commentators, this poses major problems in conceiving of them as narratives. A number of studies on ‘reading’ (or wreading) computer games refer to their multiple endings as a unique feature but none attempt in-depth analysis. Yet, as it is the peculiar nature of the endings that compounds the problem with ‘reading’ digital game-texts as a narrative medium, a discussion of this is now long overdue. This paper will examine the telic possibilities of game-texts, and explore the link between these and other forms of texts; in the process, it will show how game-endings provide a significant point of departure in the understanding of textual possibilities in narrative media.

Endings have always been a major element of interest in the world of stories. Scheherazade famously preserves her life by postponing the end of her story and weaving within it a mesh of further stories, each leading to another. Shakespeare’s endings have baffled generations of scholars. Sometimes, later Shakespearean productions have even changed the endings: for example, Nahum Tate’s nineteenth-century King Lear has a happy ending where Cordelia marries Edgar. More recently, especially in works like Italo Calvino’s The Castle of Crossed Destinies or Alain Robbe-Grillet’s novels, the literary narrative contains many endings and repetitions, because of which conceptions of temporality and telos are altered and confused. The problem of endings in these works points clearly to the fact that narrative endings have always contained the potential for multiplicity, whether on the level of text, continuation of story or interpretation.

Literary criticism has become increasingly responsive to these issues and there have been various attempts by eminent scholars to address them. In his classic study, The Sense of an Ending, Frank Kermode extensively analyses the works of Robbe-Grillet, concentrating on their telic element. Gérard Genette also refers to the same texts in terms of the events that repeat themselves. As Kermode comments:

Les Gommes is writing with an eraser. The story ends where it began, within the immediate perceptual field of the narrator. It is always not doing things which we reasonably assume novels ought to do: connect, diversify, explain, make concords, facilitate extrapolations. Certainly there is no temporality, no successiveness.1

He states that in Robbe-Grillet there is ‘an attempt at a more or less Copernican change in the relation between the paradigm and text’.2 However, he cannot help observing a similar principle in operation in earlier novels like Camus’ The Plague and Dostoevsky’s The Idiot. He observes that Camus’ novel is ‘susceptible to multiple readings … it even contains the opening of a rival novel’.3 In an even earlier example, he mentions how in both St John and St Paul there is the tendency to conceive of the End as happening every moment. When he maintains that the same immanent endings are also characteristic of narratives, Kermode comes very close to describing a narrative medium that illustrates this to a far greater degree than Robbe-Grillet’s novels: the digital game.

The multiplicity of endings in game-texts is not a unique media-specific feature and is already present in earlier narrative media. Though narratives in game-texts may employ different technologies, they are essentially not new. A major claim that the advocates of this so-called ‘newness’ make is that of replayability. For some game studies commentators like Jesper Juul or Craig Lindley, the repetition characteristic of computer games is largely incompatible with narrative. Juul states that ‘Literary qualities … actually [make] computer games less repeatable’4 while Lindley claims that the repetitive structure of computer games ‘[undermines] any strong sense of narrative development’.5 However, these claims are based on older conceptions of narrative progression such as the linear structure of Aristotelian drama; they cannot be justified under more current conceptions of narrativity, as examples throughout this analysis will illustrate.

Repetitive structures already exist in older narrative media without in any way undermining narrative development. As Kermode describes it, ‘in Robbe-Grillet’s novel the same character is murdered four times over’.6 Genette also reminds us that ‘certain modern texts are dependent on their capacity for repetition’. Like Kermode, he points to Robbe-Grillet and others as obvious examples of repetition:

We may remember , for instance, a recurrent episode like the death of the centipede in La Jalousie. On the other hand, the same event can be told several times not only with stylistic variations, as is generally the case in Robbe-Grillet, but also with variations in ‘point-of-view’, as in Rashomon or The Sound and the Fury. The epistolary novel of the eighteenth century was already familiar with contrasts of this type….7

Genette’s and Kermode’s comments clearly indicate that repetition and multiplicity have always coexisted within the very notion of narrativity. The narrative in the game-text, characterized by variance occurring within a process of repetition, is therefore not a new phenomenon. What is different, however, is the manner in which the variance and repetition can occur and the degree to which immanence can be experienced. Compared to the earlier narrative media cited by Kermode and Genette, digital games have a more complex telic structure, characterised by multiplicity and repetition. To plot this, one would have to include all the events, running back and forth and laterally along the timeline - an almost impossible task.

Some of these games are quite conscious of this aspect. Prince of Persia: Sands of Time (henceforth called Sands of Time) allows the player to rewind events within the context of the game. Should the player fail in his attempt, the Prince’s voice tells us, ‘No, this is not how it happened’ making the entire gameplay instance seem like one bad flashback among many. In an added nuance to the game, the Prince’s response is subtly different each time and the story develops by subverting, reversing or restarting the progression of events. Therefore, it can be said that the endings as well as the beginnings of the game are immanent and that they often overlap when the narrative is considered along various planes. The theme of Sands of Time is time and as the young prince tells us in the ‘beginning’ of the game, he thought that Time was like a river but now he has found out that it is like the sea.8 Time does not have a unidirectional progression according to the game. With the Dagger of Time that the Prince finds in the treasury of an Indian Maharajah, he can travel back in time and reverse his actions. His first discovery of the powers of the dagger is quite illustrative:

Unaware of the stone gargoyle plunging toward him, he notices a switch on the dagger’s hilt. He presses it. SAND spills from the dagger onto the floor.
At the last second, the Prince looks up to see the gargoyle about to crush him! His eyes widen with the terror of certain death. But just then —
The gargoyle springs back up, reversing its trajectory, and lands in its original position.
blinks, baffled as to what just happened.
As he is staring up at the gargoyle, it teeters, just as it did before, and starts to fall a second time.
This time, forewarned, the Prince jumps back out of the way. The gargoyle crashes harmlessly next to him.9

Every step inside the Maharajah’s crumbling palace ruins is fraught with danger - with spikes emerging from the floor and swinging blades - so the endings are not only immanent, they are constantly imminent. Once the player finishes the game (after many ends, rewinds and repetitions, presumably), the game shows the Prince at the bedside of the sleeping Princess Farah, who was shown as dead in the last section of the gameplay - the end of the story is another beginning. As Barry Atkins comments in his recent essay, Sands of Time is perfectly self-aware:

In drawing attention to issues of temporality in games, however, it [Sands of Time] highlighted its own structure as a videogame even as it might appear to have attempted to conceal the artificiality of this key aspect of the practice of videogame play through providing an internal justification for temporal manipulation through the Dagger of Time.10

Besides the rewind function provided through the Dagger of Time, the player is also allowed random glimpses of possible futures through the proleptic ‘vision’ mode that is present within the game. Interestingly, it is accessible from the same place (a translucent golden hourglass-like figure) as the save-game function - the vision mode is essentially a flash-forward showing one potential future while the save game function is a node from which innumerable possible futures can result or which allows a return to various saved instances of pasts. Gameplay, thus, exists in the realm of the virtual. The ‘sands of time’ can also be used to control the speed with which events occur within games: the ending of the game-text is therefore delayed or hastened, as the case may be. Of course, the player’s interaction (and skill) is also key to this deferment or hastening. Further, the selection of difficulty levels makes it more or less difficult (and often, therefore, taking more or less time) to complete all the levels of a game. The increasing number of obstacles in higher difficulty settings can also influence the narrative. Unless the player kills the monstrous antagonists and destroys them by obtaining their ‘sands of time’ using the Dagger of Time (which, in a beautifully animated sequence, sucks them in), they respawn and attack yet again.

A further complication arises with the ‘sequels’ to Sands of Time.11 Prince of Persia: Warrior Within and Two Thrones both link their plots to the Sands of Time story: for example, in Two Thrones, Princess Farah reappears but she does not remember the Prince. The action in all three games is supposed to be happening in different replays of the same story, involving the same characters, but not only does the time-frame vary, there is also a considerable shift in spatial terms: from India to a mysterious Island of Time and then to Babylon. It is difficult to conceive of a transcription of the ‘plot’ of Sands of Time because the narrative contained within the game-system is a multilayered temporal mesh. Story systems created in digital games are indeed quite different; nevertheless, they are still stories.

In critical circles, games are considered as very different and even trivial, when compared to ‘serious’ cultural products, because of their replayability and multiplicity. Gonzalo Frasca states this view quite clearly:

Whatever you do in a game is trivial, because you can always play again and do exactly the opposite…. [The player] is free to explore any ’what if’ scenario without taking any real chance. The problem is that usually ‘serious’ cultural products are essentially based in the impossibility of doing such a thing in real life.12

Frasca observes that from the perspective of real life, the reversibility of events is viewed as something that ‘trivializes the “sacred” value of life’.13 However, his concomitant definition of ‘serious’ cultural products is quite controversial. There are many instances in so-called ‘serious’ literature and films that constantly point to the possibility of the multiple within texts, as seen earlier. Films like Blind Chance or novels like The French Lieutenant’s Woman narrate the possibility of many ‘what if’ scenarios being actualized after rewinding time and restarting the action. To conclude that texts which do so are trivial is therefore not tenable. Nevertheless, the issue of difference from older media remains a moot question and endings and temporality remain major issues. Atkins, commenting on Sands of Time, states that ‘it brings to our attention … the degree to which videogame play offers a very different temporal experience than our other media’.14 This is a key point because, although it states that gameplay offers a ‘very different temporal experience’, it also qualifies the statement by saying that the difference is in degree.

For a comparison of the telic possibilities of games and older media based on their temporal structure, the nature of ludic time needs to be examined. Juul’s essay, recognized as a key contribution on the subject, is an important entry-point. Juul maintains that games apply a different set of temporal parameters. According to him, the moment of gameplay, ‘has a basic sense of happening now, when you play. Pressing a key influences the game world, which then logically (and intuitively) has to be happening in the same now’.15 For Juul, narrative conveys a basic sense in which the events do not happen now and the plot itself imposes a chronology for the events to happen. The game, however, happens solely in the now.16

There is a problem with this position when we apply it to a game like Sands of Time. This game is set somewhere in the ancient past and yet, because the player acts out the story, there is a sense of the events happening now. Further, the game consciously confuses the difference between the now and the then, within its temporal mesh. It is no coincidence that Jordan Mechner, creator of the first Prince of Persia game and member of the designer team of Sands of Time, acknowledges the influence of the ‘nested stories’ in The Thousand and One Nights in making the games: to return to Scheherazade again, her stories while happening in the now consists of other stories within the main story which relate to each other. These may begin and end as separate stories but as soon as Scheherazade ends a story, another one starts, thereby postponing the end of the main story. Juul observes a difference between Scheherazade’s situation and computer game play. For him, ‘the continuing delaying of Scherazade’s (sic) execution in Thousand and one nights is a good example of this [the reader’s desire to know the end]. In the computer game, on the other hand, the ending is often well known, but it is one you try to actualise by your playing’.17 There is, however, a problem with this description. Sands of Time makes it amply clear that the end of a game is impossible to predict. Ideally, the player should reach the point where the Prince meets Princess Farah sleeping in her bedchamber but that is not the end of the story: the Prince runs away and disappears into the jungle, leaving Farah bewildered about how he seemed to know all that he said. The game can therefore be replayed by treating this as one actualization of the possible combinations of events but one that keeps alive the player’s (or reader’s) desire to reach the end by postponing the conclusion, much like Scheherazade’s stories. A temporal map of either Scheherazade’s tales or the Prince’s adventures is not plottable owing to the complexity of the multi-level links. The conception of such a structure is not entirely new to narratives: a very famous literary example is Jorge Luis Borges’ short story, ‘The Garden of the Forking Paths’.

The illustration of the problem of the non-linearity of time in digital games, therefore, does not come as something uniquely different. Book Ten of Augustine’s Confessions discusses time as existing as an eternal present and states that all actions, whether in the future or in the past, actually occur in the ‘right now’: Juul’s conception is therefore not really unique or limited to ludic time. Using later accounts from Christian theology, Kermode identifies three main orders of time: chronos or earthly time, occurring as successive events, kairos or God’s time, consisting of moments beyond conceptions of reality and temporal sequence and aion, described as the ‘time of a world of becoming’. For Kermode, the novel exists in the time-order of the aion and incorporates a movement to and from the regions of chronos and kairos. Applying Juul’s terminology, this oscillation could be seen as occurring between the chronological order of event time and the immanent ‘right now’ of play time – quite similar to the situation in computer games.

It is evident, therefore, that temporality in digital games already has a string of antecedents in older narrative media and is part of a much larger discussion. However, like game-studies, literary criticism is also uneasy about some aspects of temporality. Though Kermode highlights the immanent endings and the temporal variation in the nouveau roman, he regards the ‘real novel’18 as one with a beginning, middle and end and is uncomfortable with such novelists as William Burroughs because their prose is in ‘random order’.19 Such a ‘justification of the ideas of order’20 is in contradiction to implications of immanence and marks the limitations of canonical criticism in analysing narrative endings. This is where a study of game-texts is called for.

In a computer game, the actions do not happen once, but both as one and many at the same time. Failure or death, in Sands of Time, has the Prince exclaiming that it is not how things happened, but the player knows that it is. In one sense the player’s action is valid as a single gameplay session and in another it is a unit within a multiplicity. Amongst the different strands of the narrative mesh, some are not even available: they exist but are not available until played into existence. Finally, as any gamer knows, these narratives keep overlapping and there is both difference and repetition amongst the countless potential or actualised trajectories. As far as the potential narratives are concerned, it is next to impossible to determine their number or nature, even for the designers of the games themselves. Using cheats, mods and patches, gamers can easily exploit the technology and affect the creation of the narrative. The player can also develop unprecedented playing strategies that can change the game narrative even without using external elements like mods or patches. As Espen Aarseth comments about the multiplayer demo of Return to Castle Wolfenstein:

Someone discovered that by exploiting the fact that players were invulnerable for the first seconds after they were revived by a medic, one could ‘fly’ over the wall if one was revived next to a live grenade about to explode. Thus, by committing suicide, one could win the game in a novel way.21

These and similar elements have caused a rethinking of game design concepts, resulting in more player-centred design options that view the narrative space of games as ‘the space of possibility’, which Katie Salen and Eric Zimmerman describe as:

the space of all possible actions that might take place in a game, the space of all possible meanings which can emerge from a game design … as a game designer you can never directly craft the possible space of your game. You can only indirectly construct the space of possibility, through the rules you design.22

Is it at all possible to analyse the ‘space of possibility’ which can neither be directly crafted, nor constructed? While the tools employed by literary criticism prove inadequate, Gilles Deleuze’s theories of multiplicity are extremely well-suited to evaluating such multi-telic and intangible systems of possible narratives. Treated as a Deleuzian multiplicity, the Prince’s narrative in Sands of Time and narratives in other game-texts become more accessible to analysis.

Multiplicity is a key concept in Deleuzian thought and, as Manuel DeLanda states, ‘is one that stands out for longevity’.23 DeLanda describes Deleuzian multiplicity in terms of the ‘manifold’, a mathematical concept developed to cover n-dimensional geometry. According to DeLanda,

A Deleuzian multiplicity takes as its first defining feature these two traits of a manifold: its variable number of dimensions and more importantly, the absence of a supplementary (higher) dimension imposing an extrinsic coordination, and hence, an extrinsically defined unity… never has a supplementary dimension to that which transpires upon it. This alone makes it natural and immanent.24

Though neither Deleuze nor DeLanda discuss computer games as such, the idea of the variable number of dimensions not subordinated to an extrinsically defined unity aptly describes the variable pathways that game narratives usually take. According to DeLanda, ‘the dimensions of a manifold are used to represent properties of a particular process or system, while the manifold itself becomes ,i>the space of possible states which the physical system can have’.25

It is important to note that the reference to the space of possibility in DeLanda’s description of the manifold and Salen and Zimmerman’s account of game design is not coincidental. The processes in game design also need to be considered in the number of relevant ways in which they can change. According to some conceptions in physics and mathematics, as DeLanda points out, the object’s instantaneous state, no matter how complex, becomes a single point within an increasingly complex manifold space comprising all its degrees of freedom. Similarly in digital games, while it is possible to have single instances of gameplay, these exist within a manifold consisting of multiple levels of possibility.

According to such a model, objects retain their identities even though they are based in multiplicities. Each game has its own narrative and ludic identity. To return to Sands of Time, here, an instance of gameplay exists within the manifold of the title Prince of Persia: Sands of Time and by extension, as the story claims, of all other PoP games. However, at the same time, the instance of gameplay is a single narrative object with its own identity. On its own, it can be told as a stand-alone story. The concept of singularity explains how objects retain their identity despite being within a multiplicity. A singularity is a special topological feature of manifolds that has a large influence on the behaviour of the trajectories and hence on the whole system. A large number of different trajectories, starting their evolution at very different places in the manifold, may end up in the same final state if they are within the singularity’s sphere of influence. It is possible to allow for transitions from one form to the other when the trajectories break free of the influence of one singularity and come under that of another. DeLanda, therefore, describes multiplicity as being ‘defined by distributions of singularities defining tendencies in a process; and by a series of critical transitions which can take several such distributions embedded within one another and unfold them’.26

In an earlier section, ludic time was located in the order of the aion between the sequential chronos and the eternal kairos. In The Logic of Sense, Deleuze gives the concept of the aion further layers of complexity: it is ‘the past-future, which in an infinite subdivision of the abstract moment endlessly decomposes itself in both directions and forever sidesteps the present’.27 However, computer game events do not just occur in the past-future; rather they occur in the ‘right now’, as noted earlier. DeLanda’s explanation is helpful here. He introduces the idea of virtuality where the ‘right now’ is understood in terms of ‘becoming’ and not discrete instances of being. The idea of the present within a virtual multiplicity, therefore, does not contradict Deleuze’s conception of the aion and even fits well with Kermode’s description of it as the ‘time of becoming’. As DeLanda states

Unlike actual time, which is made exclusively out of presents (what is past and future relative to one time scale is still the living present of a cycle of greater duration) a pure becoming would imply a temporality which always sidesteps the present, since to exist in the present is to be, no longer to become.… And unlike actual time which is asymmetric relative to the direction of relative pasts and futures, a pure becoming would imply a temporality which is perfectly symmetric in this respect, the direction of the arrow of time emerging as a broken symmetry only as the virtual is actualized.28

Time in computer games also behaves like DeLanda’s description above. Each saved game (whether past or future relative to the event of gameplay) is temporally as valid as any other in the timescape of the game. Each is a new beginning and can have various different endings (depending on how many times it is replayed). Each gameplay instance is therefore an actualization of the general virtual multiplicity that constitutes digital-game time. This idea can be extended further within the instance of gameplay itself. Gameplay, constituted of a set of actions, happens in the now ; but the now is not yet the present in terms of being. Therefore, each action is actually in the state of becoming until it is performed and actualised (or as the continuous tense changes to the perfect). Deleuze’s theory is therefore appropriate for describing multitelic entities like digital games. Importantly, his theory itself is intrinsically ludic in nature: to explain the working of the aion, he uses examples from ludic texts like Borges’ Lottery of Babylon and Lewis Carroll’s novels and comments that ‘The Aion is the ideal player of the game’.29

There are many futures and pasts that never happen when considered in relation with actualized instances (because they are ‘asymmetrical’). Deleuze addresses the issue of the unrealized and potential events in his concept of ‘virtuality’:

The virtual is not opposed to the real but to the actual. The virtual is fully real in so far as it is virtual ... Indeed, the virtual must be defined as strictly a part of the real object - as though the object had one part of itself in the virtual into which it plunged as though into an objective dimension.... The reality of the virtual consists of the differential elements and relations along with the singular points which correspond to them.30

This bears a clear resemblance to the many stories created within a computer game. Every instance of gameplay is after all part of a game-system. It follows the game rules, is represented by the game graphics and mechanics and also has the same basic narrative environment. So it can be pointed out that the game consists of a multiplicity of gameplay that corresponds to the singularities of the game-environment, the basic framework that permits certain kinds of progress and not certain others, and the designated beginning(s) and ending(s), while at the same time also following differential tracks of progress depending upon the interaction between the game and the player. Of the multiplicity that a digital game is, each played instance becomes an actualization while the other possible instances remain part of the virtual.

This points towards the other part of the problem: in the realm of the virtual, how can the possible instances be differentiated (and therefore analysed) when their identity has not yet been actualized? This problem is summed up in the philosopher W. V. O Quine’s jocular criticism:

Take, for instance, the possible fat man in the doorway; and again, the possible bald man in the doorway. Are they the same possible man, or two possible men? How do we decide? … What sense can be found in talking of entities which cannot be meaningfully said to be identical with themselves and distinct from one another?31

We, therefore, face the problem of defining differences and repetitions: is the computer game not retelling the same story all over again?

The above objection seems problematic because it is framed solely in terms of linguistic parameters. Instead, by considering games as multiplicities, it is possible to transcend the limitations that language-systems pose towards describing the problem. Games can instead be compared to scientific phase portraits that determine the structure of state spaces. In the phase portrait, according to Ronald Giere, the population of trajectories as a whole play a role in shaping any particular actual history.32 So it is not merely the actualized trajectory that should be studied to understand the state of a particular object, but the whole set of trajectories including the possible and non-actualised trajectories. DeLanda maintains that objections like Quine’s arise only when possible worlds existing alongside actual worlds are postulated in terms of essences. The alternative provided by Deleuze is to ‘avoid taking as a given fully formed individuals, or what amounts to the same thing, to always account for the genesis of individuals via a specific individuation process’.33 In computer gameplay, too, any particular instance is influenced by the possible others. For example, when players in GTA: San Andreas get to choose from multiple missions, or even to avoid missions, they are simply choosing to actualise one possibility, which is constantly influenced by others as the game progresses. Also chessplayers usually ‘see’ a few moves ahead and then select from amongst a series of potential moves. All of these potential moves determine the player’s decision to actualise one of them. Gameplay exists as a developmental process, involving a multiplicity of possibilities, and not as a transcendent essence. To counter Quine’s objection, it can be said that the possible is inextricably intertwined with the identity of the actual, as can be clearly illustrated through instances of gameplay. As Ian Bogost comments, ‘GTA crafts the game experience in terms of a set of relations between possible actions and their consequences … This is where the player must frame his next action in relation to a web of motivations, fears, and preconceptions, both within and without the game’.34

In his recent book, Unit Operations, Bogost agrees that the structure of games is a multiplicity and tries to understand it by drawing upon Deleuze and Guattari’s ideas to analyse the freeform structure in Grand Theft Auto. He sums this up as follows:

Deleuze and Guattari’s project focuses on removing boundaries, in rejecting the idea that boundaries create meaning. Instead, meaning is always provisional, in a state of openness. Freedom in GTA is thus much more like the freedom of the desiring machine [postulated by Deleuze and Guattari] than that of Kantian reason.35

However, he disagrees with Deleuze and Guattari when it comes to describing the nature of multiplicity in computer games. According to him, games exist as unit operations which he defines as a ‘configurative system, an arrangement of discrete, interlocking units of expressive meaning’.36 Unit operations are discrete and hence different from systems operations which are more holistic mechanisms. Though different, these are not in binary opposition, as Bogost clarifies. He adds that systems operations sacrifice openness for certainty and argues that complexity’s macroscopic vision is myopic. According to him, it ignores the importance of the individual within a network, focusing solely on generative structures, instead. He feels that in Deleuzian multiplicity, the constituent rhizomic and nomadic structures tend to move far away from considering individual instances of gameplay. As he states it,

the fundamental difference between nomadism and unit analysis comes to the fore: nomad thought resists thinking of the world in discrete components, devouring individual decision into an amorphous whole. This obstacle stands in the way of nomadism’s embrace of unit operations, despite the apparent similarity of their attempt to disrupt unities of meaning.37

In his attempt to see gameplay as constituted of discrete units that do not repeat themselves, Bogost tries to situate it in a different account of multiplicity.

For him, Alain Badiou’s version of multiplicity is a means of redressing the problem. Badiou applies set theory to ontology and treats the set as a collection of elements selected from a universal set containing infinite elements. The various sets formed are multiplicities by nature. In a process, which he calls the ‘count as one’ (compte-pour-un), every multiplicity is instantiated and treated as a complete whole. In Badiou’s reading, Deleuze’s insistence on continuity leads to an eternal sameness. The issue of difference and repetition is raised yet again and, in Badiou’s scheme, applying Deleuzian ideas in understanding computer games will imply an assumption that all instances of gameplay are one and the same. If this were so, then digital game narratives would certainly not support a Deleuzian analysis. Before arriving at such a conclusion, however, a more in-depth evaluation of Badiou’s position is necessary.

Badiou’s main problem with Deleuze is regarding the virtual. For him the virtual cannot exist alongside the actual and therefore Deleuze’s formulation is a ‘heroic effort … incapable of succeeding’.38 He states quite clearly that ‘contrary to Deleuze, therefore, I think that the “event dice throws” are all absolutely distinct - not formally (on the contrary, the form of all events is the same) but ontologically ... No count can group the events, no virtual subjects them to the One’.39 However, Badiou, in concerning himself with the concept of the One in Deleuze, seems to ignore the idea of immanence that is key in Deleuzian thought. As Todd May observes, ‘Badiou separates his discussion of time from his discussion of the virtual and the actual. This, I believe, is a mistake, since it is primarily from the viewpoint of time that the virtual and the actual can be considered’.40 May clarifies this further in the following comment

This conception of time allows Deleuze to conceive difference in both its virtual and actual aspects without resort to any sort of transcendence. The past coexists with the present in a single time; it is not ontologically transcendent to it. This coexistence is in some sense ontologically One (there is one time) and in some sense not ontologically One (the past is, by virtue of being a virtual difference in kind, ontologically distinct from the present, which is difference in degree) … it is a thought of difference and unity, the Many and the One.41

This relates well to DeLanda’s use of the concept of the singularity and individual trajectories, which are still influenced by the others in the multiple system.

In analysing the problem of difference and repetition, either within the Possible or the actualized instances of gameplay, Deleuze again provides a useful point of entry:

Repetition is no longer a repetition of successive elements of external parts, but of totalities which coexist on different levels or degrees. Difference is no longer drawn from an elementary repetition but is between the levels or degrees of a repetition which is total and totalising every time; it is displaced and disguised from one level to another, each level including its own singularities or privileged points.42

He also goes on to say that repetition includes difference, and in one and the same movement. For game-narratives, this is important because it resembles the unique phenomenon where separate (and different) narrative instances evolve out of the same and binding narrative framework. The basic narrative framework is repeated on different levels (and instances of gameplay) but is also displaced and differentiated because each actualisation follows its own singularity and has its own unique outcome. The story is the same but is played out on a different level, each time. This is similar to Deleuze’s comparison of the multiple levels of repetition to metempsychosis. He says:

Since each is a passing present, one life may replay another at a different level, as if the philosopher and the pig, the criminal and the saint, played out the past at different levels of a gigantic cone. This is what we call metempsychosis.43

This shows a striking resemblance to digital games, especially those like Fahrenheit where the player can play the murderer and the detective in different instances of gameplay. Even in general, the fact that the player in the digital game has many lives and is ‘reborn’, as it were, finds a close parallel in Deleuze’s idea of difference and repetition. Finally, Deleuze’s use of ‘replay’ and ‘played out’ is hardly accidental, considering his consistent use of the ludic metaphor to illustrate key ideas about virtuality, multiplicity and the order of time. This indicates how notions of the ludic and the multiple are necessarily informed by each other.

For Deleuze, the dice game (another ludic metaphor) that signifies events, consists of different ‘throws’ that are formally distinct but ontologically the same. Deleuze also refers to the ‘Divine Game’ that he describes as ‘the most difficult game to understand, impossible to deal with in the world of representation’.44 The computer game is hardly like the Deleuzian ‘Divine Game’ because it cannot be entirely smooth and rhizomic. Yet, Deleuze’s concept is more apposite in thinking about gameplay than Badiou’s assertion that the event dice throws are ‘absolutely distinct’ in terms of their coming to existence. We must remember that often the many ‘different’ games are the outcomes of a single saved game and share a single origin. It is, therefore, problematic to imagine ontologically distinct sets of events emerging from the game. Even the narratives that emerge are actually one narrative. When playing a game like Sands of Time, which is quite tightly scripted in terms of its plot, we still don’t play exactly as instructed in walkthroughs. And even the walkthroughs differ - as do the fates of the players, while in the game. One wrong move might blow the player to smithereens and within a few seconds of having started, she must press the F5 key to load the save again. If therefore, one is to analyse computer games in terms of ‘unit operations’, it would be necessary for those unit operations to be separate but to allow the possibility of crossing over into each other.

On considering the virtual aspect of the multiplicities, the inadequacies of applying Badiou’s model to digital games become more obvious. Consequentially, Bogost’s argument that Deleuze and Guattari’s nomad space does not locate the significance in the gaps between the states is also not relevant. The instances of gameplay are never disjointed and hence it is not necessary to account for any ‘gaps’. Contrary to Bogost’s claim, Deleuze and Guattari definitely recognize the significance of individual events - in terms of virtuality, as noted previously, and even in terms of the nomadic space. Even Bogost agrees that ‘Deleuze and Guattari do offer occasional allowances for gaps or pauses in the nomad’s progress. The nomad, they argue, ‘has a territory; he follows customary paths; he goes from one point to another; he is not ignorant of points’.45 This is especially clear when he acknowledges as the ‘most fungible practical guideline’ of A Thousand Plateaus, the passages and combinations in the operations of smoothness and striation and how the ‘punctuations between deterritorializations and reterritorializations appear to come closest to demarcating the individual “units” of a flow’.46 Bogost raises another issue: he maintains that the ‘local operations’ that occur during the transitions do not allow for preordination or deliberation. However, within the virtual multiplicity, the changes of state occurring under the influence of singularities do not preclude the possibility of deliberate action. Rather, they also account for the ever-important aleatory factors (especially in emergent games) in addition to deliberate actions. Finally, Bogost also concedes that unit operations are not in binary opposition with system-operations: ‘unit-operational structures might also reaffirm systematicity … systems are fluctuating assemblages of unit-operational components rather than overarching regulators’.47 By defining ‘unit operations’ as constitutive of assemblages, a connection with another Deleuzoguattarian idea is made implicit.

The latter point is illustrated well in games like Grand Theft Auto: San Andreas. As Bogost comments ‘GTA does not just provide several different styles of gameplay, it also allows free-form transitions between those play styles’.48 GTA allows players to move within a game space called San Andreas: either to roam the ‘city’ as they please or to play the game’s missions. On selecting a mission they (and therefore the gameplay) enter a singularity that defines the actions and affordances during the length of the mission. Spatially, however, the transition is not clear-cut since the player is free to visit the same places within the mission. Temporally, too, it is very fluid because of all the possibilities enabled by the saves and reloads. GTA exhibits the characteristics of an assemblage because it allows various trajectories of possible narratives to flow into one another in the freeform transitions that Bogost mentions. Within the assemblage of the game-machine, the events therefore exist in a state of becoming till they are actualized within a mission or some other particular ludic situation.

In many ways, therefore, Bogost’s conception finds itself similar to Deleuzian ideas. To represent the functioning of the game system the ‘unit operation’ needs to situate itself within the virtual and the ‘units’ then need to exist in a state of becoming. When actualized, each ‘unit’ will have its own telos but also take part in a common telos (a very literal example being the ‘Game Over’ or exit screen). Finally, within the space of possibility the telic exists both as the divergent as well as the tautological. It is this situation that helps to demonstrate how the computer game narrative can be both never-ending (the Prince of Persia keeps on returning to his story) and finalised, at least until it is played again (the Prince leaves Princess Farah, bewildered by his time-travel tales, and disappears into the jungle). Within the multiple spaces of the computer game, the telos, thus, is not lost - it merely changes, turning into beginnings and repetitions that rework elements encountered before. Nor is it something phenomenally new: it operates in similar ways in other media, though in the digital game-world it is more clearly outlined. In fact, the theoretical apparatus used for the analysis is equally applicable to other kinds of texts. It is possible to say that the reading of the multi-telic digital game also influences our experience of other narrative media. On reaching the end, the reader is left with the Prince’s voice saying, ‘No no … that isn’t how it happened’.

Souvik Mukherjee is currently pursuing a PhD at Nottingham Trent University on videogames as newly emerging storytelling media. He is interested in how videogames relate to canonical ideas of narrative as well as how they also inform and challenge current conceptions of technicity, identity and culture. Souvik blogs about his research and gaming experiences at http://readinggamesandplayingbooks.blogspot.com/.

1. Frank Kermode, The Sense of An Ending (Oxford: Oxford University Press, 1966), p. 21.

2. Kermode, Sense of An Ending, p. 23.

3. Kermode, Sense of An Ending, p. 22.

4. Jesper Juul, ‘A Clash between Game and Narrative’, Master’s thesis, v. 0.99 (2001), http://www.jesperjuul.net/thesis/ [accessed 22 March 2008].

5. Craig Lindley, ‘The Gameplay Gestalt, Narrative and Interactive Storytelling’, http://www.tii.se/zerogame/pdfs/CGDClindley.pdf [accessed: 22 March 2008].

6. Kermode, Sense of An Ending, p. 25.

7. Gérard Genette, Narrative Discourse: A Study in Method, trans. Jane E. Lewin (Oxford: Blackwell, 1980), p. 115.

8. Hence the name, Sands of Time, signifying the sands on coast of the ‘sea of Time’.

9. Screenplay of Sands of Time from POP Legacy.com http://poplegacy.planets.gamespy.com/PoPSoTHealers_ScreenplaySOT.php [accessed 22 November 2007].

10. Barry Atkins, ‘Killing time: time past, time present and time future in Prince of Persia: Sands of Time’, in Videogame, Player, Text, ed. Barry Atkins and Tanya Krzywinska (Manchester: Manchester University Press, 2007), p. 243.

11. Though released after Sands of Time, and in keeping with the story, it is difficult to establish any chronological order and hence to call them sequels would not be accurate.

12. Gonzalo Frasca, ‘Ephemeral Games: Is It Barbaric to Design videogames after Auschwitz?’, http://www.ludology.org/articles/ephemeralFRASCA.pdf [accessed 23 November 2007].

13. Frasca, ‘Ephemeral Games’.

14. Atkins, ‘Killing time’, p. 251.

15. Jesper Juul, ‘Introduction to Game Time’ in First Person: New Media As Story, Performance, and Game ed. Noah Wardrip-Fruin and Pat Harrigan (Cambridge, Mass.: The MIT Press, 2004), p. 134.

16. James Newman makes this point in Videogames (London: Routledge, 2004), p. 103.

17. Juul, ‘A Clash between Game and Narrative’.

18. Kermode, Sense of An Ending, p. 139.

19. Kermode, Sense of An Ending, p. 117.

20. Kermode, Sense of An Ending, p. 124.

21. Espen Aarseth, ‘Playing Research: Methodological Approaches to Game Analysis’, http://www.spilforskning.dk/gameapproaches/GameApproaches2.pdf [accessed 23 November 2007].

22. Katie Salen and Eric Zimmerman, Rules of Play: Game Design Fundamentals (Cambridge, Mass; London: MIT Press, 2004), p. 67.

23. Manuel DeLanda, Intensive Science and Virtual Philosophy (London: Continuum, 2002), p. 12.

24. DeLanda, Intensive Science and Virtual Philosophy, p. 12.

25. DeLanda, Intensive Science and Virtual Philosophy, p. 13.

26. DeLanda, Intensive Science and Virtual Philosophy, p. 20.

27. Gilles Deleuze, The Logic of Sense, trans. by Charles Stivale and Mark Lester (London: Athlone, 1990), p. 77.

28. DeLanda, Intensive Science and Virtual Philosophy, p. 127.

29. Deleuze, Logic of Sense, p. 64.

30. Gilles Deleuze, Difference and Repetition, trans. Paul R. Patton (London: Athlone, 1994), p. 209.

31. Nicholas Rescher, ’The Ontology of the Possible’, in The Possible and the Actual: Readings in the Metaphysics of Modality ed. Michael J. Loux (Ithaca: Cornell University Press: 1979), p. 177.

32. Ronald N. Giere, ‘Constructive Realism’ , cited in DeLanda, Intensive Science and Virtual Philosophy, p. 32.

33. DeLanda, Intensive Science and Virtual Philosophy, p. 37.

34. Ian Bogost, Unit Operations (Cambridge, Mass.: The MIT Press, 2006), p. 155.

35. Bogost, Unit Operations, p. 156.

36. Bogost, Unit Operations p. i.

37. Bogost, Unit Operations, p. 142.

38. Alain Badiou, Deleuze: The Clamor of Being, trans. Louise Burchill (Minneapolis: University of Minnesota Press, 2000), p. 45.

39. Badiou, Deleuze, p. 46.

40. Todd May, ‘Badiou and Deleuze on the One and Many’, in Think Again: Alain Badiou and the Future of Philosophy, ed. Peter Hallward (London: Continuum, 2004), p. 74.

41. May, ‘Badiou and Deleuze’, p. 76.

42. Deleuze, Difference and Repetition, p. 358.

43. Deleuze, Difference and Repetition, p. 105 (my italics).

44. Deleuze, Difference and Repetition, p. 353.

45. Bogost, Unit Operations, p. 144.

46. Bogost, Unit Operations, p. 144.

47. Bogost, Unit Operations, p. 4.

48. Bogost, Unit Operations, p. 155.