Marie-Laure Ryan, the well-known scholar on narrative theory and electronic textuality, states in her recent

book Avatars of Story (2006) that classical narratology,

originally designed for the analysis of standard written text, does not work well for interactive textuality or
computer games. She suggests that narratology, which
she characterizes as an unfinished project, should reposition itself on a transmedial and transdisciplinary track
to redefine itself as a transmedial narratology (2004,
p. 1). The main problem of the transmedial narratology in relation to computer games is according to her the
question of how to reconcile narrativity and interactivity.
With her plea for a project of transmedial narratology Ryan explicitly takes position in the debate between
ludologists and narratologists, that started in 2001 with
the publication of the editorial of the first issue of the
online journal Game Studies (Aarseth 2001). Ludologists
claim that games and narratives are mutually exclusive,
whereas narratologists not only claim that games have
narrative dimensions, but also that gameplay can only
be understood in relation to the narrative dimensions.
After almost a decade of debate between ludologists
and narratologists all participants acknowledge that the
main problems of the interaction between narrative
structures and the mechanics of gameplay remain to be
solved. The debate between ludologists and narratologists, heated by the political urge to demarcate the new
field of game studies, obfuscates the fact that the emerging discipline of narrative analysis, in particular the
structural narrative semiotics as defined by Barthes and
Greimas, already dealt with these problems. A structural narrative analysis of Tetris will not only show that
Tetris has narrative structures, but also that the strategic and communicative dimensions of computer games
can be studied using narrative models and categories
developed in the 60s and onwards.
1. Tetris-studies
Although ludologists and narratologists disagree about
how narrative concepts should be used in game studies,
they agree about the fact that some games have narrative dimensions and others not. Chess, Go, or Tetris
are non-narrative games because they do not fill the
basic conditions of narrativity, namely offering an image of life by creating a concrete world populated by intelligent agents whose actions make this world evolve
(Ryan 2005). The question of the abstract games plays
a very prominent role in the debate between ludologists and narratologists with Tetris as piece de rsistance.
Ludologists refer to Tetris to explain that games not
necessarily need a narrative dimension and that the
performative dimension can be studied separately from
the narrative dimension. Narratologists accuse the ludologists of neglecting the narrative dimensions of games and of looking only at things that apply to Tetris
(Montfort 2004). They caricature ludology tauntingly
as Tetris-studies.

E|C Serie Speciale

Anno III, nn. 5, 2009, pp. 31-36

ISSN (on-line): 1970-7452

ISSN (print): 1973-2716

EC

Bridging the NarratologyLudology Divide. The Tetris Case

Jack Post0
But does Tetris really lack any narrative dimension? At
first we should distinguish concepts like narration and
narrative from narrativity. To construct a medium-free
and universal transmedial narratology, Ryan defines
narrative as a cognitive construct that transcends media, disciplines, and historical as well as cultural boundaries (2006, pp. 1-2, 102). As a consequence thereof,
anything, even life itself, can provoke stories in the mind
of a cognizing subject. To differentiate these more diffuse experiences of narrative from the narratives in the
proper sense of the word, she introduces two narrative
modalities: having a narrative indicates that a semiotic object is able to invoke a narrative script and being
a narrative means that a semiotic object is consciously
produced with the aim to evoke a narrative image and
is recognized as such (ivi, pp. 10-11). Narrative is thus
on the one hand defined as a mental image constructed
by the interpreter and on the other hand as a particular meaning that is encoded in a text. We should differentiate narrative as defined by Ryan from narrativity,
which according to Greimas, forms the very organizing
principle of all discourses, whether narrative, non-narrative, figurative or abstract (Greimas & Courts 1982,
p. 209). The fact that some games are more abstract
and less figurativized and iconic than others, does not
necessarily mean that they lack narrativity. The definition of narrative in terms of time, settings, characters,
and events, limits the ludologists as well as narratologists approach to what Roland Barthes calls the referential surface level of the text (1977, p. 111).
2. A structural analysis of narrative
When Ryan states that narratology as the formal study
of narrative has been dormant for forty years and has
never developed into a full-scale transmedial narrative
theory (2004, p. 1), she refers to the publication of the
Special Issue on Structural Analysis of Narrative of the

2009 AISS - Associazione Italiana di Studi Semiotici

T. reg. Trib. di Palermo n. 2 - 17.1.2005

32

French journal Communications in 1966 (Barthes et alter

1966). Especially the essays of Barthes and Greimas paved the path for a more general semiotic approach of
the narrative dimensions of discourse. In the very first
paragraph of his seminal essay Introduction to the Structural
Analysis of Narrative Barthes states that the narratives of
the world are numberless and distributed amongst very
different substances (languages, gestures, images) and
present in many genres to which we could of course
add computer games (1977, p. 79). Because an inductive analysis of this variety of narratives is doomed to fail,
narrative analysis, argues Barthes, is only conceivable as
a deductive project, that departs from a hypothetical
model of description and gradually works down from
this model towards the different narrative species which
at once conform to and depart from the model (ibidem,
p. 81). Which means that narrative analysis can never
be, as Ryan argues, essentially a taxonomical project
(2006, p. 120).
Narrative analysis has according to Barthes to decronologize the narrative continuum and to relogize it
(1977, p. 99). Referential time, space and characters
should in other words be distinguished from the narrativity defined as the logical organization of the underlying narrative structures. On deeper analytical levels the superficial level of the referential illusion is
analyzed in terms of logic, paradigmatic oppositions,
functions, actants and actions.
Although Ryan claims that narratology is the formal
study of narrative (2004, p. 1), her transmedial narratology doesnt seem to account for the underlying
narrative deep structures. Nor does Ryans definition
of computer games as narratively organized systems
for playing (2006, pp. 8-9, 197; 2004, pp. 349-350)
or the study of computer games as a functional ludonarrativism that studies how the fictional world, realm
of make-believe, relates to the playfield, space of agency (2006, p. 203), really answer the question how the
strategic dimension of gameplay and the imaginative
experience of the fictive world are related. In her own
words:
could the same system of rules (provided we are able to
determine what is a rule and what is not) be narrativized in
many different ways, or is there an organic, necessary connection between rules and narrative ? Do the problems presented to the player grow out of the narrative theme, or are
they arbitrarily slapped upon it? (Ryan, in Montfort 2005)

It was Barthes, who forty years earlier, formulated a

possible answer to this question. He observed that the
subject of many narratives is often a dual subject,
that is, a subject based on the archaic structure of two
adversaries who dispute over a stake. This dual subject
relates the structure of narrative to that of (modern)
games:

Jack Post Bridging the Narratology-Ludology Divide. The Tetris Case

two equal opponents try to gain possession of an object

put into circulation by a referee; a schema which recalls the
actantial matrix proposed by Greimas, and there is nothing
surprising in this if one is willing to allow that a game, being
a language, depends on the same symbolic structure as is to
be found in language and narrative; a game too is a sentence. (1977, p. 108)

Hence games, narratives and language all share the

same symbolic structure, which Greimas calls a simple narrative (1982, p. 203). Does this mean that the
narrativity of a simple and abstract game like Tetris can
be linked to the performative dimension of its gameplay?
3. Narrative analysis of Tetris
Following Barthes proposal to analyze a game as an
actantial matrix, we should be able to describe the
manipulation of the blocks in Tetris as an interaction
between actants. In Greimasian semiotics, which attends to signification in a broad sense, actants are defined
as abstract syntactical units of the discourse which undergo an act (ivi, p 5). Actants are thus not individuals
or material things, but formal actantial positions which
define each other reciprocally. The actant-subject only
exists in conjunction or disjunction with the actantobject and vice versa. The subject strives after the
object which is always a value-object for a subject. The
object in Tetris invests for instance the goal of the game
with all its stakes. The actants, who only exist on the abstract depth level of the semio-narrative structures, are
anthropomorphized as actors on the more superficial
discursive level and finally textualized as a referential
world which consists of a playfield, falling blocks and a
player1. The simple narrative of Tetris can be analyzed as a base Narrative Program (NP) which consists
of several sub-programs (instrumental NPs) which are
necessary for the realization of the NP2. A NP can be
described as a transformative doing (Dt) in which a beneficiary subject (S2), who initially is in disjunction ()
with the value object (Og = gameplay), is at the end of
the game in conjunction () with the value object: Dt
[S1 (S2 Og) (S2 Og)].
Given that the state which precedes the transformation
is presupposed by the process, it is common practice
to write the formula in a shorter way: Dt [S1 (S2
Og)]. Our hypothesis is that the operator subject and
the beneficiary subject of the abstract semio-narrative
level correspond on the discursive level with only one
actor, the player, then they are in an actorial syncretism
(S2 = S1) and should the formula be rewritten as Dt
[S1 (S1 Og)] (ivi, p. 326). The base program of
Tetris can only be realized through a whole series of
embedded instrumental NPs, such as the opening of
the game, the reading of the help files, the starting of
the game, and the series of manipulation of the blocks.
The following schema shows the complex base NP of
Tetris with some of its instrumental NPs.

Fig. 2 Hierarchy of narrative trajectories in the canonical schema of the quest

Fig. 1 Base Narrative Program of Tetris

4. The canonical schema of the quest

The starting of the game installs a subject operator
with the modality of a /wanting-to do/ (wanting to
play Tetris) and a value object the gameplay which
is strived after by the operator. But before the subject
is able to play the game, it has to acquire the necessary competences to play the game. A first time player
acquires these competences by opening the help files
which installs two new actants: the Sender (Sr) and
Receiver (R). The subprogram of the acquisition of the
competence is different from the base NP, because the
Sender manipulates and sanctions the Receiver from a
level that transcends the narrative universe in which the
subject operator accomplishes its narrative trajectory
(Greimas 1982, pp. 206, 294). The Sender invests the
Receiver thus with a new value object (Okn) namely
a /knowing how to do/: Dt [Sr (R Okn)]. Note
that the Receiver on the discursive level is in an actorial
syncretism with subject operator (R = S1), because both
correspond to the actor player. The Sender not only
endows the Receiver with the competences to play the
game, but it also informs the Receiver about the rules
of the game and asks the Receiver to obey the rules.
The Receiver in turn has to accept or to reject the terms
the contract offered by the Sender. Hence the Sender
not only determines which values are at stake in the
game (the object) but also how the Receiver has to act (according to the rules of the game). The Sender returns at the

end of the game as a subject judicator who determines

whether the operator has acted conform the terms of
the contract (Greimas 1982, p. 267). The Sender judicator decides in other words about the final modality of
the /being able to do/ and publishes the game score.
The whole programming of the action in Tetris corresponds exactly to what Greimas (ivi, p. 204) following
Propp (1968), defines as the canonical schema of the
quest3. The schema of the quest is not just a conflict
between two actants over an object, but it always implies the transfer and definition of the values which give
meaning to the trajectory of the subject. The quest schema puts two couples of actants into play (Sender and
Receiver, and subject and object) which each its own
narrative trajectory, for the Sender and the Receiver:
Contract Action Sanction and for the subject and
object: Competence Performance Consequence
(Fontanille 2006, pp. 73-75). Because the Sender defines the values at stake and determines the action programs of the subject and object, both trajectories are
hierarchically organized in the sense that the trajectory
of the Sender and Receiver controls and subsumes the
trajectory of the subject and object. To sum up we can
schematize the Narrative Program of Tetris as follows:
With the opening of the game a subject operator is installed and modalized with a wanting-to-do, the competence (knowing-to-do) is acquired with the acceptance of the rules and the definition of the value object,
the performance (being-able-to-do) is realized through
the execution of the rules and the quest for the value
object, and finally the action is sanctioned with the registration of the scores.
5. The dual subject in games and narratives
Because each transformation necessarily takes place in
a field of forces, in which the operator has to overcome resistances, every narrative program is necessarily
doubled with a counter-program. Although the players
of Tetris do not play against real adversaries, the gameplay presupposes a polemical relation between two
diametrically opposed NPs, each endowed with its own
operator, beneficiary subject, value object, instrumental
NPs et cetera. The player of Tetris plays not against an
explicitly manifested anti-subject but against an in the
computer game objectivized or neutralized counterprogram. Tetris is in other words an automaton, the

E|C Serie Speciale Anno III, nn. 5, 2009

33

34

Fig. 3 Narrative program of Tetris

simulacrum of a programmatic doing defined by

Greimas as a (neutral) operator subject in possession
of a group of explicit rules and an order requiring the
application of these rules or the carrying out of instructions (1982, p. 20). The game is after all nothing else
than a translation of the official Tetris guidelines into
the algorithms of the computer language. Hence the
player plays against the machine and has to develop
strategies to undo or avoid the execution of the counter-programs of the automaton. This is what Barthes
meant when he stated that the subject of many narratives and games is a truly dual subject (see also
Landowski 1989, p. 242-44).
Aarseth argues that games are not accessible for narratological analysis because the player pursues a goal
in the future and games therefore are always forward
oriented (2004, p. 333). Playing games is an act or performance, and to account for the complex and unpredictable event structures of games, the strategic analysis of the performative dimension should carefully be
distinguished from the reflective retrospective analysis

Jack Post Bridging the Narratology-Ludology Divide. The Tetris Case

of the narrative dimension (ivi, p 369). Aarseth makes

an exception for non-narrative games like Tetris and
Quake:
In Tetris there is no final solution, just harder and harder
situations, until the player makes one mistake too many. And
in a Quake death match, like in chess, there are no specific
ways to progress; you win by killing your opponents any which way you can and more times than they. (ivi, p 369)

Both narratologists and ludologists consider valid the

assertion that narratives are backwards oriented and games forwards oriented: an important argument to distinguish games from narratives. Both characterize Tetris
as an abstract, non-narrative game, because it lacks the
retrospective dimension of the narrative. Ryan states
that some games can be considered narratives because
we are able to attribute meaning to them retrospectively
in the retelling of the game (2004, pp. 333-334). But as
Jacques Fontanille argues, the meaning of any action,
narrative or non-narrative, is only apprehensible a posteriori, thus when the process is closed. Narrative schemas

are thus always based on a retrospective reading of the

course of action:
starting from the end the sanction recognition, compensation, or punishment one is invited to discover which fact
carries the sanction, that is, to identify the consequence of the
action. Starting from the consequence, one may reconstitute
what led to it, beginning with the performance itself. Then,
starting from the performance, one may calculate the conditions that had to be established first and for all, the competences that it was necessary to acquire, etc. The intentionality of
the action can thus only be retrospective. (2006, p. 133)

The term Narrative Program already suggests that a

narrative trajectory is always programmed, that is, calculated from the point of view of the expected outcome. A prospective programming of the action differs
from a retrospective programming in that the relations
between the elements in the concatenation are not necessary but possible and contingent (see also Bremond
1980). This does not necessarily mean that the actant
has no control over the action. The actant may, according to Fontanille, program the action in three ways
(2006, p. 134). The actant may calculate the trajectory
backwards, starting from the desired end situation or
make use of stereotyped schemas. These two forms of
retrospective logic, which are characteristic for Tetris,
are still closely related to the logic of the action and
can therefore be called tactics. Landowski defines tactics as the science of the actualizing maneuvers in
which the interactional effect comes from the fact that
S1 masters the objective circumstances of the doing
of S2 (1989, p. 239). Tactics only concern an operator
and the application of a set of rules and should therefore be located on the local level of the actualization
of instrumental NPs. The third way of taking control
of the action is by using strategies. When confronted
with unexpected situations in which no backwards reasoning or accumulated knowledge suffice, such as in
complex games as World of Warcraft in which the player
plays against the Artificial Intelligence of the program
as well as against other players, the operator has to develop a strategy. Strategies are based on improvisation
and try to respond to the unpredictable circumstances
of the event structures of the play. A strategy differs
from tactics because it concerns next to the operator
one or more subjects which can be considered as real
anti-subjects who have a relative mastery over their own
programmatic doing (ivi, p. 239). Strategies operate on
a global level and not on the local level of the instrumental NPs, because the operator manipulates the antisubjects competence for making decisions. Although
the strategy induces to what Fontanille calls an open
trajectory envisioned prospectively (2006, p. 135), the
mode of reasoning remains nevertheless retrospective. The invention of counter-strategies relies namely
always on the construction of virtual NPs of which the
intentionality is oriented retrospectively. Consequently

strategies as well as tactics enlarge the number of narrative programs which also means that the (narrative)
identity of the subjects and anti-subjects not only is acquired in relation to the value-object but also in a continuous polemical interaction with projected counterprograms. We should therefore conclude that Aarseths
strategic analysis is still governed by a retrospective narrative logic and should be based on narrative models
and categories.
The discussion of the narrativity of computer games, and Tetris in particular, can be extended beyond
an analysis of the actants of the narration alone. As
Barthes in 1966 stated, a narrative is a narration and
an object of communication between a donor and a
receiver at the same time (1977, p. 109). Donor and
receiver, the actants of the communicative situation,
are like the actants of the narration never real living
persons (ivi, p. 111) but paper beings, immanent to
the narrative and only accessible to a semiotic analysis. Hence the communicative act of playing games is
itself a minimal story, an action which can be analyzed as a NP. The performative dimension therefore has
its own communicative doing (enunciation), actants
(enunciator and enunciatee), objects (utterance), strategic dimensions and instrumental NPs which control the
narration. A detailed analysis of the communicative act
of playing Tetris (with the analysis of its interactivity
and the soft- and hardware interfaces) lies beyond the
scope of this article, but would indicate that not only
the game itself but also the playing of the game (its
performative dimension according to Aarseth4 ) can be
analyzed by using narrative models.
6. Conclusion
Talking about narrativity in relation to games should
thus go beyond the common sense definition of a narrative (even defined as a cognitive construct) with characters that figures in place and time. Defining narrative in these terms leads to rather crude formulations as
that the narrative element of the computer games is subordinated to the playing action and therefore nothing
more than an accessory affective hook or narrative garb that lures the players into the game (Ryan
2004, pp. 10, 349; 2006, p. 197), that stories are just
uninteresting ornaments or gift-wrapping to games
(Eskelinen 2001) or that studying the narrative of Tetris
is just a waste of time and energy (Aarseth 2004, p.
365). Narrativity is on the contrary, not ornamental
or accessory, but the very organizing principle of all
discourse (Greimas & Courts 1982, p. 209). The very
fact that the narrative deep structures are constitutive
of semiotic processes, opens up the signification systems
of computer games to semiotic analysis: for a narrative
semiotics of the classical structuralist kind, but also for
the recent semiotics of discourse (Fontanille 2006). This
would lead us out of the unfruitful dichotomy of ludology versus narratology.

E|C Serie Speciale Anno III, nn. 5, 2009

35

Evaluating the interplay between narrative and gameplay thus starts with the analysis of the narrative (semiotic) deep structures that govern the more superficial
discursive structures of figurative and referential nature. An abstract and non-narrative game like Tetris has
narrative structures, not because it has settings, events and characters, but because of its complex NP and
tactic dimensions, and because the interactivity of its
gameplay can be analyzed in narrative terms. To bridge
the divide between ludology and narratology, that is, to
reconcile narrativity and interactivity, we need paradoxically where Barthes in 1966 called for, a structural
analysis of narrative.
Notes
0

36

This article is part of a research project funded by the

Netherlands Organisation for Scientific Research (NWO).
1
The Greimasian approach uses a generative semiotic model
of the constitution of meaning which moves up from the most
simplest and abstract fundamental elementary structures
toward the most complex and concrete surface manifestations
(Greimas 1982, p. 132-134).
2
In Greimasian semiotics a Narrative Program is the most
elementary syntagm used to represent an action. A NP can
be simple or complex. A complex NP requires the preliminary realization of one or more other NPs. The general NP is
called a base NP and the NPs presupposed and necessary for
its realization instrumental NPs (Greimas 1982, p. 245-246).
3
See for a recent discussion of the model of the Quest in
relation to computer games Aarseth 2004; Montfort 2004;
Lvlie 2005.
4
Aarseth (2004, p. 369) contrasts, following Tronstad (2001),
quests as performatives which belong to the order of the act,
with stories as constatives which belong to the order of meaning.

Bibliography
Aarseth, E.J., 2001, Editorial: Computer Game Studies, Year
One, in Game Studies, n. 1, vol. 1, available: http://www.
gamestudies.org/0101/editorial.html.
Aarseth, E.J., 2004, Quest games as Post-Narrative
Discourse, in M.-L. Ryan, ed., pp. 361-376.
Barthes, R., 1966, Introduction lanalyse structurale des rcits, in Communications, n. 8, pp 1-27; En. tr. Introduction
to the Structural Analysis of Narratives, in Image, Music,
Text, London, Fontana, 1977, pp. 79-124.
Barthes, R. et. alter, 1966, Lanalyse structural du rcit,
Communications, n. 8.
Bremond, C., 1966, La Logique des Possibles Narratifs,
in Communications, n. 8, pp. 60-76; En. tr. The Logic of
Narrative Possibilities, in New Literary History, n. 11, pp.
387-411.
Eskelinen, M., 2001, The Gaming Situation, in Game
Studies, n. 1, vol. 1, available: http://www.gamestudies.
org/0101/eskelinen/.
Fontanille, J., 1998, Smiotique du discours, Limoges, Pulim; En.
tr. The Semiotics of Discourse, New York, P. Lang, 2006.

Jack Post Bridging the Narratology-Ludology Divide. The Tetris Case

Greimas, A.J. & Courts, J., 1979, Dictionnaire raisonn de la thorie du langage, Paris, Hachette; En. tr. Semiotics and Language.
An Analytical Dictionary, Bloomington, Indiana University
Press, 1982.
Landowski, E., 1989, La Socit rflchie. Essais de socio-smiotique,
Paris, Editions du Seuil.
Lvlie, A.S., 2005, End of story? Quest, narrative and enactment in computer games, Digital Games Research Conference
2005, Changing Views: Worlds in Play, Vancouver, available:
http://www.digra.org/dl/db/06276.38324.pdf.
Montfort, N., 2004, Computer Games at SSNLs Narrative
Conference (Against Tetris Studies), in GrandTextAuto,
available: http://grandtextauto.org/2004/04/25/computer-games-at-ssnls-narrative-conference/.
Montfort, N, 2005, Aarseths Anti-Quest. GrandTextAuto,
in GrandTextAuto, available: http://grandtextauto.
org/2005/01/10/aarseth-anti-quest/.
Propp, W., 1968 [1928], Morfologija skazki, Leningrad, Nauka;
En. tr. Morphology of the Folktale, Austin, University of
Texas Press, 1968.
Ryan, M.-L., ed., 2004, Narrative Across Media. The Languages of
Storytelling, Lincoln, University of Nebraska Press.
Ryan, M.-L., 2005, Narrative and the Split Condition of
Digital Textuality, in dichtung-digital, n. 1, available:
www.dichtung-digital.com/2005/1/Ryan.
Ryan, M.-L., 2006, Avatars of Story, Minneapolis, University
of Minnesota Press.
Tronstad, R., 2001, Semiotic and Non-Semiotic MUD
Performance, COSIGN conference, Amsterdam, available:
http://www.cosignconference.org/downloads/papers/
tronstad_cosign_2001.pdf.

Computer games
Tetris, by Pazhitnov, A., 1985, Various.
World of Warcraft, by Blizzard Entertainment, 2004, Vivendi
Universal.
Quake, by id Software, 1996, Various.