University of Southern California Dissertations and Theses

Of worlds and avatars: a playercentric approach to videogames

(USC Thesis Other)

Of worlds and avatars: a playercentric approach to videogames

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
OF WORLDS AND AVATARS:
A PLAYERCENTRIC APPROACH TO VIDEOGAME DISCOURSE

by
Robert Buerkle

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CINEMA-TELEVISION (CRITICAL STUDIES))

December 2008

Copyright 2008 Robert Buerkle

ii

Acknowledgements

Though I cannot, in such a brief space, give due credit to the countless people
who have helped in this project, there are a few that I would be remiss to not mention by
name.
Foremost, I thank the co-chairs of my dissertation committee, Tracy Fullerton and
Anne Friedberg. Tracy has influenced my thinking on videogames more than any other
person during my doctoral career, and Anne saved me by generously agreeing to come
onboard and contribute to this project. To both, I am forever grateful. I am additionally
grateful to Doug Thomas for serving as the outside member of my committee, as well as
Marsha Kinder and Rick Jewell for all their help during my qualifying exams.
I thank all of the good friends who have kept me sane and confident through their
unwavering support, including Paul Reinsch, Chris Hanson, Dan Herbert, Scott Ruston,
Brian Jacobson, and many, many others. I thank my wonderful family—my mother and
father, Jim, Patty, and Sam—for all their encouragement, despite the thousands of miles
between us. And I thank the kind workers at Panera Breadworks in Newport Beach,
who served me lots of hot tea and Fuji Apple Chicken Salads throughout the writing of
this lengthy document.
Lastly, I thank Caitlin: I would not be the scholar nor the man I am today without
your love and support.

iii

Table of Contents

Acknowledgments ii

Abstract iv

Introduction: VIDEO/GAME 1

Chapter 1: SYSTEM 16
Objects and Experiences: Videogames as Game-Texts 23
Videogames-as-Texts vs. Videogames-as-Games 42
Discursive Systems: the Semiotics of Videogames 66
Systems of Rules and Spaces of Possibility: Play and the Aesthetics of 104
Emergent Space
Discursive Systems of Rules: New Media, Iterative Discourse, and the 131
Combination of Signification and Play

Chapter 2: WORLD 137
Temporary Worlds: The Play-ground and the Diegetic Space 145
Narrating Space: A Diegetic Model of Videogames 162
Formal Transparency and Maintaining the Play-Ground: Videogames 190
as Referees
Participatory Storytelling and Narrating the Diegesis: Videogames as 209
Dungeon Masters

Chapter 3: AVATAR 257
Boomerang Etymology: The Evolution of a Word 264
A Spectrum Approach to Dual-Positioning 276
A Multiplicity of Positioning Cues: Textual Influences on Player 283
Positioning
The Ends of the Spectrum 328

Chapter 4: PLAYER 346
This is Your Life: Videogames and Second-Person Narration 353
Fiction and Function: The Dual Roles of the Avatar 375
Frames of Embodiment: The Experience of Narration 397
The Borders of the Frame: Immersion, Reflexivity, and the Fourth Wall 424

Conclusion 443

Bibliography 451

Appendix A: Videogames Referenced 460

iv

Abstract

A study of the medium-specific properties of videogame discourse, this
dissertation considers the videogame situation as the interaction of four component
parts: a PLAYER, participating in a formal SYSTEM that projects a fictional WORLD, via
the mediation of an AVATAR, and these four components form the basis for the four
chapters of this project.
Chapter one begins by looking at the SYSTEM of a videogame—the formal
structure of rules and aesthetics that comprises the thing itself. Questioning what sort of
“thing” a videogame is, this chapter proposes the conceptual model of a game-text, a
unique object combining the discursive properties of texts with the spontaneous activity
that is a game.
Chapter two examines the WORLD of the videogame by suggesting a diegetic
model of videogame discourse, one rooted in narration’s ability to “mentally transport” its
audience to an imaginary space. Videogames are virtual activities, occurring apart from
the player’s here and now, and as such, require a mode of discourse capable of
projecting the player into a non-existent (that is, fictional) time and space. In this regard,
narration is addressed as a manner of “spatiotemporal projection.”
Chapter three is concerned with the most fundamental device of the videogame
interface: the AVATAR. Probing the avatar’s relationship to the player (and vice versa),

v
this chapter proposes a dual-positioning model of videogames, wherein the player is
always positioned in two places at once—inside the game, as a diegetic participant, as
well as outside the game, as a disembodied observer, experiencing the game through
both roles simultaneously.
Finally, chapter four considers the videogame PLAYER by addressing the
medium as manner of second-person narration. Unlike conventional narration, which
has long centered on first- and third-person address, videogames privilege the narratee
rather than the narrator, suggesting the audience as the central actant in the game’s
fiction. Ultimately, then, this dissertation offers a “playercentric” approach to videogame
discourse—one that examines the discursive situation not via its author or designer, but
rather through its receiving audience: the player.

1

Introduction: VIDEO/GAME

Before we begin, let’s clear some assumptions from our minds:
A videogame is not the same thing as a movie. Nor a television program. Nor a
novel. Nor a stageplay, comic book, painting, photograph, sculpture, symphony, dance
routine, or any other type of art or media. Seems obvious enough, right?
Further, a videogame is not the same thing as a board game. Nor a card game.
Nor a sport. Nor a parlor game, guessing game, dice game, paper-and-pencil
roleplaying game, carnival attraction, casino table, logic problem, jigsaw puzzle, or any
other type of game, contest, or amusement. Seems obvious as well…right?
And so long as we’re at it, let’s make one more clarification: a videogame is not
the same thing as hypertext, virtual reality, the internet, computer animation, DVDs,
interactive movies, digital photography, cellphones, iPods, DVRs, office productivity
software, or any of the many other objects we toss under the broad aegis of “new
media.”
These all seem fairly self-evident, don’t they? Well, oddly enough, not always.
From their inception, videogames have always been labeled as somewhat of a
second-class medium—something commonly thought of as being “for kids,” a mere
trivial entertainment, and not something on which serious adults, serious aesthetes, or

2
serious scholars would spend their time. Roger Ebert, for example, gave voice to this
societal disesteem in claiming that videogames are “inherently inferior to film and
literature,” citing their opposition to our traditional conception of great art:
Video games by their nature require player choices, which is the opposite of the
strategy of serious film and literature, which requires authorial control. …I
believe the nature of the medium prevents it from moving beyond craftsmanship
to the stature of art. To my knowledge, no one in or out of the field has ever
been able to cite a game worthy of comparison with the great dramatists, poets,
filmmakers, novelists and composers. That a game can aspire to artistic
importance as a visual experience, I accept. But for most gamers, video games
represent a loss of those precious hours we have available to make ourselves
more cultured, civilized and empathetic.
1

Not surprisingly, these comments led to an extensive bout of message board griping
from fanboys across the web, yet outside the community of gamers, these remarks
represent a standard notion of videogames in contemporary society—yes, they look very
impressive, but ultimately, they are a trivial medium that amounts to wasted time on
childish pursuits.
Ebert’s comments do seem mildly hypocritical, of course, coming from a man
who champions cinema, a medium that received strikingly similar criticisms during its
own infancy—that it could never measure up to literature, theater, and other serious art,
and would remain only a trivial amusement for children and the ignorant masses. But
we’ll let that point pass.
Note the assumption that Ebert’s comments make—that in order to prove their
aesthetic worth, videogames must compete with traditional media at their own game.
“Art,” it seems, is what literature, cinema, painting, music, and other such media already
do, and thus in order for videogames to be considered a legitimate pastime that is not “a
loss of those precious hours,” they must prove themselves by the standards of previous

1
Roger Ebert, “RogerEbert.com: Answer Man,” Digital Chicago, Inc.,
http://rogerebert.suntimes.com/apps/pbcs.dll/section?category=ANSWERMAN&date= 20051127
(accessed 7/26/2007).

3
forms. And sadly, those who champion videogames have often felt the need to do just
this. As a result of the medium’s low societal and cultural esteem, even those who take
videogames seriously have felt the need to associate them with more “legitimate”
objects, resulting in a desire—particularly among critics and scholars—to link
videogames to longer, more esteemed traditions by attaching them to more culturally-
ingrained forms. This generally manifests in one of two ways.
The first is to place videogames among more traditional arts, particularly cinema,
television, and other screen media. Media studies scholars—those with backgrounds in
art, film, communications, English, or some related field—are the most common in using
this approach, attempting to discuss videogames using the same methodologies that
they would to discuss movies, literature, painting, and the like. They discuss Super
Mario Bros. and Half-Life as though discussing Citizen Kane or A Tale of Two Cities,
thus hopefully lending a modicum of credibility to such scholarship. Some common
approaches are to associate videogames with the narrative tradition by focusing on
videogames’ unique methods for telling stories (as well as the ways in which they
appropriate or adapt previous narrative techniques), or to instead focus on the
audio/visual design and stylistic tropes of videogames, linking videogames to the
representative traditions of painting and photography, as well as cinema and modern
media. Regardless of the approach, the point, in the end, is to place videogames within
a tradition of “serious” art.
2

Yet this alone does not seem enough. Videogames’ ill reputation—whether as
childish, as violent, as gender-exclusive, as politically problematic, as pulp

2
This is not a new phenomenon, however, as it mirrors the tactics of many previous critics in
other media. When film theory was still in its infancy, scholars similarly attempted to raise the
medium’s esteem by association with more revered forms, porting the terms of literary and
theatrical criticism to cinematic analysis (hence film scholars’ deployment of theories of narration
and mise-en-scene).

4
entertainment, as technophilic, or as all the above—still tends to outweigh the merits of
such connections, as the very mention of the word “videogame” immediately turns up
aesthetic noses. Even amongst the academic community who have fought to legitimize
such popular media as film and television, videogames are still rarely considered
worthwhile in their own right. Thus many have felt the need to hide videogames by
lumping them into the scholarly catch-all of “new media,” a term that seemingly
references an amalgamation of every product involving a computer in some phase of its
production or reception.
On the bright side, this method does at least recognize that videogames are
something new, and not merely a continuation of a past tradition. But notice the double-
buffer it creates: not only do we gain legitimacy by association with past forms, but we
also avoid addressing videogames directly. This approach buries videogames within the
contemporary wave of computational media, assuming that when we lump all these
objects together, there must be something worth studying. If we combine the Internet,
computer animation, hypertext, virtual reality, *cough* videogames *cough*, and every other
computational product out there, obviously something worthwhile would have to emerge!
In this way, videogames can also be lent credibility by association with interactive art
installations, avant-garde hypertext novels, and other experimental (read: artsy) uses of
computers. Once again, we can address “new media” through the standard traits that
we typically associate with “art”: experimentation, transgression, political investment, and
intellectual gymnastics. Even the videogame industry itself often tries to veil its product
by referring to videogames as “interactive entertainment,” “entertainment software,” or
some other such nonsense in order to create an association with other entertainment
media, or with high technology, and thus distance themselves from the negative
connotations of gameplay.

5
The alternate approach is typically positioned as oppositional to—or at the very
least, distinct and separate from—these medial approaches. Rather than linking
videogames to the traditions of art, we instead place them within the tradition of games,
thus again placing this relatively young medium within a much longer lineage stretching
back for millennia. Rather than associating videogames with cinema, literature, or
painting, this approach instead associates them with sports, cards, and board games—
and especially, the always-esteemed chess. Indeed, a large portion of videogame
scholars who use the “game studies” approach sooner or later employ chess as
example, for while videogames may be for children—a mere trivial pastime—only high-
minded, intelligent adults play chess, a cultured and elitist endeavor associated with
intellect and maturity that still retains a cerebral mystique. Other games which serve a
similar purpose include baseball (a sport with the cultural regard of being “America’s
pastime”), Monopoly (also an old game—born in the 1910s—typically too complicated
for young children), poker (what’s more “grown-up” than poker night?), and especially
go—not only a mature game with a long history, but also one with the cultural cache of
foreign origins (just as in cinema, where the only thing more prestigious and high-minded
than an art film is a foreign art film, the only game with more potential esteem than chess
is its Chinese equivalent). The methodology may be different, but we can see similar
tactics at work here, attempting to escape the societal disesteem of videogames by
association with a more historic, culturally-ingrained, and/or “serious” object than this
relatively young medium. Yet this approach really just creates the same problem in a
different context by lumping videogames into a different super-category. It hardly seems
any less fair to assume videogames to be interchangeable with sports or board games
than with movies or novels, or hypertext or virtual reality, and yet the number of

6
videogame critics and theorists who use an old-fashioned non-digital game to
demonstrate a videogame concept is surprising.
This latter assumption—that videogames can simply be lumped into the bigger
category of games—is especially common due to our choice of terminology: the word
“game” is right there inside the name of the medium. Further, when discussing
videogames we often choose the shortened version of the word, simply calling them
“games,” while videogame players refer to themselves as “gamers” rather than
“videogamers.” As a result, the majority of us—videogame scholars, videogame
designers, videogame players, and even the broad public who may be indifferent to the
medium altogether—assume that we can discuss games and videogames
interchangeably. Yet we most certainly should not.
3

Now don’t allow my cynicism to mislead you. It’s understandable why we discuss
videogames alongside movies and television, or alongside sports and board games, or
alongside hypertext and virtual reality. It would seem rather unreasonable to not discuss
videogames among such objects, as there obviously is something to be learned from
such comparisons (and I myself will certainly indulge such comparisons in the pages that
follow). Bolter and Grusin argue that all new forms of media begin by imitating and
repurposing older forms, integrating those forms into their evolution,
4
and as such, we
certainly ought to consider videogames in relation to previous objects.
5
Further,

3
Consider: we do not question The Sims as a videogame…but is it a game? It has no fixed
objectives, no end-state, no winning or losing—even the game’s designer himself, Will Wright,
shies away from calling it a game, preferring instead to think of the product as a “toy.” Indeed,
one would have a difficult time justifying The Sims as a game—yet it certainly is a videogame. If
such is the case, then the assumption that that we can discuss these two categories
interchangeably appears flawed.
4
Jay David Bolter and Richard Grusin, Remediation: Understanding New Media (Cambridge: MIT
Press, 2001), 14-5.
5
For at least two reasons—to recognize their lineage (e.g., comparisons with cinema help
recognize the manners by which videogames have repurposed the conventions of that medium)

7
associating a new object with that which has come before allows one to make fruitful
observations, to identify similarities and differences. Grouping objects based on a
perceived similarity can allow for a better study of whatever factor joins those objects—
by considering videogames alongside other narrative forms, or other screen media, or
other computational media, or other game types, one can get a better picture of the
commonality that joins those items. And many preceding works addressing videogames
in such a larger context have handily accomplished these feats.
Yet most media scholars would balk at the notion that we ought to simply lump
cinema and television together, or lump novels and stageplays together, or lump music
and dance together. While many of the arts can be fruitfully discussed side-by-side,
each of these forms have enjoyed decades, if not centuries, of medium-specific
scholarship as well. Film scholars, for example, are regularly differentiated from their
television counterparts, despite the common public perception that they’re “kinda the
same thing.” Yet despite the videogame industry’s enormous influence in contemporary
culture, and despite the medium’s significant difference from any other popular
entertainment, many still insist on lumping videogames in with some bigger catch-all,
ignoring the vitality of medium-specific attention. Videogames deserve such attention.
It is for this reason that I choose the singular word “videogame” rather than the
more commonly used two-word derivation “video game.” The latter suggests “video” to
be an adjective, making the medial component (video) a mere descriptor distinguishing
the noun “game.” This assumes video games belong to the genus of games, with the
medial properties only a specifying trait to distinguish them from the board, card, dice,
parlor, and guessing varieties. Alternately, we could assume “video” to be the word of

and to learn from the past (e.g., controversies over videogame content bears striking similarity to
controversies over many other media in their infant stages).

8
emphasis, with “game” functioning as the variant; here we have a type of audio/visual
media—the gamic variety. Either way, the two separate words assume “video games” to
belong to some bigger grouping, either games or electronic media, without an identity of
their own—a subset divided only by the addition of an adjective, a particular genre rather
than a unique object. “Videogame,” on the other hand, by providing a new word,
acknowledges that a new medium has arrived, one that may indeed find origins in past
forms, yet is still deserving of its own title. Would we look at the cinema in the same
manner if we called it “mechanical theater” or “screen literature?” Would we look at
television in the same way if we called it “home cinema” or “visual radio?” And would we
look at sports in the same way if we called them “athletic contests” or “stadium games?”
Why, then, define videogames as just another variety of game, or just another variety of
screen media? Instead, we ought acknowledge them as a unique medium in their own
right.
So my question, then: What is the “art” in a videogame?
I don’t mean this in the lofty, evaluative sense of capital-A “Art,” what some might
clarify as “high art.” Instead, I use the term in a more liberal and objective sense,
derived from the same etymological roots as “artiface” and “artifact” (the Latin ars or
artis, “craft”). What is the nature of a videogame as a deliberately crafted, man-made
thing, a communicative, cultural object? As both a thing to “read” (a text) and a thing to
do (a game), how do videogames convey information and create meaningful
experiences for their players?
To answer this, I will consider the videogame situation, broadly, as the interaction
of four component parts: a PLAYER, participating in a formal SYSTEM that projects a
fictional WORLD, via the mediation of an AVATAR.

9

The videogame itself is comprised of two separate but interrelated parts: its SYSTEM
and its WORLD. This is, essentially, the form and content of the videogame; its world is
the information transmitted and its system the underlying structure of that data. The
PLAYER, external to the game, participates in the game, but only indirectly (since it is a
virtual activity rather than a real one). Thus the AVATAR, internal to the game and part
of its formal interface, serves as a bridge to the game’s content.
These four components form the basis for the four chapters that follow.
I’ll begin in chapter one by looking at the SYSTEM of a videogame—the formal
structure of rules and aesthetics that comprises the thing itself. My first order of
business will be to question what sort of “thing” a videogame is, a question that I’ll
answer with the conceptual model of a game-text, a unique object combining the medial
properties of texts with the spontaneous activity that is a game. The chapter will start by
defining these core principles—what is a text, and what is a game?—and after isolating
the fundamental characteristics of each, I’ll use these features to distinguish several
manners by which looking at the videogame-as-text and the videogame-as-game would
seem to have us looking at entirely different instances. Texts and games are in many

10
ways as apples and oranges—not necessarily opposites, but certainly two different
manners of “thing,” giving us two vital but separate parts of the videogame experience to
examine.
As game-texts, then, videogames combine the discursive properties of texts
(semiotics) with the emergent properties of games (play), and the latter half of this
chapter will consider each of these in turn. To look at the semiotics of videogames, I’ll
delve into their deployment of representation and abstraction, not only in the sounds and
images of their aesthetics, but in the very actions they entail. To examine play, on the
other hand, I’ll look to the systems theory concept of emergence to see what “organized
chaos” can teach us about videogame play. To wrap up the chapter, I’ll then consider
the “emergent discourse” of game-texts—their combination of semiotics and play—in
relation to the computational nature of “new media.”
In chapter two, I’ll examine the WORLD of the videogame, and to do so, I’ll make
a vital methodological leap: using narration as my discursive model. Not all games tell
stories, to be sure. Yet I will argue that every videogame must be “narrated” because,
unlike traditional games, the “game” isn’t really happening. Videogames are virtual
activities, occurring both “somewhere else” and “somewhen else,” apart from the
player’s here and now, and as such, we’ll need focus on a mode of discourse capable of
projecting the player into a non-existent (that is, fictional) time and space. In this regard,
we will discuss narration as “spatiotemporal projection.”
6

I’ll begin chapter two where one left off—with play. Specifically, I’ll look at the
intimate connections between play and fiction, both activities that foster imaginary worlds

6
Some might dispute my use of narration in this regard; perhaps we need a new term for this
mode of discourse, one free of the excess baggage that “narrative” tends to carry. Yet I don’t like
to frivolously coin new terms, and I think narration is an adequate model, so long as we
distinguish it as separate from “storytelling” in a conventional sense. Further, since narratology
already offers a foundation for addressing the projection of a fictional spacetime, I see no reason
to reject that history.

11
of a sort. With this as a foundation, I’ll propose a diegetic model of videogame
discourse, one rooted in narration’s ability to “mentally transport” its audience to an
imaginary space, but forgoing narratology’s traditional emphasis on a chain of events by
instead stressing the creation of a diegetic world. Much of the tension around narrative
models of videogames results, I suspect, from a misuse of terms, a confusion of what it
means to call an object “narrative.” Rather than consider videogames as narratives, we
should think of them as narrators—as world generators. Using that metaphor, I’ll
consider the videogame “narrator” in relation to two different gameplay figures: the
referee and the dungeon master. As referee, I’ll address videogame’s transparent
deployment of rules, considering the ramifications of a game system in which players no
longer need take responsibility for upholding the rules themselves. As dungeon master,
on the other hand, I’ll address videogames within the lineage of participatory storytelling,
adapting the “two world logic” of story and discourse for application to videogame
worlds.
Chapter three will be concerned with the most fundamental and medium-specific
device of a videogame: the AVATAR. In particular, I will probe the avatar’s relationship
to the player (and vice versa), and resultantly, the manners by which the player is
“positioned” in relation to that figure. Here, I’ll propose a dual-positioning model of
videogames, where the player is always positioned in two places at once—inside the
game, as a diegetic participant, as well as outside the game, as a disembodied
observer. I’ll start by examining the history of the term “avatar,” from its origins in Hindu
mythology to its contemporary application, noting the similar dual-positioning that
accompanied the term’s original usage, and following the circuitous route that it has
taken to its current deployment. I’ll then establish this dual-positioning model via a
spectrum approach, with a purely immersive, “internal” experience at one end and a

12
completely removed, “external” one at the other, arguing the player to always be
positioned somewhere in between. In many ways, we might think of these as the ideal
of virtual reality (complete and seamless immersion in a fictional space) and the
spectating situation of cinema (observing a fictional world from the other side of the
screen). Videogames are neither one nor the other, but instead a medium that
combines these two (seemingly opposed) impulses.
To see how the player moves between these poles, I’ll suggest ten categories of
“positioning cues,” distinguishing the textual features that influence the player’s
relationship to the avatar in one way or another. These cues are frequently at odds, as
some features may act to pull the player inside the gamespace while others establish the
player as an outsider, separate from their avatar, resulting in complex, fluid experiences
in which the player modulates between these roles. Finally, I’ll examine the polar
extremes of the positioning spectrum, considering a few games that position their
players at the far ends of the continuum—two highly immersive games that merge the
avatar’s identity with the player’s, and two games that move in the opposite direction,
refusing the player any tangible presence or embodiment inside the videogame. Here,
we’ll see a somewhat surprising similarity between these ends, suggesting that
movement inside or outside the gamespace may not be as different as one might
assume.
Finally, chapter four will expand on this dual-positioning model by examining the
PLAYER—not only in their relationship to an individual avatar, but to the videogame as a
whole. After establishing a diegetic model in chapter two, along with the player’s
relationship to their avatar in three, four will examine the unique role that avatars provide
for videogame narration, a variation that privileges not the narrator, but rather the
narratee. Unlike conventional narration, which has long centered on first- and third-

13
person address, videogames differ by offering second-person narration, suggesting the
audience as the central actant in the game’s fiction. As such, I’ll re-examine the
convoluted and vastly incorrect evolution of the terms of the narrative person
(particularly in screen media), attempting to set the terms straight so that we might have
a better appreciation for the ways in which videogame narration differs from conventional
storytelling. I’ll follow this by examining the dual roles of the avatar in videogame
narration—both as a fictional character and as an ergodic vehicle—to suggest a
psychological-utilitarian model by which an avatar mediates the player’s experience.
I’ll then revisit the dual-positioning of chapter three in a new light, considering the
player’s experience as negotiating several different “frames of embodiment”—the
multiple discursive contexts that frame a diegetic world. Combining both narratological
and ludic models of framing contexts, we’ll see that even the most rudimentary of
videogames entail at least three separate frames that contextualize the game’s content,
and resultantly, three separate embodiments: as audience, as player, and as avatar. I’ll
finish the chapter by reconsidering “immersion” in light of these frames—what it means
to be immersed and the different phenomena that we describe as “immersive”—and
examine why it is that reflexivity (the opposite of immersion) operates differently in
videogames than in the classical narrative model.
As epilogue to the entire dissertation, I’ll briefly touch on the current state of the
medium, as well as the way we position videogames in contemporary culture.

Before jumping into chapter one, allow me to address some minor business on
my manner of presentation. In addition to my use of the word “videogame” rather than
“video game,” there are a few other matters of word choice that I ought to clarify.

14
First, despite my insistence on dividing videogames from the large category of
games as a whole, common parlance tends to confound these terms, as we colloquially
use “games” as shorthand for “videogames.” Such parlance is not without merit—
repeatedly saying “the videogame does this” and “the videogame does that” can become
annoyingly repetitive—and I myself will often use the abridged term “game” to discuss a
videogame whenever my object of reference is clear (such as: “In Pac-Man, the game’s
hero navigates a maze-like world…”). Yet when speaking generically, I will stick to
“videogame” when referring to such an object and “games” when addressing that larger
category, so as not to confuse the two. Thus, when I discuss “games” or a hypothetical
“game,” I am addressing games of all sorts, digital or not. When speaking specifically of
the medium at hand, I will refer clearly to “videogames” in particular.
Secondly, when I discuss “videogames” I am discussing both console games and
computer games alike, and I make no distinction between these two. Some wish to
distinguish these as separate objects (at times differentiating “video games” as those
played on a television and “computer games” as those played on a Mac or PC), yet as
far as I’m concerned, Commodore 64s, Apple IIs, command-line IBMs, and Windows-
based PCs are all merely platforms, just as an Atari 2600, Super Nintendo, Playstation
2, or Xbox 360 is a platform. I see no need to distinguish these objects unless
discussing the specifics of an interface or some other issue of hardware. The gradual
blurring of the distinction between console and computer (consoles utilizing PC
hardware, the growing interchangeability of HDTVs and computer monitors, Microsoft’s
bridge between Windows and Xbox, and the frequency of console and computer titles
being ported between one another, just to name a few examples) evince the growing
uselessness of the console/computer distinction. So unless discussing controller
configurations, playing context, or some other issue that requires attention to the

15
platform at hand, console and computer games will all happily be grouped under the
common heading of “videogames.”
Finally, a word on my use of pronouns. Many writers have wrestled with the
English language’s limited choice of pronouns when discussing a hypothetical individual
of undisclosed gender, and as the unfortunate result of a long patriarchal tradition, we’ve
commonly fallen back on “he” whenever in doubt, neglecting 50% of the world’s
population. When discussing videogames—a medium that has long been dominated by
the assumption of a male player—this slight seems all the more egregious. Some of
those conscious of this gendered bias (both male and female alike) have responded by
leaning in the opposite direction, using “she” instead, yet by my mind, this doesn’t solve
the problem so much as it counterbalances with a new one; it merely neglects the other
half of the population (while additionally drawing attention to itself by opposing the
expectation of a generic “he”). I instead choose a third option: using “they” as a non-
gendered singular. Though perhaps not grammatically ideal, this word choice seems
less rupturing than “she,” while simultaneously managing to not exclude either gender,
and as such, I find this to be the least problematic option—even if it does lead to the
somewhat paradoxical word “themself.”

16

Chapter One: SYSTEM

Dawn of Civilization

It is the year 4000 BC. Your ancestors were nomads. But over the generations
your people have learned the secrets of farming, road-building, and irrigation,
and they are ready to settle down.

Lincoln, your people are Expansionist and Industrious and have recently
mastered Pottery and Masonry.

The people have vested absolute power in you, trusting that you can build a
Civilization to stand the test of time!

I click “OK” with my mouse, and this introductory text box disappears, leaving
behind a lone figure garbed in a powder blue tunic—a Settler—standing in the middle of
a two-dimensional, map-like world. This primitive tribesman appears as a giant in this
environment, standing next to a tiny river snaking its way across the open grassland, a
mere sliver of blue next to this towering figure. To his north stands a stretch of
mountains, barely taller than the Settler himself. To the Settler’s west is a village—a
cluster of thatch-roofed huts which rise just below his waist. To the east, a large cluster
of grapes hangs suspended over a green patch of hills. And due south stretch some
open plains, a single cow standing in their midst.
I direct the Settler to walk east, and the animated figure takes three hurried
strides to his left, stopping in the middle of the tiny village, which promptly disappears.

17
The face of my Science Advisor (a smiling, balding fellow dressed in the tunic and
wreath of an ancient Roman senator) appears in the upper right corner of the screen to
inform me that “The Inuit tribe has taught us Ceremonial Burial.” Ahh, wonderful! Ten
seconds into the game, and my newborn civilization has already learned a new primitive
technology! More importantly, however, the move has expanded my view of the world—
as my units move about the map, exploring the terrain, my worldview expands, reflecting
my increased awareness of the surrounding terrain. I see that my Settler is now
standing on the shore of an ocean (or perhaps, just a sea…I can’t see far enough yet to
tell), and in the water, I can see a single fish jumping above its surface.
Meanwhile, two more figures occupy the space vacated by the Settler, each
dressed in the same shade of powder blue—a Warrior and a Scout. I move the Warrior
south into the plains, and the Scout east into the hills, and the map continues to expand.
With all three units having exhausted their turns, I click the spacebar and 50 years pass
by in an instant—it is now 3950 BC.
Strategically located at the mouth of a river, and nestled comfortably between
three valuable resources—fish, wine, and cattle—my Settler seems positioned in an
ideal spot to begin my new civilization. I instruct him to build our nation’s first city, then
watch as the Settler kneels to the ground, sets down his pack, then vanishes as a tiny
city appears in his place. In this, the year 3950 BC, I, Abraham Lincoln of the
Americans, have founded our fledgling nation, building our capital city of Washington on
the western coast of this strange little continent—a city that begins with a population of
one.
With my Settler gone, I focus attention on my Scout, exploring the terrain to the
east, and it’s not long (about 300 years) before I encounter another city. A scant
distance northeast of Washington—it seems a mere 20 steps by the Scout’s pace—I find

18
the newly established city of Rome. The city is guarded by a Warrior identical to my own
(aside from sporting a green loincloth, rather than powder blue), and as I near the city’s
border, I am addressed by the Roman people’s leader, Julius Caesar: “Caesar of the
Romans has noticed your cute little civilization. I offer you this trade in return for
amicable relations between our cultures,” he declares, with an offer to exchange my
knowledge of Ceremonial Burial for his own knowledge of Masonry. Wanting to make
nice with my continental neighbors (and equally wanting the ability to build city walls), I
accept Caesar’s offer, and our relationship seems off to a harmonious start. Meanwhile,
Washington has just finished work on a Granary, allowing the city’s population to grow
faster, and has already increased its size to four citizens. I’ve founded the city of New
York on the plains east of Washington. My scientists have discovered the secrets of
Alphabet, allowing me to build seafaring units, and I’ve just begun construction on the
Pyramids in Washington…
Eight hours later, I’m in the middle of the Industrial Age. America is now a
Communist state—as my scientists have yet to “discover” Democracy—and I’ve gained
powerful allies thanks to diplomatic alliances with the Iroquois and the French. The
Iroqouis’ discovery of Motorized Transportation has allowed them to build Tanks, making
Hiawatha a peerless military force, and the French’s hearty supply of oil makes my trade
agreement with Joan d’Arc especially valuable. And having built the Sistine Chapel in
Boston and Shakespeare’s Theater in Chicago, America’s cultural influence knows no
equal.
Yet things are not all roses on this revisionist Earth—my relationship with the
Romans has deteriorated. Annoyed with my refusals to share vital resources, Caesar
has declared war on America, and his ally Catherine the Great has been amassing
Russian troops just outside Philadelphia. I have spent the past three turns (three years

19
by the game’s clock) bombarding Rome’s defenses with an Artillery unit and thinning its
military ranks with my Infantry. With a mere two units left, I brace for my last push. I
command the first to move into the city, and the animated little soldier charges forward to
combat the final Rifleman that defends Rome. They exchange fire, and my Infantry falls,
defeated. Yet he’s weakened the Rifleman, and I send forward my last soldier. He fires
three shots, and the Rifleman keels over, allowing my soldier unit to charge into Rome.
My Military Advisor—a lantern-jawed chap in a WWI-era army helmet—informs me, “Sir,
once again our magnificent armies are victorious! We captured Rome and ‘liberated’ 78
gold.” I install a new governor, and Rome becomes an American municipality. Caesar
appears on my screen, asking for a peace treaty, and after he concedes to throw in the
secrets of Medicine, I accept his deal. Catherine agrees to peace as well, and with the
Roman and Russian military forces withdrawing from my borders, I turn my attentions to
“building” the Theory of Evolution in Atlanta.
It may not be entirely logical, but it’s quite a revisionist account of world history
nonetheless, eh?
This is Civilization III, the third installment in Sid Meier’s monumentally
successful strategy game franchise for the PC. Providing a wide range of cultural and
historical touchstones, from Mahatma Ghandi to the Manhattan Project, Civilization
allows its players to rewrite history by building a civilization from the ground up and
conquering the world through diplomatic finesse, cultural supremacy, scientific pre-
eminence, military conquest, or simple world domination.
I doubt I need draw attention to the absurdity in this account. Cities are founded
instantaneously by a single Settler, begin with a mere one citizen, and reach “metropolis”
status at a population of thirteen. A Scout or Warrior can be seen running across the
continent, towering higher than the cities themselves—nearly as tall as the mountains—

20
and a single soldier conquers Rome by defeating the city’s lone defender. “Great
Wonders” like the Theory of Evolution and Universal Suffrage (ideas rather than tangible
objects) get “built” in a city just the same as the Pyramids or the Hanging Gardens.
Technological advancements manifest spontaneously, and include religious
“discoveries” (Ceremonial Burial, Polytheism), civic discoveries (Monarchy, Economics),
and cultural discoveries (Literature, Free Artistry), as well as the more logical scientific
and technical ones (The Wheel, Nuclear Fission). And leaders like Abraham Lincoln and
Julius Caesar live for thousands of years, commanding their nations as supreme leader
all the while—which not only defies any practical life expectancy, but would also seem to
negate the whole point of being able to choose a Democratic government. Add to all this
the strange historical revisions of building the Sistine Chapel in Boston or allowing the
Iroquois nation to develop tanks, and you’ve got quite a nonsensical account of world
history.
And yet, Civilization is a game that purports to depict a variation on the history of
the civilized world, repeatedly stressing its connection to the realities of world politics
and culture, as demonstrated in the product description on the game’s packaging:
Sid Meier’s Civilization III is an exciting journey through time where you are
challenged to create your own version of history and match your wits against the
world’s greatest leaders.

Start off as a world leader in 4000 BC with one settler unit and build a civilization
to stand the test of time. Over the years, explore the world, expand civilizations
by building new cities, developing trade, investing in science research to develop
new technologies and creating your own culture. Using diplomacy, deal with
other civilizations by negotiating peace treaties, trade embargos, military alliance.
Raise armies and conduct wars when talks become useless!
7

The game’s presentation overtly connects itself to the history of the world we know, yet
despite suggesting a variation on world history—a premise rooted in some sense of

7
Civilization III packaging, European release, (Lyon, France: Infogrames Interactive, Inc., 2002).

21
plausible reality—Civilization III seems enormously illogical, and would appear to make
very little sense.
And yet, no one seems to mind.
We might make sense of Civilization in one of two ways. We can encounter it
textually, as a representation of understandable human experience depicted via
abstraction. Rather than conceive of a single infantry unit defeating a single rifleman in
a single attack to conquer all of Rome, I can read these as abstractions of larger
concepts, an entire platoon arduously defeating Rome’s defenses during the course of a
battle—but reduced to a brief and containable representation. In this regard, the combat
animation is a rough equivalent to stating: “…and then an American infantry platoon
entered Rome, defeating a brigade of riflemen.” Rather than a lengthy depiction, it
provides a quick and reductive encapsulation.
Or, we can conceive of the Civilization landscape as a gameboard, one in which
players arrange game tokens and abide by the internal rules of that space. Much as I
might maneuver chess pieces with little concern for their loose representations of
knights, kings, and bishops, I can equally move my “Infantry” into “Rome” while attending
to those signifiers only insofar as they help delineate the rules of the game. In this
regard, the combat animation is much closer to denoting “queen to F3” than any manner
of literal action.
Yet individually, neither of these seem a satisfying account. Dividing its
conceptual meaning (as a text to be read) from its perceptual meaning (as a game to be
played) proves problematic, as Civilization goes to great lengths to blur these two
together. And such is the case in a great many videogames.
Many communicative products merge literal and conceptual meaning, yet
typically, we have no problems negotiating the two and can easily keep them

22
independent. 1952’s High Noon foregrounds the literal story of a sheriff confronting a
band of outlaws, while retaining a conceptual reading as a metaphor for McCarthyism,
without any conflict between the two. Alternately, the literal appearance of Monopoly’s
gamepieces (top hat, thimble, race car) take a backseat to representing the game’s
players, while the king and queen in poker, though understandable as literal royalty,
remain conceptual as indicators of value. Yet this division is not so easy in Civilization.
Compare Civilization to the board game Risk—a game whose context and
mechanics are relatively similar—and we begin to see the leap that videogames take in
this regard. Much like Risk, most of the games’ signifiers operate conceptually—when
we see a lone soldier invading Rome, we’re meant to read that soldier as a whole
platoon. A single citizen indicates a population of thousands, a cow signifies land rich in
cattle, the “tech tree” of 83 advancements represents the entirety of human knowledge
(scientific and otherwise), and Abraham Lincoln embodies a lineage of leaders spanning
several millennia, as well as the player themself. Unlike Risk, however, Civilization also
designs these signifiers to function literally. The single infantry unit is made to be
readable as an individual soldier: every movement and action is animated as a literal
event, whether running across the continent, charging into attack, or melodramatically
dying in defeat, and such units are even given voices and other sound effects to
accompany their actions. Resultantly, the player is encouraged to interpret the soldier
as an autonomous, living entity—as a character—while simultaneously reading it as a
conceptual “game piece”—an indicator of a platoon. Similarly, the game’s leaders are
designed as animated caricatures of their real world counterparts, allowing centuries-
long relations between rival nations (in the game’s conceptual realm) to be read literally
as an ongoing feud between two individual characters. A slippage arises in that
Civilization makes no effort to clearly delineate the conceptual from the literal—to the

23
contrary, the game does its best to blur the distinction between these two, resulting in a
cartoonish account of caricaturized world leaders guiding their tribes through the ages.
Civilization is hardly the only videogame to do so. The Sims provides an
abstracted version of domestic life in which people live for roughly sixty days, character
motivation is reduced to eight wants and needs, and the breadth of human knowledge is
contained by seven specific skills (leading to such oddities as needing a high cooking
skill to become a scientist). The characters themselves even operate in both realms, as
both conceptual signifiers of human behavior and as their own literal breed of creature:
“sims.” In Grand Theft Auto: San Andreas, days last 24 minutes, an entire U.S. state
occupies a scant 17 square miles, and usable objects float in mid-air, waiting to be
picked up. And in Super Mario Bros., mushrooms make Mario double in size, bullets
travel slow enough to be dodged (or jumped on), and pipes operate as fantastic portals
to other lands. All of these abstract some aspect of intelligible experience, yet unlike
High Noon or Monopoly, the distinction between their literal and conceptual reading
remains much more ambiguous.
To make sense of such phenomenon, this chapter will examine videogames as
game-texts, a unique cultural object combining the discursive, textual properties of art
and media with the spontaneous, rule-based operations of gameplay, resulting in a
formal system that is more than merely the sum of its parts. But to do so, we’ll first have
to dissect those individual parts.

Objects and Experiences: Videogames as Game-Texts
What do I mean by the “system” of a videogame? Let’s begin by revisiting the
diagram of the videogame situation that I proposed in my introduction.

24

For now, our concern is with the system and the world, the dual halves of the videogame
itself. Most critics recognize some similar pairing in every medium—this division is
essentially that between form and content, a divide acknowledged in most approaches to
art, media, communications, language, narrative, or design. The terms vary, depending
on context—form vs. content, form vs. function, discourse vs. content, discourse vs.
story, signal vs. referent, signifier vs. signified, fabula vs. syuhjet, and so on—but in each
case, the pairing distinguishes the how from the what: how something is conveyed,
communicated, or presented, and what is conveyed, communicated, or presented. In
the case of a videogame, the world is the what, the system is the how.
We will pursue the world of the videogame—the fictional spacetime offered to the
player as an arena for action—in chapter two. In this chapter, we are instead concerned
with the formal half of the game: the construct that frames that content. “A group of
interacting, interrelated, or interdependent elements which form a complex whole,” a
system is a composite singularity that operates via the interrelationship of mutually

25
dependent parts.
8
When we discuss the system of a videogame, we are discussing its
structure, its arrangement, its aesthetics, and those elements which comprise its formal
composition. This includes the videogame’s rules, play mechanics, and procedures; its
interface, including the controller or input device, the avatar, and the screen itself; the
style and aesthetics of the game, including all sensory output (such as images, sounds,
and haptics); and any other expressive, communicative, and/or presentational devices
(those techniques which help convey information). All these parts operate together in
order to create the world of the videogame—its fictional space and all that exists or
occurs therein. As such, the system is a frame, in both senses of the word: It is a
structural frame, the architecture that holds up and holds together the videogame while
remaining hidden beneath its surface, just as the girders of a building or the skeleton of
a body. But it is also a presentational frame, like the periphery of a painting or the
boundaries of a camera’s lens, presenting the content and facilitating our access to it.
The system is the formal architecture that allows the content to manifest, as well as the
means of access to that content.
What sort of system does a videogame provide? To best answer this, we need
consider the two basic systems in which they find their origins—games (a system of
rules) and “texts” (a system of signifiers)—and examine videogames’ relationship to
both. As I explained in my introduction, my usage of the singular word “videogame”
rather than the more common two-word derivation reflects this very balance: while the
separate words seem to suggest a primary noun qualified by an adjective—that is, that
“video games” are either a specific type of game (the video variety) or a specific type of
audio/visual text (the game variety)—the combined word indicates a unique medium

8
“System,” Dictionary.com, The American Heritage Dictionary of the English Language, Fourth
Edition (Houghton Mifflin Company, 2004), http://dictionary.reference.com/browse/system
(accessed: July 08, 2008).

26
finding its roots in both of these forms. To understand this curious amalgamation, we
need consider the types of systems offered by texts and games, and consider how
videogames operate as both things at once by combining the signifying discourse of
texts with the interactive play of games.
Considering videogames’ existence at the intersection of these two systemic
types might then offer us the best chance to understand that which occurs in Civilization
III. Rather than attempt to understand Civilization through traditional textual discourse,
and rather than attempt to understand it through traditional gameplay, we need instead
determine what exactly it is that we’re looking at. What manner of discourse do
videogames offer? What sort of play do they offer? Which is essentially to ask: what is
a videogame, anyway? To understand this, we need first determine what sort of “thing”
texts and games are.

What is a text?
As a critical term, “text” is fairly enormous, and somewhat ambiguous. Barring
colloquial and everyday usage (referencing schoolbooks, printed wording, instant
messaging, etc.), and limiting ourselves solely to critical applications of the term used to
describe a work of art or media—which is to say, not just “text,” but “a text”—we would
still be hard-pressed to find an agreed-upon definition:
A text is any discourse fixed by writing.
9

—Paul Ricoeur

[A text is] any instance of spoken or written language that could be considered in
isolation as a self-sufficient entity.
10

—William Covino & David Jolliffe

9
Paul Ricoeur, Hermeneutics and the Human Sciences: Essays on Language, Action, and
Interpretation, trans. John B. Thompson (Cambridge University Press, 1981), 145.
10
William A. Covino and David A. Jolliffe, Rhetoric: Concepts, Definitions, Boundaries (Boston:
Allyn and Bacon, 1995), 4.

27
[Texts are] the authentic products of social interaction.
11

—Suzanne Eggins

A text is a semantic unit defined by the textual component.
12

—Michael Halliday

By “text” I shall mean any communication that temporally controls its reception by
the audience. Thus, texts differ from communicative objects such as (non-
narrative) paintings and sculptures, which do not regulate the temporal flow or
spatial direction of an audience’s perception.
13

—Seymour Chatman

I will define “text” as a certain collection of descriptions of an artifact where the
artifact must be one that materializes a symbol system, and the descriptions that
are offered of it must be sanctioned by society. Thus a “text” is more than the
materials of an artifact and more than the symbols materialized.
14

—Edward Branigan

A text is a finite, structured whole comprised of language signs. The finite
ensemble of signs does not mean that the text itself is finite, for its meanings,
effects, functions, and background are not. It only means that there is a first and
last word to be identified; a first and last image of a film; a frame of a painting,
even if those boundaries…are not watertight.
15

—Mieke Bal

Most texts can be described as communicative artefacts.
16

—David Graddol

Even in this small sampling, the definitions range from the very specific (written works
alone) to the very broad (any manner of communication), with a considerable amount of
variation in between. Minimally, we can at least say that all these definitions recognize a
text as an instance of discourse (some implicitly, others explicitly), and thus to get at the
nature of texts would require an understanding of discourse. Unfortunately, that term
can prove just as complicated.

11
Suzanne Eggins, An Introduction to Systemic Functional Linguistics (New York: Continuum,
2005), 2.
12
Michael Halliday, “Text as semantic choice in social contexts,” Linguistic Studies of Text and
Discourse (Collected Works of M.A.K. Halliday), ed. Jonathan J. Webster (New York: Continuum,
2002), 47.
13
Seymour Chatman, Coming to Terms: The Rhetoric of Narrative in Fiction and Film, (Ithaca,
NY: Cornell University Press, 1990), 7.
14
Edward Branigan, Narrative Comprehension and Film (London: Routledge, 1992), 87.
15
Mieke Bal, Narratology: Introduction to the Theory of Narrative (Toronto: University of Toronto
Press, 1997), 5.
16
David Graddol, “What is a Text?,” Media Texts: Authors and Readers, eds. David Graddol and
Oliver Boyd-Barrett (Clevedon, UK: Multilingual Matters, Ltd., 1994), 41.

28
In its strictest definition, “discourse” describes an exchange of words or some
other verbal expression, whether in speech or in writing, and is often used synonymously
with “discussion” or “dialogue.” We use the term to reference such things as “political
discourse” or “the discourse around” a particular subject, and often employ it in relation
to rhetoric. In discussing classical rhetoric (in his appropriately titled A Theory of
Discourse), James Kinneavy recognizes four fundamental components in any discursive
exchange, claiming that all discourse entails “a person who encodes a message, the
signal (language) which carries the message, the reality to which the message refers,
and the decoder (receiver of the message).”
17
By this model, then, discourse occurs
anytime one person communicates information to another via language. Because any
discursive exchange retains the potential for miscommunication, as there is no
guarantee that what the sender puts in will necessarily be equivalent to what the receiver
takes out, the general intention of discourse is to avoid such an occurrence, and to
effectively transmit the intended information by minimizing discrepancies between the
encoding and decoding.
A text, then, has traditionally been considered a manifestation of discourse, thus
standard definitions of the text have generally mirrored those of Ricoeur (“discourse
fixed by writing”) or Covino and Jolliffe (an “instance of spoken or written language that
could be considered in isolation as a self-sufficient entity”). This leaves a mild amount of
flexibility—some use the term very broadly to include any instance of language, others
reduce it to more tangible or deliberate manifestations; some limit the term to written
language, others include spoken language as well. Most commonly, however, the term

17
James Kinneavy, A Theory of Discourse (Englewood Cliffs, NJ: Prentice-Hall, 1971), 19. The
term “reality” poses numerous problems—language can certainly refer to fiction, as well as
subjective and ethereal concepts (love, an idea, etc.)—thus I focus on the broader notion of that
“to which the message refers” rather than the more limiting “reality,“ choosing instead the term
“referent.” This does not, I believe, in any way alter Kinneavy’s arguments.

29
is used by literary critics to address written works—to refer to “the text” is to refer to the
object of literary analysis (note that the majority of the word’s other meanings still revolve
around written language). And thus all was well and good with such terminology…until
literary analysis found its way outside of the literary realm.
Sometime around the middle of the 20
th
century, as critics invested in other
media—particularly film scholars—came to realize that discursive theories could be
fruitfully applied to other art forms, the field of reference for these terms began to
expand. Eager to up the prestige of cinema and attain credibility for a medium that
many dismissed as trivial entertainment, film scholars sought to legitimize their
burgeoning field by association with more traditional forms of art, particularly through the
application of literary theories. “The film as text” became a means toward such
respectability. Stam, Burgoyne, and Flitterman-Lewis explain:
Semioticians preferred to speak not of films but of texts. The concept of text
(etymologically “tissue,” “weave”) tended to emphasize the film not as an
imitation of reality but rather as an artifact, a construct. The term had the
corollary effect of a cultural upgrading for the cinema; in a single terminological
stroke, film-as-text took on all the prestige of literature.
18

As critical theorists began to expand “text” to reference not only verbal exchange, but
any communication, the definition of discourse began to shift with it. And indeed, the
application seems quite apt.
Although Kinneavy limits himself only to verbal discourse as his object of study,
the four components of this discursive model—encoder, signal, referent, and decoder—
can easily be applied to non-verbal communication as well. Cinema and television use
sensory imitation as their “signal” rather than words and sentences—that is, images and
sounds meant to resemble those things to which they refer—yet a film equally entails
both a sender (filmmakers) and receiver (audience), as well as content to which the

18
Robert Stam, Robert Burgoyne, and Sandy Flitterman-Lewis, New Vocabularies in Film
Semiotics (London: Routledge, 1992), 191.

30
signal is meant to refer (that which the film is about). This opened up a whole field of
enunciation theory in media studies, using such linguistic models to address non-
linguistic manners of conveying information—a model of “telling” used to address media
of “showing.” By this expanded definition, discourse can manifest in a wide variety of
media (literature, cinema, painting, music) and through numerous systems of reference
(verbal language, sensory imitation, symbolic systems, conceptual abstraction).
Evincing the fact are the numerous types of media that use discursive terminology in
their very names: narrative films, documentaries, instructional videos, and expressionist
paintings are all defined by a particular mode of discourse.
Scholarly heavyweights ranging from Christian Metz to Jacques Derrida have
since moved toward a more open definition of the text as “any finite, organized discourse
intended to realize communication.”
19
Yet the text metaphor did not stop with art and
media. Following their lead, innumerable theorists of aesthetics, communication, and
culture now use the text metaphor to discuss such diverse objects as paintings,
symphonies, stageplays, magazines, television programs, advertisements, celebrities,
commercial products, clothing, buildings, national landmarks, corporations, and social
institutions. Meanwhile, many literary and rhetorical critics still maintain these words in
their traditional parameters. As a result, there’s no critical consensus as to what exactly
constitutes a text, nor what constitutes discourse, as these terms have been deployed to
reference nearly anything and everything. Any use of these terms, then, requires some
clarification—collectively, we simply can’t seem to agree on their meaning.
Discourse, as I will use the term, is a means of communication, a manner of
organizing, presenting, and/or exchanging information. We typically speak of discourse

19
Ibid, 51. This passage is part of the authors’ summation of Christian Metz’s position on the text
in his Language and Cinema.

31
as that which facilitates content (the information itself), and distinguish these two as
separate from one another. Discourse, then, is the formal component of
communication—the perceptible transmission (images, sounds, etc.) between sender
and receiver. Discursive events might include:
• A film that tells the story of a cowboy
• An essay that argues against tax increases
• A song that expresses romantic longing
• A painting that describes a landscape at daybreak
• A sculpture that suggests the angst of modern life
• Your grandmother explaining how to bake her famous oatmeal cookies.
In each of these cases, information is being conveyed by a sender, to a receiver, via the
formal constraints of a medium. Though departing from Kinneavy’s focus on language,
we see that his model still holds: a discursive event need only entail two participants (a
sender and receiver), a means of transmission (the “signal”), and content that is
separate from that means (the information conveyed).
The means of expression is thus a delivery mechanism for the content, a way of
conveying the intended information between sender and receiver. The combination of
these two components—the content, conveyed via the means of expression—thus
comprises the message of the discourse.
20

20
Hence Marshall McLuhan’s famous claim that “the medium is the message.” The medium—the
means of transmission—is indeed part of the message, as the content cannot be examined

32
As a manifestation of discourse, a text is simply the mediating portion of this
exchange, the content and means of expression manifesting in a tangible instance:

Even when expanding “discourse” beyond written or verbal communication, however,
critics still dispute what objects can be fairly described as a text. Chatman, for example,
limits his application only to “communication that temporally controls its reception by the
audience,” thus discounting such works as painting or sculpture,

while Barry Brummett
goes so far as to insist that any cultural instance loaded with meaning, and thus able to
be “read” into, is fair game, claiming that “Jennifer Lopez, for instance, is certainly a
complex enough artifact to be readable as a text in her own right.”
21

My usage falls somewhere in the middle of these. As I will use the term, texts
are medial objects. They are a manner of organizing information within a fixed object,
and they convey that information through discursive means, allowing for a
communication of ideas between two or more participants. By my usage, films, books,
plays, paintings, photographs, music albums, and television programs, as well as
everyday objects and documents like credit card statements, advertisements, instruction
manuals, and the book report you wrote for your fifth grade English teacher, are all texts.
I find my definition rather sympathetic with Edward Branigan’s:
I will define “text” as a certain collection of descriptions of an artifact where the
artifact must be one that materializes a symbol system, and the descriptions that
are offered of it must be sanctioned by society. Thus a “text” is more than the
materials of an artifact and more than the symbols materialized.
22

separate from its conveyance, making the two inherently tied up in one another. Marshall
McLuhan, Understanding Media: The Extensions of Man (New York: Routledge, 2001), 7.
21
Barry Brummett, Rhetoric in Popular Culture, 2
nd
ed, (Sage Publications, Inc., 2006), 34.
22
Branigan, 87. Italics are Branigan’s.

33

Branigan defines a text as providing information that refers to something outside of itself
(“descriptions”), and that information is organized together in some way, shape, or
fashion (“a collection”). This collection manifests in a tangible object (“materializes”),
and its information is conveyed through discursive means (“a symbol system”) that we
can reasonably assume to be understood by the person receiving that information
(“sanctioned by society”). A text is a fixed collection of discursive statements, operating
as a systemic whole—it is discourse manifesting through a medial object.
23

We can thus isolate three requisites of a text. First, a text is a fixed object. It is a
tangible thing, something to which we can unambiguously refer without any doubt as to
our referent. This tangible quality is often expressed by addressing the text as an
“artifact,” whether a communicative artifact, discursive artifact, rhetorical artifact, or
cultural artifact. Paul Ricoeur describes texts as possessing a fixity that “preserves
discourse and makes it an archive available for individual and collective memory.”
24
In
this way, we rule out conversations, utterances, live performances, and other instances
that cease to exist as soon as they are enacted, leaving no record or referent aside from
memory and thus remaining too unstable to be addressed with the specificity of an
artifact.
Secondly, this object is a discursive instance—by utilizing a symbol system, the
text becomes a product of discourse, communicating ideas between a sender and
receiver through a medium. As I noted earlier, discourse can manifest through any
number of media (literature, cinema, painting, etc.) using any system of reference

23
I should clarify that I do not include the broadest usage of the term which allows for any
culturally loaded object—celebrities, products, institutions, etc. While I do agree that we can
fruitfully apply theories of the text to such items, I think it necessary to remember the
metaphorical nature of such an application. Yes, Jennifer Lopez does communicate something
as a cultural object, but if we pursue that metaphor to its logical end, attempting to address her as
an instance of discourse, the metaphor will eventually break down.
24
Ricoeur, 147.

34
(language, sensory imitation, symbolic systems), so long as it allows for a transmission
of information between these two parties.
Lastly, this symbol system then serves to convey exterior meaning. In one way
or another, a text exists to represent things outside of itself. Branigan defines a text as a
collection of “descriptions”—it serves to depict something else. The letters on the page,
patterns of light on the screen, brush strokes on the canvas, or features carved in stone
are not, in and of themselves, the purpose of the text. Rather, they serve to reference
things and ideas that can be projected onto the world. While discourse can serve
innumerable purposes—expression, description, narration, argument, instruction,
documentation, evaluation, and much more—all exist to address something outside the
discourse. Which is to say: discourse cannot exist simply for its own sake.
We can readily see how videogames qualify as texts. Like any other instance of
media, a videogame is designed by an artist (or team of artists) and encoded with
meaning—in this case, both literally and figuratively—and that object is then received by
an audience. Certainly, the interactive qualities of the object do allow for a flexibility of
content that is uncommon in many other sorts of text, but the range of content—the
boundaries of what can and cannot occur when engaging the text—is indeed fixed.
When I refer to Super Mario Bros., there is a certainty as to what artifact I am discussing,
and that artifact clearly functions within our model of the discursive situation: the
videogame is the mediating object that delivers content between sender (designer) and
receiver (players). The game’s designers envision a fictional world and encode it into
the text, and the player receives that world by interpreting the pixels of light that emanate
from the screen (and although the player may not interpret exactly the same world that
the designers envisioned, information has indeed been transferred from sender to
receiver). Further, Super Mario Bros. displays external meaning in that the game

35
references a fictional world that does not exist within the text—after all, the text merely
provides a long series of ones and zeroes. We understand the videogame’s images by
interpreting them as another world, perceiving characters, places, and actions that exist
beyond the tangibility of the videogame itself and thus exist outside of its materiality.

What is a game?
The term “game” has similarly seen many critical attempts to define just what the
object of study is:
A game is a form of art in which participants, termed players, make decisions in
order to manage resources through game tokens in the pursuit of a goal.
25

—Greg Costikyan

Games are an exercise of voluntary control systems, in which there is a contest
between powers, confined by rules in order to produce a disequilibrial outcome.
26

--Elliot Avedon and Brian Sutton-Smith

To play a game is to engage in activity directed towards bringing about a specific
state of affairs, using only means permitted by rules, where the rules prohibit
more efficient in favour of less efficient means, and where such rules are
accepted just because they make possible such activity.
27

—Bernard Suits

Reduced to its formal essence, a game is an activity among two or more
independent decision-makers seeking to achieve the objectives in some limiting
context. A more conventional definition would say that a game is a context with
rules among adversaries trying to win objectives.
28

—Clark C. Abt

A formal game has a two-fold structure based on ends and means: Ends. It is a
contest to achieve an objective. …Only one of the contenders, be they
individuals or teams, can achieve it, since achieving it ends the game. …Means.
It has an agreed set of equipment and of procedural “rules” by which the
equipment is manipulated to produce a winning situation.
29

—David Parlett

25
Greg Costikyan, “I Have No Words and I Must Design,” Greg Costikyan’s homepage,
http://www.costik.com/nowords.html (accessed July 13, 2007).
26
Elliot Avedon and Brian Sutton-Smith, The Study of Games (New York: John Wiley & Sons,
Inc., 1971), 405.
27
Bernard Suits, Grasshopper: Games, Life, and Utopia (Boston: David R. Godine, 1990), 34.
28
Clark C. Abt, Serious Games (New York: Viking Press, 1970), 6.
29
David Parlett, The Oxford History of Board Games (New York: Oxford University Press, 1999),
1.

36
A game is a closed formal system that subjectively represents a subset of reality.
…Games provide [an] interactive element [that is] crucial in their appeal.
…Conflict arises naturally from the interaction in a game. The player is actively
pursuing some goal. Obstacles prevent him from easily achieving this goal.
…[And] a game is an artiface for providing the psychological experiences of
conflict and danger while excluding their physical realizations.
30

—Chris Crawford

If we’re willing to add some translated accounts defining the activity of “play”—which in
Dutch and French is used synonymously with “games”, as those languages do not offer
separate words for the two concepts—we might add these seminal definitions as well:
[Play/gameplay is] a free activity standing quite consciously outside “ordinary”
life as being “not serious,” but at the same time absorbing the player intensity
and utterly. It is an activity connected with no material interest, and no profit can
be gained by it. It proceeds within its own proper boundaries of time and space
according to fixed rules and in an orderly manner.
31

—Johan Huizinga

[Play/gameplay is:] Free: in which playing is not obligatory…; Separate:
circumscribed within limits of space and time, defined and fixed in advance;
Uncertain: the course of which cannot be determined, nor the result attained
beforehand…;Unproductive: creating neither goods, nor wealth, nor new
elements of any kind…; Governed by rules: under conventions that suspend
ordinary laws, and for the moment establish new legislation, which alone counts;
[and] Make-believe: accompanied by a special awareness of a second reality or
of a free unreality, as against real life.
32

—Roger Caillois

With games, the definitions become significantly more complex, and resultantly, we see
even wider variance among them. However with the term “text,” the variety of definitions
exhibits a genuine disagreement on what objects qualify—whether to include manners of
information beyond written documents, whether to include ephemeral discourses beyond

30
Chris Crawford, The Art of Computer Game Design, Washington State University,
http://www.vancouver.wsu.edu/fac/peabody/game-book/Chapter1.html (accessed July 13, 2007).
31
Johan Huizinga, Homo Ludens: A Study of the Play Element in Culture (Boston: Beacon Press,
1955), 13. In this passage, Huizinga’s referent is officially translated as “play,” which by our
distinction in English would be significantly different than a game, however Huizinga’s native
Dutch uses the same word for both concepts; “play” and “gameplay” translate equally, and as
such, Huizinga makes little differentiation between these two concepts. Considering Huizinga’s
mention here of “fixed rules,” the primary characteristic that separates a game from generalized
play, it seems fair to use his definition for a discussion of games.
32
Roger Caillois, Man, Play, and Games, trans. Meyer Barash (Champaign: University of Illinois
Press, 2001), 9-10. Just as with Huizinga’s translation, Caillois’ native French also makes no
distinction between “play” and “games,” therefore while the official translation uses “play,” Caillois’
inclusion of being “governed by rules” suggests a symmetry with the English “game.”

37
concrete objects, etc. Yet with “game,” there seems a fairly unswerving consensus as to
what items, in the main, fall within this category. Generally speaking, we all know a
game when we see one. The only dispute is in how to define the term so as to know
where we draw the boundaries to include or exclude edge cases (such as riddles,
puzzles, or gambling). As with “text,” we can dismiss some colloquial and everyday
uses of the term (such as when we call goofing off “playing games”), but when deploying
the word as a formal term for a certain type of activity, we normally seem to agree on
what things are games and what things are not; the only question here is in what
characteristics make them so.
A wide range of scholars and critics have taken stabs at an authoritative
definition, but fortunately, Katie Salen and Eric Zimmerman have done much of the
comparative legwork for us. In their comprehensive textbook on game design, Rules of
Play, Salen and Zimmerman provide an insightful comparison of the eight definitions
listed above, handily distinguishing the commonalities and differences among them, and
cobbling together a nice, compact definition out of the effort:
A game is a system in which players engage in an artificial conflict, defined by
rules, that results in a quantifiable outcome.
33

They also go on to elaborate that:
Players interact with the system of a game in order to experience the play of the
game [and] …rules provide the structure out of which play emerges.
34

By collating these previous accounts, Salen and Zimmerman arrive at a moderately
democratic definition that is at once simplistic, specific, and all-encompassing. A further
benefit is that the object of Salen and Zimmerman’s book is the super-category of
games; unlike many contemporary definitions that are particularly biased toward

33
Katie Salen and Eric Zimmerman, Rules of Play: Game Design Fundamentals (Cambridge: MIT
Press, 2004), 80.
34
Ibid, 80.

38
application to digital games, Salen and Zimmerman demonstrate little predisposition
toward any particular subcategory.
This definition will handily suffice, yet I do think it worthwhile to briefly touch on
the fundamentals that seem requisite in these (and most) accounts of what comprises a
game—despite Salen and Zimmerman having already completed much of the task. By
my count, there are four critical elements that most definitions reference; the exact
phrasing often varies, but most attempts to qualify the term “game” entail these four
components in one manner or another.
35

First and foremost, a game is an activity. Unlike a text, which can be examined
outside of any audience, a game relies on activity—a participant’s engagement—for its
very existence. Games require active participation on the part of their audience rather
than the more passive engagement of watching, listening, reading, observing, or
otherwise consuming that is required of other instances of art and culture. Some
address games in just these terms, as an activity,
36
while others get at this notion by
specifying the game as involving effort.
37
Yet this characteristic is most often conveyed
by highlighting the role of the person performing the activity. Whether described as a
player,
38
contender,
39
decision-maker,
40
or even a “power,”
41
this role is inevitably
described in active terms, foregrounding the participatory nature of the game (as
opposed to a more passive description, such as “receiver” or “audience”). With the

35
Salen and Zimmerman’s analysis distinguishes six components rather than four, yet the
differentiation is fairly negligible; they, for example, differentiate “system” and “rules” as separate
concepts, while I join these together to acknowledge the two as part-and-parcel to one another.
But ultimately, it’s an arbitrary matter of semantics—both versions make similar points, I merely
wish to emphasize different qualities for my purposes.
36
Huizinga, 13; Suits, 34.
37
Jesper Juul, Half-Real: Video Games Between Real Rules and Fictional Worlds (Cambridge:
The MIT Press, 2005), 36.
38
Salen and Zimmerman, Rules of Play, 80.
39
Parlett, 1.
40
Abt, 6.
41
Avedon and Sutton-Smith, 405.

39
coming of the computational era (along with its component buzzwords) more recent
accounts discuss not merely the activity of games, but their interactivity.
42
In any case,
all definitions necessitate—whether deliberately or in more implicit terms—a distinctly
active rather than passive experience. Games involve a “doing” rather than a receiving.
Yet a game is not just any activity; it is an activity defined by a system of rules,
the principal feature that divides games from the larger category of “play.” Children play
house, play with toys, and play pretend, but while all these activities belong in the realm
of play, few would describe them as gameplay, primarily because they are not bound by
rules in the way that a game is. Some draw attention to the rules as the means by which
players engage the game,
43
others as a deliberate choice of arbitrary inefficiency,
44
and
still others as the “limiting context” which structures or confines the activity of the
game.
45
Yet regardless of the exact wording, each account seems to agree that the
rules “bound” play in one way or another, and it is this structuring—the provision of a
formalized system that delimits and organizes the activity—that most distinguishes
gameplay from the more freeform nature of generalized play. Fundamentally, the rules
are what define the game itself.
Related is the third commonality in defining games—that games create a goal-
oriented conflict. This is what the system of rules governs, and while the rules provide
the form of the game, we might consider this conflict as the content of gameplay. The
rules provide a goal (or goals) for the player, but also provide obstacles which stand
between the player and the achievement of that goal—whether opposing players or the
system of the game itself—and thus, conflict results. A wide variety of terms have been

42
For example: Crawford
43
Parlett, 1.
44
Suits, 39.
45
Abt, 6; Huizinga, 13; Avedon and Sutton-Smith, 405.

40
used to describe the player’s goal—objective,
46
ends,
47
outcome,
48
or the “bringing about
[of] a specific state of affairs,”
49
—and some choose to describe the conflict more
specifically as a contest
50
or a competition, but regardless of semantics, most agree that
games must encourage the player toward some achievement, while simultaneously
obstructing the accomplishment of that achievement.
The fourth and final characteristic is perhaps the most complicated. While
activity defines what the thing is, system of rules defines its form, and goal-oriented
conflict defines its content, the final component is much more ambiguous in terms of why
it is requisite of a definition—and yet, few would doubt that it is indeed vitally important.
Its ambiguous nature is reflected in the wide variety of manners in which it is addressed,
yet in one way or another, it is always essential—by their nature, games are artificial. A
game is “not real” in that meaning within the game does not translate to the same
meaning outside the game, and games do not have any substantial ramifications outside
of their context. Some reference the voluntary nature of games,
51
others the safety of
games,
52
and still others games’ status as a “subset of reality,”
53
all indicating that games
remain outside the realm of consequence—that they do not provide a real conflict. But
this unreality is best described by those not concerned specifically with games, but
rather the broader category of play. These critics address play (and by extension,
gameplay) as existing “outside ordinary life” and bearing no relation to material interest
or profit,
54
as a type of make-believe or unreality,
55
or as being “untrue” and possessing

46
Abt, 6
47
Parlett, 1.
48
Avedon and Sutton-Smith, 405; Juul, Half-Real, 36.
49
Suits, 34.
50
Parlett, 1; Avedon and Sutton-Smith, 405.
51
Avedon and Sutton-Smith, 405; Suits, 34.
52
Crawford.
53
Ibid.
54
Johann Huizinga, 13.

41
“non-existent” meaning,
56
all indicating the artificial nature of play. We might find some
dispute amongst these details—for example, whether gambling or professional sports
qualify—yet there seems a broad agreement that in one way or another, games exist as
separate from “the real world.”
Drawing these four characteristics together, then, we find that: a game is a
system of rules which allows players to act within an artificial, goal-oriented conflict—and
videogames obviously fit this definition as well. When I engage Super Mario Bros., it
functions just as readily as a game as it does a text. It is certainly an activity, as I must
actively participate in the game via my control of Mario in order for the game to proceed;
it is a system of rules in that I do not have unrestricted freedom to do anything, but am
instead bound by the affordances of the game; it a goal-oriented conflict, requiring that I
reach the end of its lengthy gauntlet while unleashing a plethora of obstacles in my path;
and it is artificial in that, outside of the game context, there is no Mario, there is no
princess to save, and there is no consequence to whether I win or lose.
Thus to make sense of the videogame as a system, we must consider
videogames both as texts and as games, yet this endeavor is not as simple as the
suggestion may seem. There are numerous critical differences between these two
concepts—texts and games are in many ways as apples and oranges, two mildly related
but inherently separate phenomena. Consider: three components are requisite of a text:
it is a fixed object, it is a discursive instance, and it provides external meaning. Four
components are required of a game: it is an activity, it is bound by a system of rules, it
provides a goal-oriented conflict, and it is ultimately artificial. Using these few
fundamental characteristics as our basis, we can find at least seven ways in which

55
Roger Caillois, 9-10.
56
Gregory Bateson, “A Theory of Play and Fantasy,” Steps to an Ecology of Mind: Collected
Essays in Anthropology, Psychiatry, Evolution, and Epistemology (Chicago: University of Chicago
Press, 2000), 183.

42
videogames-as-texts and videogames-as-games would have us looking at two distinctly
different things—beginning with the very question of what sort of “thing” they are.

Videogames-as-Texts vs. Videogames-as-Games
As a text, videogames are textual—they are tangible, material objects that we
can point to. By examining the videogame-as-text, we are discussing the videogame as
a discursive artifact, an object that can be referenced with certainty as to the thing in
question. Like words on a page or images on celluloid, we can point to the code on a
disc (or cartridge, or hard drive, or server) and know our object of reference. Further,
texts are fixed and unchanging. The materiality of texts assures us that they cannot
change—when a painter finishes a canvas, an author completes a novel, or a filmmaker
submits their final cut, the text is complete, solidified for future examination. One can
perform a “textual analysis” of George Melies A Trip to the Moon (filmed in 1902),
Chaucer’s The Canterbury Tales (written circa 1380) or even cave paintings over 10,000
years old because of this very permanence—in the time since their creation, we assume
them to have not changed (aside from the physical wear of the ages). As texts,
videogames become cultural relics, available for future reference and able to be
discussed in specific terms due to their tangible nature.
When I discuss Civilization III, any reader familiar with that videogame could
assume with relative certainty that the object they know as Civilization III is the same
object to which I refer, with no discrepancy between the two; as a fixed and definitive
object, Civ III does not change (gameplay may change—we will address this shortly—
however the possibilities defined by the game’s code are inherent to the text, and thus
fixed). It is for this reason that we are able to discuss texts (of any sort) with others,
even without experiencing their reception together. Texts carry with them an assurance

43
of consistency, a guarantee that the thing in question remains constant, even when the
moment of reception—or even the instance of reception—differs.
To alter the object in any way, then, would be to create a new text—this is why
revised editions, translations, director’s cuts, and remakes are cited as different texts
than their originals. Similarly, expansion packs, downloadable content, mods, expanded
editions, or any other modification that alters the original videogame must be considered
as separate texts. There’s good reason that such objects are typically given their own
titles (such as the expansion Civilization III: Play the World)—they introduce a new
“thing” to be considered, something distinct from the original (even if this new text
requires the original to operate, and thus cannot be “read” on its own). And if we are to
be completely honest, even updates and patches need be delineated from the original,
as they offer a (slightly) different object—hence the usual delineation of “v1.1.”
Yet games are much more fluid and amorphous than texts. After all, where do
we locate “baseball?” Can one go and buy a copy of “baseball?” We could point to a
printed rulebook that details the rules of baseball, which would indeed be a text, yet that
would not be baseball; that object would be a separate discursive act that merely
describes the game, much like sheet music may describe a symphony, but is not the
symphony itself. After all, do we need a rulebook to play baseball? Similarly, we could
point to a bat and ball, gloves and bases, and even a field for playing, yet these things
would not be the game either—they’re merely equipment, the things necessary for the
act of playing, the “hardware,” so to speak. If the equipment defined the game, how
would we explain the vast differences amongst chess sets? Surely an oak chess table
with hand-carved marble pieces is a different object than a colorful cardboard
chessboard with a plastic Mickey Mouse as the king—and yet, both would be offering
the same game: chess.

44
In a similar vein, the years have seen numerous renditions of Pac-Man and Tetris
ported to a variety of different platforms, while contemporary games like Tomb Raider:
Legend and Tom Clancy’s Splinter Cell have been released for multiple consoles
simultaneously. As specific objects, we’d have to address each port of these games as
separate texts, however we would generally think of them as offering the same game. In
many cases, we’d even think of sequels that merely update game franchises—Mortal
Kombat 3 as opposed to Mortal Kombat II—as offering “the same game,” though these
are clearly different texts, much like we think of professional baseball as offering the
same game every season, despite yearly updates and revisions to the MLB rulebook.
So while the original Civilization III, the expansion Civilization III: Play the World, the
Macintosh port of Civilization III, and the sequel Civilization IV are all separate texts, we
might recognize all of these as offering “the same game”: Civilization.
As activities, games exist as an amorphous idea as much as anything. They are
not things that we read, watch, look at, or listen to; they are instead experiences,
something that we do, and as such exist in the playing rather than in any object. Salen
and Zimmerman clarify that in games, the equipment and material components of play,
to the extent that they “embody the game rules and enable them to be followed,” are
“mere extensions of the rules,” and it is the active playing out of those rules that
comprise the game.
57
Texts, on the other hand, are much more patently intertwined with
their material nature. David Graddol notes that
A great deal flows from [texts’] status as artefacts. They are commodities which
can enter social and economic relations: they can be advertised, sold for profit,
presented as gifts, owned as property. …[Further,] artefacts are the product of a
technology and their material form in part reflects the nature of that technology.
58

57
Katie Salen and Eric Zimmerman, The Game Design Reader (Cambridge: The MIT Press,
2006), 9.
58
Graddol, 41.

45
Texts are a physical product of their medium, and as a result, inherently reflect the
technology that is available to produce them. Just as the style and purpose of writing
has necessarily reflected the technology available to it—from letters carved in stone to
script inked on parchment to type duplicated by printing press—so too changes in
videogame technology have drastically changed the possibilities and expectations of the
medium. The elaborate stories, graphical complexity, and spatial environments of the
Half-Life, Grand Theft Auto, or Final Fantasy franchises could hardly be accomplished in
the era of Pong and Space Invaders, when graphical videogames were limited to simple
and repetitive actions and any attempts at narrative elaboration was forced into text-
based adventures.
59
As objects, videogames are continually defined by the technology
that produces them (the most recognizable distinction between Civilization II and III, after
all, is in the technological leaps between them).
This materiality, however, can often be elusive, as the object’s boundaries are
not always self-evident. This is especially the case between software and hardware.
Where do we draw the boundaries of the text when discussing the 1983 Star Wars
arcade game? Using 3-D color vector graphics to simulate the POV of Luke Skywalker
flying his X-Wing fighter, Star Wars featured an elaborately designed, sit-down arcade
cabinet built to mimic a cockpit, with an elaborate steering mechanism that allowed the
tactile sensation of controlling an aircraft while staring out the cockpit window of the
screen. In such a case, would the game’s code be enough to constitute the “text” of the
videogame? Or would we need look at the entire cabinet as the object in question? In
this case, we’d probably say the latter. However when we consider home consoles, the
hardware involved remains just as relevant, though the console exists as a separate

59
Further, while such abstract and arcade-style gameplay then died-off in the 1990s, thanks to
the expectations that came with the more advanced technology available, such games have since
been revived with the advent of Xbox Live Arcade and Nintendo’s Virtual Console, not to mention
cellphones, iPods, and other technologies that require simplified gameplay.

46
object which accommodates a great many texts. We could hardly discuss Nintendogs or
Puzzle Quest without addressing the touchscreen of the Nintendo DS, yet we purchase
those cartridges separate from the platform. Should we now isolate the software from
the hardware? Considering Puzzle Quest as a game, such questions are moot—the
hardware and software both become mere “equipment” that allows the activity to occur,
as the game can only be located in the act of playing. Yet as tangible objects that are
the products of a mediating technology, such questions are vitally relevant.
This provides us our first distinction between the videogame-as-text and the
videogame-as-game:
Distinction #1: As texts, videogames are fixed, tangible objects, while as
games, they are activities, elusive happenings that eschew materiality.

Texts are a physical realization of the data they convey, made possible by the medium
that created them—they are delivery mechanisms for the information communicated.

Encoded and decoded via a semiotic system, a videogame is a manifestation of its
medium—of the means of transmission—just as a film is a manifestation of cinema or a
novel is a manifestation of literature.
In this regard, videogames convey exterior meaning. Texts reflect ideas and
concepts that are not inherent to the text, but rather suggested to exist outside of it—
referents that we do not locate within the object itself, but are rather representations that
we understand independent of the text. Whether descriptive, argumentative, narrative,
expressive, instructive, informational, or utilizing any other manner of discourse, texts, as

47
discursive objects, refer to other things, providing information that can be interpreted in
the mind of the receiver and projected onto the world outside the text.
60

Yet can we compare games in this regard? Games do convey something to their
players, so there is some system of meaning at work within games. But can games be
discussed in the context of a communication in the same way that we discuss stories,
documents, works of art, and language in general? And if games are discursive, then
what sort of discourse do they offer? How would we apply this model to baseball? Or
chess? Or solitaire?
We can at the very least say that games do involve a conveyance of
information—when a baseball is hit deep into the outfield, when a queen captures a
rook, and when a solitaire player places a three of diamonds over a four of clubs, we
understand what those actions mean within their respective contexts, much as when a
novel tells us “they lived happily ever after” or a film shows us a shot/reverse shot, we
understand what those things signify. Though markedly different in many respects, texts
and games do have some relationship to one another, as both involve a conveyance of
meaning.

60
It’s telling that most abstract arts—those which need not reference anything beyond
themselves—are performative rather than textual. Instrumental music, for example, is abstract in
and of itself. Musical texts are referential, pointing to the music itself (sheet music is instructional,
describing how the music is played, while a recording documents the playing of the music), yet
the music in its own right does not represent. Non-interpretative dance can be described
similarly. Comparatively, even the abstract works of modern art—deliberate transgressions
against the classical ways of representation, deliberate attempts to break from the textual mold—
still extend outside of themselves via their expressive nature, indicating something beyond the
physicality of the text (after all, many of these works were products of expressionism). But note
that such works, in their attempts to move outside representation, often moved into experience
and performance as a means of such transgression. The process of “action painting” is closely
allied with abstract expressionists like Jackson Pollack, who broke away from representation
through an emphasis on the physical act of painting itself, focusing on how the painting is made
rather than what it is meant to depict, and associating his craft more with the jazz musicians of
the era than the painters who had preceded him (Pollack often listened to jazz during the process
as inspiration). Yet despite transgressing traditional representation, even these convey exterior
meaning via their expressive qualities.

48
But games also manifest from another sort of system—a system of rules.
Games are rule-sets that provide for a player experience, that bind and delineate the act
of gameplay, and it is here that we would typically locate a game, as the rules are the
very thing that defines a game. How would one go about defining “baseball?” Most
likely, by providing some abbreviated form of the rules. Yet the rules alone are difficult
to consider as the game itself. Consider the way in which Salen and Zimmerman
describe the game of soccer:
In a game of Soccer…the players, the ball, the goal nets, the playing field, are all
individual elements. When a game of Soccer begins these elements gain
specific relationships to each other within the larger system of the game. Each
player, for example, plays in a certain position on one of two teams. Different
player positions have roles that interrelate, both within the system that constitutes
a single team…and within the system that constitutes the relationship between
teams... The complex whole formed by all of these relationships within a system
comprises the game of Soccer.
61

Take particular note of that last sentence: The whole formed by all these relationships
comprises the game of Soccer. The game of soccer is a system of rules, and yet the
actual “thing”—what we would equate with the “discursive instance”—is that which
manifests from an active playing out of those rules. In other words: while a text is a
discursive instance that arises from a particular semiotic system, a game is, in itself, a
semiotic system. This is ultimately what discourse is: a system of signification, a system
that provides terms through which elements gain meaning and thus allows for
information to be exchanged in a manner that is understandable to all parties involved.
As games, videogames create a system of interior meaning within which players are
able to perform the activity of gameplay. The most practical comparison to texts would
be to compare games not to the objects themselves, but rather to the discursive system
that allows for them—to their respective medium. Just as a filmmaker interacts with the
medium of cinema—having to abide by the rules and boundaries which that medium

61
Salen & Zimmerman, Rules of Play, 50.

49
affords—in order to create a text, so too the player interacts with the game’s system of
rules—having to abide by the rules and boundaries which that system affords—in order
to create the playing of the game.

The answer to the question “are games discursive?” is, “yes…sort of.” “Discourse” may
not be the most appropriate term to describe this system of meaning, but there is a
communicative exchange occurring, one which makes a reasonable comparison. But
rather than functioning as an object through which a sender transmits an idea to a
receiver, games create their own context within which meaning is transmitted.
The content of such “discourse,” then, is quite the opposite of textual content.
The content of a game is the goal-oriented conflict that the game itself creates, a conflict
that does not exist outside the game’s system of rules—this is why games are, by their
very nature, artificial. In its most minimalist form, a game conveys not exterior, but
rather interior meaning—the meaning created within its own context. The rules of the
game—the formal properties that bound and define it—exist solely for the experience
that they allow, and need not refer to anything beyond the game itself. It is for this
reason that Roger Caillois notes that “the game has no other but an intrinsic meaning.
That is why its rules are imperative and absolute, beyond discussion. There is no
reason for their being as they are, rather than otherwise.”
62

Tic-tac-toe does not refer to anything beyond itself. Hockey does not refer to
anything beyond itself. Checkers does not refer to anything beyond itself. Games can

62
Caillois, 7.

50
exist for their own sake, and as such, can be wholly abstract. Gameplay differs from
textual discourse, fundamentally, in that it need not signify anything outside its own
system of rules—it can, of course, yet it is not a necessity. Books don’t exist simply for
the act of reading or writing—they exist for what it is that that the reading conveys. Films
don’t exist for the mere act of filming or viewing—they exist for what the viewing
conveys. Yet games may exist solely for the act of playing—the ultimate purpose of the
game is the experience of play, the experience of engaging the game’s rule-based
structure.
Even when games do include representative elements, the references
themselves are often not the point, but are instead used to achieve an intuitive
understanding of gameplay. Home plate in baseball, for example, has a representative
connection to the concept of “home,” yet the game is not meant to in any way comment
on domesticity. Instead, the representation helps explain the role of that location—the
runner begins by making a journey away from home, and is not “safe” until they return
there. Similarly, chess uses kings, queens, and pawns—roles that can be found outside
the game—yet the most immediate purpose of these references is to distinguish the
importance and functionality of each piece. If we were to posit that chess’ fundamental
purpose is to depict the functions of a royal court, however, we’d be hard-pressed to
maintain that comparison: Why does the bishop move diagonally? Why is the queen an
offensive powerhouse? Why can pawns be exchanged for a captured queen? Such
events can remain sensible within the game context because of the extremely tenuous
connection between the game’s rules and its representative components. It is for this
reason that the vast number of themed chess sets that depict trademarked characters
(Looney Tunes, Star Wars, The Simpsons) still provide the same game—chess—despite
their difference in representation. The same could be said for the many, many different

51
editions of Monopoly. With no need for such exterior investment, games feel no
obligation to represent concepts from the outside world. They can, but they need not.
By focusing on the experiential qualities of play, a game needs not provide explicable
content as the experience itself may serve as the content. A text, on the other hand,
cannot exist without exterior meaning; without some external signification, the object
would cease to be a text (If I paint my garage door, no one would call it a “text” because
it fails to depict or express anything—it’s just a white door).
63

This brings us to our second distinction:
Distinction #2: As texts, videogames are the product of a semiotic system,
while as games, they create their own system of meaning.

The thing in question—the videogame itself—has moved in the diagram of game
discourse. In textual discourse, the text is the discursive instance. The author creates it
and encodes it with meaning, working within the boundaries of the medium, and the text
is what carries the content. Yet in game discourse, the game essentially takes the place
of the author, and we have a new element as the discursive instance—the individual
match. If we view Casablanca ten times, every viewing involves the same discursive
instance—the film Casablanca. However if we play baseball ten times, every playing will
provide a different discursive instance—that particular game. Because we typically use
the word “game” to reference both the overall activity (the game of baseball) as well as
an individual match (a game of baseball), we often confound these two things. We
should not.

63
Keep in mind: these statements are made regarding only the most bare-bones fundamentals of
a text and game, considering theoretically minimalist examples. In truth, nearly every game
makes reference to something outside of itself, and nearly every text has some pleasure in and of
itself. My point is that, theoretically, they need not, and if looking only at the fundamental purpose
of their design, we can generalize that the point of a text is to go outside of itself, while the point
of a game is its own experience.

52
If we wanted to discuss the “author” of gin rummy, we would look to Elwood T.
Baker, who created the game in 1909. Yet when referring to an actual match of gin
rummy, would we credit that match to Elwood T. Baker? No, we’d credit the players—
they are the ones who create the instance. Yet they do so by working within the bounds
of the system of rules, so we need give some authorial credit to the game as well. It is
through the interaction between Mr. Baker’s creation and the players’ actions that the
match takes place, thus we need distinguish the game gin rummy from a particular
match of gin rummy. Similarly, a particular round of Tetris is a different “thing” from the
overall game of Tetris and we need distinguish these two. Are we discussing a game of
Tetris or the game of Tetris? While as a text, Tetris may be authored by game designer
Alexey Pajitnov, the actual playing—an instance of game discourse—is “authored” by
the player operating within the system of rules.
Note the manner by which we deploy the verbial forms of media: painters paint
paintings, sculptors sculp sculptures, writers write writings, narrators narrate narratives,
builders build buildings, and singers sing songs. In each case, the “enacting” of the
medium is an act of production—the creation of the thing itself—and not surprisingly, our
examination of those objects usually focus on the work put into the design, on the
intentions and actions of the artist, the originator of the discourse. But gamers do not
create games. “To game” is not to design. Simply as a matter of semantics, to enact a
game is not an act of production, but rather reception—the medium is enacted by its
audience, and this shift is telling. In game discourse, our attention moves from the
traditional focus on the sender to the receiver—the player. To make a rather common
sense distinction, we could say that texts are “written,” but games are played.
Because texts are the products of a semiotic system, they are most invested in
the presenting of information. They provide data, and are thus concerned with

53
description, depiction, explanation, expression, instruction, classification, documentation,
or some other manner of discourse—all actions that flow from the figure who creates the
text. Because that discourse manifests in a fixed object, there’s no way for the audience
to send a signal back to the artist, forcing a one-way communication and leaving the
audience in a position of passivity. With all the information flowing in one direction, our
natural tendency is to consider that information in relation to its source rather than its
destination, as the receiver has little control over the discourse, save for interpretive
liberties. As such, it makes sense that we look at texts from the perspective of whatever
figure we identify as providing the expression, whether literally from an author or artist,
or more figuratively from a fictional narrator or a character’s point of view, rather than
from the perspective of the audience.
Yet information does not only flow in one direction in a game. While texts are
presentational, games are interactive, allowing information to flow both ways by
providing for both input and output. The audience is now able to take an active role
rather than a passive one, and resultantly, one cannot ignore the player. As such, game
discourse is invested in the playing of the game just as much—if not moreso—as the
authoring. Consider: we associate a novel with its author (rather than the readers who
enjoy it), a painting with its painter (rather than the patrons of a museum), and a film with
its director and actors (rather than the audience in a theater). Yet with whom do we
associate a sport? Rarely the “inventor” of the sport, the league that regulates it, nor the
referees who enforce its rules; instead we associate it with the athletes who play it.
Further, consider how often we refer to the prowess of an athlete in similar terms as we
might discuss the artistry of an author, filmmaker, or musician—both are comments on
that figure’s ability to work within the boundaries of a system of rules, to achieve more
within those limitations than a lesser figure could achieve.

54
Generally speaking, we do not focus on the person who “authors” or “designs” a
game because we don’t perceive a game as a communication between the inventor of
the game and the player. Although we can fruitfully apply the “text” metaphor to games
in the same manner that others apply it to celebrities or institutions (“soccer as a cultural
text”), and although there are plenty of instances of games designed to express
something in the same manner that a book or painting might, this is a separate system
of meaning than that occurring within gameplay. For while it would be possible to
discuss the discursive act of designing a game, the system of meaning in gameplay is a
separate matter entirely, and it is this system to which I refer when discussing “game
discourse.”
This gives us a third distinction:
Distinction #3: As texts, videogames are presentational, allowing a one-way
flow of information sourced in the author, while as games, they are
interactive, allowing a two-way flow of data between the player and the
game itself.

Or, to put it more plainly:
Distinction #3 (version 2): As texts, videogames are “written,” while as
games, they are played.

Interactivity necessitates a different model of the audience than presentational
discourse. While these simplified models of textual and gamic discourse provide an
appropriate visualization of discourse in its most simplified form—between the bare
minimum of participants—the comparison becomes more complicated when considering
a larger audience.
Textual discourse is inherently geared toward a mass audience. With a fixed
discursive object, and an audience unable to provide substantive input, each audience
member ends up receiving identical information. Not only do we receive the same text

55
every time we view Casablanca, but every other member of the audience also receives
the same text every time they view the film.

Game discourse differs, however, in that it addresses an individualized player. Though
each player interacts with the same system of rules, every player experiences a different
game. Single-player games provide the most obvious example—if each match is co-
authored by the player, then each match will be tailored to that single member of the
audience, based on the decisions they make and the actions they take.

Yet even within the same match, games still address every player individually. When
two players engage in our hypothetical round of gin rummy, those players are given
significantly different information—they’re provided different knowledge (the cards in

56
their own hand), and they are provided different resolutions: a positive ending (a win) vs.
a negative one. Even on the same team, players still experience a different discursive
instance—each player on a football or baseball team has different concerns, and thus
experiences a different game.

Here, all the players are interacting within the same match, yet the “content”—the
individual experience—differs for every player.
When eight players engage in a multiplayer round of Halo, those eight players
are all engaging the same text—the videogame Halo. Yet they each experience a
different game, as they all receive different information—what occurs on each player’s
screen differs, as each player is addressed individually by providing the point-of-view of
their particular avatar. Because texts must address a mass audience, we could hardly
imagine a film or novel in which every audience member is positioned behind the eyes of
a different character. Yet this is exactly what multiplayer games are able to achieve.
Distinction #4: As texts, videogames address a mass audience, while as
games, they address individual players.

If we are to critically discuss a game, then, where are we to locate it? We know where to
find a text—because texts are tangible objects, they are locatable. If one wishes to
examine the Mona Lisa or Citizen Kane, there’s no question as to where to look. And as

57
fixed texts that address each audience member indiscriminately, we can rest assured
that what we are viewing is “the definitive thing.” Yet where do we go to examine
baseball? Simply reading the rules—that which we use to define the thing—would not
seem an adequate object for study. As Salen and Zimmerman’s soccer example
demonstrated, it is not the rules but rather the “complex whole” that manifests in the
playing that comprises a game. Yet with such variance in that instance, how can we use
one match as exemplar? Despite games lacking the materiality or fixity of a text, there is
still a certainty with which we can discuss a game—we all know what baseball is. Yet
where do we locate it?
This is the problem of “live” discourse, discourse that does not manifest through
a locatable object. While we can point to the film, the book, the painting, or the sculpture
and specify that that is the text, that is the discourse realized, performative media such
as music and theater, like games, cannot be referenced directly outside of their live
context. They exist as an experience, as an event that is occurring now, and any
attempt to capture that activity is merely to create a new object, a text that merely
represents—that is, re-presents—the original experience.
We can point to sheet music as a text, however this is not the music itself—it is
merely a representation of the notes that make up the song. Similarly, we can reference
the words of a playwright, bound on paper, as a tangible artifact, but that text is merely a
script. Both of these objects could be compared to a game’s rulebook—they are a
description of the thing, instructions for how it is to be played, but they are not the music,
the stageplay, nor the game itself.
Equally, we can point to a music recording, such as a CD or an mp3 file, yet this
is also not the music itself—it is merely a depiction of the music, a recording, a
representation of the musicians performing. We can look to a videotaped recording of a

58
stageplay’s performance, yet this is a vastly different thing than the play. Both of these
might be compared to a recorded sporting match, a representation of a game that has
already been played, or to screenshots or motion captures of a videogame. These are
documents of how it was played, yet they are similarly not the music, stageplay, nor
game itself.
Instead, a live performance breaks the boundaries of textuality by existing as an
ethereal, intangible instance that cannot be completely captured in an object.
Performance’s only life is in the present. Performance can not be saved,
recorded, [or] documented…: once it does so, it becomes something other than
performance. To the degree that performance attempts to enter the economy of
reproduction it betrays and lessens the promise of its own ontology.
Performance’s being…becomes itself through disappearance.
64

The experience of an action occurring right now is part of what defines the activity, and
we can extend this notion beyond performative media to any experiential event.
Whether a musician playing an instrument, an actor playing a role, or a player playing a
game, to play (just as “to perform”) necessitates an action that escapes the boundaries
of textual objects.
To put it another way, try this head-scratcher: if a game is designed but never
played, does the game exist? Without an active playing of the game, is there a game?
We could pose this same question of other “live” arts: if I compose a symphony that is
never played, is there a symphony (or merely sheet music)? If I write a play that is never
performed, would we consider it a play (or merely a hypothetical script for such a thing)?
On the one hand, videogames are objects—they are discursive artifacts, the
product of designers who have created and completed a predetermined text. Everything
that exists within the videogame—the code that makes up the game—is fixed. And yet,
videogames are equally not objects, not predetermined, not fixed, as the details of what

64
Peggy Phelan. Unmarked: The Politics of Performance. Routledge, 1993. 146.

59
happens in the game will only be realized upon playing. While in baseball or chess, the
object that contains the system of rules—the rulebook—is most certainly not the game
itself, a videogame differs in that we do take the object which contains the system of
rules—the disc or cartridge that holds the code—to be the game.
65
And yet like chess,
the experience of playing is a manifestation of those rules that cannot be realized in the
object alone.
This is because, like the performance of a song or of a play, the playing of a
game is merely a derivative—it is not a definitive product. Games are inherently
derivative in that their matches manifest from the system of rules, from the structure that
defines them. Generated from the rules, the playing of a game is but one of many
versions, and while perhaps similar to many other versions, it will inevitably be—at least
in some infinitesimal way—different from every other iteration. Every playing of the
game thus becomes another derivative of that game, another playing out of the system
of rules, and every playing thus carries an aura of uniqueness. This is what makes a live
experience special and exciting—that it is in some way distinctive and individual, even
when guided by a fixed text. Any concertgoer can attest that the experience of a live
show is quite a different phenomenon than listening to a CD—than the experience of a
text. After all, why do millions of concertgoers spend the time and money of going to live
performances when they could simply buy an album and listen to it whenever they like?
The live performance must offer something that an album cannot. For while a studio
recording may capture one particular, idealized version of a song, the experience of a
live iteration provides an energy and excitement that is lost in a text, and is often
attested to by its very imperfection—it is the energy of an event that is happening now,

65
Note that while videogames often come with an instruction booklet, these texts do not contain
the entirety of the game’s system of rules—they only explain what little information the player
need know in order to play. The vast majority of the rules—the physics of the gameworld, its
spatial layouts, the sequences of events, etc.—need not be explained to the player.

60
of bearing witness to its occurrence, and of knowing that it cannot be idealized or
predetermined (hence the violation often felt after discovering that a performer is lip-
syncing—or that a game has been fixed). Similarly, while a videogame might allow for
an idealized version of the game (a “proper” sequence of events, as dictated by the
designers), or several potential versions, or merely a range of possibilities, none of these
can be pointed to as “what definitively happens.” Despite the videogame’s status as a
text, it still allows for the “non-textual” energy of a live iteration.
A live experience remains special not only in its aura of uniqueness, but also in
our knowledge that the experience cannot be possessed. Unlike objects, an experience
carries with it the implicit knowledge that it will be gone as soon as it is finished, unable
to be possessed and never to be seen again. Texts can be obtained and possessed,
can be held and enjoyed, and can be revisited by later viewings or readings. One can
lay claim to ownership of a text, allowing some mastery over the object through its
possession.
66
Yet experiences cannot be possessed in this way; they are fleeting and
cannot be captured, try as we might (demonstrated most evocatively in vacationers who
obsessively photograph their travels in the hopes of retaining a moment that inevitably
escapes as soon as it is over). This is the nature of an experience—it cannot be
possessed in the same manner as a text, and we can see this at work in videogames. I
can vividly recall the experience of entering the ruins of Las Vegas in Wasteland, of
defeating the final boss in Super Metroid, and of first encountering the Flood in Halo, yet
none of those experiences can be replicated. I can go back and play those games
again, but I would inevitably make different choices, encounter the events in a different
manner. This would not be the same experience; it would merely be a new iteration.

66
Consider the obsession many people have with owning the DVDs of movies that they love—
even if they never actually take the disc out of its cellophane wrapping—simply to satisfy some
need or desire to possess the object.

61
This is why a game lacks a definitive object. Because it exists only in the
playing, and because every iterative match will be in some way different, we cannot
reference a definitive game—we can only glean it through its derivatives. Thus in order
to study a videogame, we must look at the text, but we must also look at hypothetical
derivatives of the text—hypothetical because every player’s playing will be a different
one, thus we cannot account for any one specific playthrough. While the playing
emanates from the text, the iterative experience of play is just as much “the thing” in
question as the discursive artifact of the text itself. We cannot separate these two.
Distinction #5: As texts, videogames provide a definitive object of analysis,
while as games, they can only be gleaned through their derivatives.

Because games exist only in their derivatives, and these change from play to play, the
“content” of a game actually exists in the playing, rather than in the system of rules that
comprises it. The content of a baseball game results from what the players do during
nine innings, and it is because every match is a different instance, with content changing
from game to game, that fans continue to watching all season long. While the content of
the game can broadly be considered as the goal-oriented conflict, the content that
players are concerned with is the outcome of their particular iterative. After all, when
reading the rules of baseball, the reader has no investment in whether the home team or
the away team should win; this conflict only arises in an actual competition.
Two further distinctions follow from this immediacy of gameplay. First, the game
is happening “immediately,” right now. The very premise of a game relies on a context
in the present-tense—the game is played because the outcome is not fixed, because
winner and loser cannot yet be decided, therefore the actions within the game retain
their importance. Games do not assure the rigidity of texts; to the contrary, they assure
flexibility and disparity. This is precisely why sports fans are able to watch their favorite

62
team play week after week, year after year: despite watching countless matches, every
instance offers a new experience, as the outcome always remains uncertain. In fact, the
very notion of a “fixed match”—a match that has been predetermined—generally violates
all sense of gamesmanship, as it is the very openness of a game that makes its playing
worthwhile. If the outcome is assured, it hardly seems a game at all (the very reason
that few of us would consider WWE wrestling to be a game of any sort). So while a text
is inherently predetermined, a game relies on its having not yet been determined.
67

Secondly, as an event happening now, games are able to appear im-mediate.
We often forget that “immediacy” entails a lack of mediation, a suggestion of direct
interaction rather than separate acts of production and reception. When dealing with a
live event—a musical performance, stageplay, speech, or even a simple conversation—
we don’t feel as though anything stands between sender and receiver; rather, the
experiential nature of the live discourse provides the illusion of not being mediated.

As Kinneavy demonstrated, the nature of discourse is such that no communication goes
unmediated—ideas cannot be directly transferred between people. Instead, discourse
must always entail a means of expression (a “signal”) that carries the content of that
communication from sender to receiver—if nothing else, language mediates the

67
This is further demonstrated in televised sporting events, where immediacy seems essential—
the discourse of such events repeatedly emphasize their “liveness,” stressing it as an experience
that is occurring right now and attempting to replicate the pleasures of attending the game. Such
live televisual discourse (not just in sports, but on the news and in other live events) is presented
in a manner that is distinctly non-textual, differentiating itself from the predetermined-ness of
traditional texts. Further, sporting events are rarely rebroadcast or played on tape-delay, as
knowing that the outcome has already occurred seems to dilute the point of watching.

63
information. Yet live communication appears transparent because there is no tangible
object between sender and receiver. Nothing “perceivable” mediates the
communication, and as a result, the discourse seems direct and unmediated—the words
of language, the sounds of instruments, or the performance of theater all occur
instantaneously, and thus appear im-mediate. This illusion is key to “liveness.”
Gameplay equally entails liveness in that the player interacts with the game—the
system of rules—in the immediate present, and the resultant gameplay is spontaneous
and instantaneous. The game is live in that the playing—the discursive manifestation of
content—is occurring now, manifesting from the player’s interaction with the game (and
possibly with other players), and this playing appears to occur without mediation. Much
like music and theater, games do provide mediation. As a soccer player, I interact with
the opposing goalie differently than I would otherwise, solely because the game
positions them as my opponent—my relationship to the other players is mediated by the
game itself. Similarly, the computer hardware and the videogame’s formal interface
mediates the videogame experience, always present between the player and the action
they perceive on screen. Yet just as language occurs as an invisible mediation in verbal
discourse, the rules and roles of gameplay equally occur transparently, providing an
“immediate” interaction.
Texts entail quite the opposite. Textual discourse cannot help but acknowledge
a tangible mediation—via the object itself—and as a result, struggles to create such an
illusion (television news, for example, uses a variety of cues to offset the TV-as-mediator
and create a sense of “liveness,” yet one cannot help but be aware of the spatial
displacement occurring through the screen). Here, the mediation of the text is apparent
and visible—we are aware that we are viewing a painting, watching a movie, reading a
book, or listening to a CD, and aware that an artist created that object at some point in

64
the past. The tangible object is our guarantee of mediation (in this sense, excessive
formalism—overt style that draws attention to its textual status—is a move in the
opposite direction of immediacy, a celebratory reveling in the medium rather than an
attempt to conceal it).
Further, texts usually imply a past-tense—as a finished object, a text carries an
underlying implication of the past, both in form and in content. Its discourse originates in
the past, even while being received in the present, and resultantly its content is equally
perceived as occurring in the past, as whatever the text depicts must precede the text
itself. Note that stories—even those set in the future—are told using the past-tense.
Similarly when we view a film, painting, or photograph, it carries with it an assumption
that the material being viewed has already occurred (as evinced by its very
documentation). In this regard, texts assert predetermination, events etched into the
fixed materiality of the object. This is the very tension that accompanies linear
gameplay—those videogames that entail a predetermined sequence of events, the most
“text-like” of videogames. As a deliberately transmitted story, they evince the
predetermination of past events, while as a game to be played, they attempt to suggest
events in the present-tense, implying some degree of uncertainty.
68

This offers our final two distinctions:
Distinction #6: As texts, videogames imply a past-tense, evincing
predetermination, while as games, they are happening in the present-tense,
evincing uncertainty.

Distinction #7: As texts, videogames foreground their mediation, while as
games, they suggest immediacy.

With all these oppositions between texts and games—at minimum, the seven I’ve listed
here—how do we reconcile a videogame as both things at once? How do we address

68
We will see a prime example of this phenomenon, modulating between past and present-tense,
in chapter four when we look at Max Payne.

65
Civilization III as an object, but at the same time not an object? As a fixed instance, but
at the same time a variable one? As occurring in the past, but also in the present? In
short, how do videogames manage to merge textuality with gameplay?
As texts, videogames are a product of discourse, a tangible, mediating object
that facilitates communication between sender and receiver, an object that we can point
to as “the text” itself:

As games, videogames are the facilitator of another sort of discourse, fashioning a
separate instance via the player’s interaction with the game:

To best address videogame discourse, then, we need a model that addresses this full
situation:

In truth, there are two locations to which we could point as where Civilization III resides:
(1) the game-text—the object that comes packaged in a box, a gamedisk that I can
assume to be identical to millions of other disks so that whenever I discuss Civilization III

66
with someone else, I can reasonably assume that we are discussing the same object; as
well as (2) the actual playing of that game—an individual iteration that I can reasonably
assume to be different from the experience of any other player.
Semiotic meaning is the fundamental property of a text—it exists to represent
things outside of itself, thus exists for the sake of something else. Yet play is the
fundamental property of a game—it exists for the experience that it allows, thus exists
for its own sake. With videogames caught between the two, we will find our best
understanding of the medium’s discursive properties by considering videogames in
relation to both of these impulses. As such, these two purposes provide the basis for the
latter part of this chapter. To discuss videogames as discursive objects, we will first
examine their semiotic properties—the manners through which videogames provide
signification. Then, to discuss videogames as rule-based activities, we will examine their
ludic properties (from the Latin ludus, “to play”)—the manner through which videogames
provide a spontaneous, emergent experience. Rooted in both texts and games, two
drastically different modes of human experience, we will find our best chance at divining
the “art” of videogames—the nature of their aesthetics—by first considering the medium
in relation to their most deep-rooted impulses.

Discursive Systems: the Semiotics of Videogames

Take a look at the above image. Seem familiar? Think of all the possibilities for
what that image might represent. Now look at that image in context:

67

Was “ball” one of the possibilities you considered?

How about “missile” (center screen), “radar blip” (top), and “abducted humanoid” (bottom
right) simultaneously? Is that the image you thought you were looking at?

Or did you say, “that’s me—a knight on an epic quest to recover a golden chalice?”

68
Such are the peculiar ways of videogame signification.
Videogames are a representational medium—the formal properties of its texts
serve to represent people, places, objects, ideas, and all sorts of other things. And as
these three examples demonstrate, a single formal instance—such as our simple
square—can be deployed to represent a variety of different things. In Pong, our square
represents a ball; in Defender, squares serve multiple purposes, depending on size and
context; and in Adventure, a square not only signifies “knight,” but also the player
themself by serving as the game’s avatar. Our concern, however, is not so much with
what these videogames mean, but rather how they mean. How is it that this simple
shape can represent so many different things, none of which bear any resemblance to a
square? How is it that we read entire worlds into pixels of light emanating from our
screen? Because videogames, like any other manner of text, are semiotic by nature.
When we address videogames as texts, we are discussing them as systems of
signification, considering the manner in which the text’s formal properties—those things
perceivable via the senses—communicate meaning to the receiver. The images we see,
the sounds we hear, and the sensations we feel all function together to create a
coherent whole (the medium has yet to attempt smell or taste, yet theoretically, these
senses could be utilized as well), and we, as receivers, interpret and translate those
formal properties to infer some meaning from the form. Pixels of light come together on
the screen and we recognize a mustachioed plumber defeating a prehistoric turtle to
save a princess. Sound waves reach our ears and we hear nearby gunshots or the
British accent of a tomb-raiding heroine. Our gamepad rumbles and we feel the
shockwaves of an explosion or the tumblers of a lock falling into place. Despite this
sensory data being entirely artificial and the things we perceive remaining non-existent,
we nonetheless comprehend the worlds presented to us by videogames. In other words:

69
videogames mean something. Their formal properties signify something to us, and by
comprehending them, we are able to engage their fictional contexts. So long as we’re
able to translate a square block as “ball,” or “humanoid,” or “me,” that fictional world is
able to come to life, and we’re able to participate in it. It is semiotics that allows this to
occur.

From Mimesis to Abstraction
Semiotics was born in the late 19
th
century of a Swiss linguist and an American
philosopher. Working independently and unaware of one another’s pursuits, Ferdinand
de Saussure and Charles Sanders Peirce both envisioned a field of study dedicated to
the manner in which humans communicate and create meaning via the exchange of
signs. A linguist by trade, Saussure was concerned with the signification of language,
fascinated by the human ability to exchange meaning via letter combinations and sounds
that are arbitrarily assigned meaning. The pragmatic philosopher Peirce addressed
language as well, but was also concerned with symbols, icons, and other indicators, both
natural and arbitrary, from which meaning is extracted. Yet on different continents,
working in different fields and examining different instances, both men were pursuing
similar endeavors: a better understanding of the human ability to communicate through
signs. Saussure called his field “semiology”; Peirce termed his “semiotics.”
The basic notion of a sign is quite simple—it is any instance which evokes
meaning. As Peirce put it, a sign is simply “something which stands to somebody for
something in some respect or capacity”—words, images, sounds, and gestures are
among the most common, yet nearly anything might function as a sign, so long as its

70
deployment evokes some mental concept beyond itself.
69
When we hear the word “bird”
and think of a certain winged creature, when we see a red hexagon and know to stop
our vehicle, or when we smell smoke and infer that a fire is near, a sign is at work.
Saussure provided the most straightforward model of signification, claiming all
signs to be comprised of two separate but interrelated parts: a signifier and a signified.
The signifier is the formal instance of signification—the letters B-I-R-D, a red hexagon, or
the smell of smoke. The signified is the mental concept evoked by that signifier (not the
actual thing itself, mind you)—the mental archetype of a bird, the concept of stopping, or
the awareness of a nearby fire. By Saussure’s thinking, it is these two components
combined which comprise a sign—the formal instance along with the “content” of that
instance (although Saussure did not necessarily intend it as such, distinguishing these
as the sign’s form and content is a logical step). Note that the content need not correlate
to some actual “thing” in the world, and whether or not it does has no consequence
here—the words “turtle,” “freedom,” and “vampire” all signify in the same manner,
regardless of indicating tangible, conceptual, or fictional signifieds. All that matters is
that the receiver interprets meaning from the sign.
70

If we take our pixilated square as our signifier, then, how is it that it may carry so
many different signified concepts? How can that same shape represent such disparate
notions as “ball,” “missile,” and “knight?” We need a two-part answer to explain this one.
First, it is because, as Saussure has assured us, meaning is not inherent to a
sign, but is instead assigned by convention. While audiences generally take meaning for
granted, assuming that discursive instances simply “mean what they mean,” semiotics
reveals the arbitrary nature that often dictates the associations between signs and their

69
Charles Sanders Peirce, Collected Papers of Charles Sanders Peirce, vol. 2, ed. Charles
Hartshorne and Paul Weiss (Cambridge: Harvard University Press, 1935), 228.
70
Ferdinand de Saussure, Course in General Linguistics, eds. Charles Bally and Albert
Sechehaye with Albert Riedlinger, trans. Roy Harris (Chicago: Open Court, 1986).

71
referents. Signs are instances that invite interpretation, and while they may or may not
encourage one particular interpretation, little if any meaning is inherent to the text. The
letters B-I-R-D reference a particular winged creature simply because we have assigned
such meaning to that letter combination, and we can apply this same logic to our
pixilated square. Umberto Eco points out that semiotics need not only address
conventional signs, such as words and symbols—semiotics are at work any time one
thing “stands for” something else.
71
As a sign, then, our square can mean whatever the
game’s designer designates it to mean, so long as that designation can be
communicated to the player. Yet we need not even go so far as to consider the
designer’s intentions. Peirce tells us that signs need not be intentionally transmitted to
create meaning; signs can be incidental, unintentional, or inadvertent, revealing that
videogames (like any text) may carry unintentional messages. Just because a designer
did not intend a particular signification does not mean that it is not valid or inherent to
one’s experience of the text. If a player of Defender misinterprets the “abducted
humanoid” squares by reading them instead as “supplies,” signification would still be
occurring (miscommunication is communication nonetheless).
So if signification is conventional, and thus variable, what dictates how our
square will be interpreted? Here’s where we get to the second part of our answer:
context—or, more specifically, the signifier’s position within the system of the text.
Any videogame contains innumerable representations, all working together as
one cohesive system. Looking at Pong—quite possibly the most simplistic and
minimalist videogame one could imagine—we still see numerous signifiers at work: the
pixilated square signifies a ball; the lines to the right and left of the screen signify
paddles, but also players 1 and 2; the entirety of the screen signifies a space of action;

71
Umberto Eco, A Theory of Semiotics (Bloomington, IN: Indiana University Press, 1976), 7.

72
the player’s movement of the game’s trackball signifies movement of one paddle; the
behavior of the “ball” signifies particular concepts of physics, including speed, time,
momentum, and deflection; the numbers at the top of the screen signifies a score; the
action of “missing” the ball signifies one point for the opponent. Yet each of these
components only gain significance in relation to one another. We read the pixilated
square as a ball because of its position amongst these other signifiers—its movement
across the screen (which we read as space), its behavior (which we read as imitative of
real world physics), its relation to the lines (which we read as paddles), and so on.
Outside of this particular context—the system of Pong—any of these signifiers might
indicate something else entirely. Such as a missle. Or an abducted humanoid. Or a
knight.
When we approach a more complex example of videogame—say, Civilization III
or Grand Theft Auto—unraveling the semiotic system becomes much more unwieldy.
Rather than attempt such an endeavor, then, let’s focus instead on an individual
instance of signification—an individual signifier pointing to a particular signified—to delve
deeper into the semiotics of videogames.
We’ll continue with our ubiquitous square, but this time consider it within the
context of Adventure, where it is used to signify a knight, and compare it with the knight
signifiers used by four other videogames—The Elder Scrolls IV: Oblivion, Ghosts ‘N
Goblins, Gabriel Knight: Sins of the Father, and Brimstone (a text-based adventure)—to
somehow reference that same signified concept.

73

Here we see five exceedingly different renditions of a knight. Each point to a similar
referent—a particular type of medieval crusader—yet do so in drastically different
fashions. These are not mere cosmetic or technological differences, mind you—each
signifier uses a fundamentally different means of representation to evoke the same
signified concept: the mental archetype of a “knight.” Let’s see if we can unpack these
differences.
Common sense would tell us that the knight signifier from Oblivion is the most
“knight-like” of the five—it is the signifier that we would most readily identify as the very
thing it is meant to signify. In a basic ontological sense, this seems the signifier most
patently connected to its referent, as it is the one that looks the most like the thing to
which it refers. We would call this a mimetic signifier.
The notion of mimesis stems from several disciplines, particularly aesthetics and
narratology. It’s a somewhat elusive term, as it gets used in a variety of manners.
Descended from the Greek, the word itself is synonymous with both “imitation” and
“representation,” and it is the leeway between these two that leads to its ambiguity. As
I’ll use it, mimesis is the process of sensory imitation—the attempt to look, sound, feel,
smell, or taste like the signifier’s referent. When we discuss “mimetic art,” we address
that which attempts to visually, aurally, or otherwise reproduce our perceptions of the
natural world. Portraits, landscapes, and other “photorealistic” paintings, sculptures

74
meant to resemble people and objects, theatrical and cinematic works that offer some
explicable world, and most photographs would all fall under the mimetic category. We
could extend this notion beyond the realm of art to include other manners of sensory
imitation: faux leather, “sounds of nature” CDs, scented candles, artificial flavors, and
innumerable other products. In each case, the object in question attempts to mimic
some phenomenological aspect of its signified referent, to create the illusion of a
sensory experience. Similarly, videogames which use mimetic signifiers—such as
Oblivion—attempt to reproduce the images, sounds, or haptic sensations of the things
they mean to represent.
The concept of mimesis descends from Aristotle and Plato, who both discussed
mimesis as an attempt to create associations with the world around us via its similarity to
that which we perceive everyday. The two had differing opinions on mimetic art:
Aristotle discussed mimesis as a primary part of the human condition, arguing that we
use imitation to learn about the natural world and get closer to “the real”—he noted that
even as small children, one of our first inclinations is to imitate—while Plato was more
skeptical, fearing the artificial façade of mimesis as one that we must “get beyond” in
order to experience “the real.” Yet both agreed that while we obviously do not confuse
mimetic art with reality, such art allows its audience to accept the world within it as valid
and verisimilar by providing images that look very much like people and places, sounds
that sound very much like voices and natural sounds, and/or other sensations strikingly
familiar to our everyday experience.
72

If we take Oblivion’s knight signifier as a mimetic one, how then do we distinguish
our other examples? While nearly all critics of aesthetics, narratology, and media

72
Note that the Aristotelian tradition commonly holds that mimetic art also serves to beautify,
idealize, or universalize the world it imitates, one of the many points of contention around this
term. My use of mimesis does not include this assumption, as I focus only on the act of
phenomenological imitation.

75
acknowledge some dialectic to address these representational differences, they’ve had a
difficult time coming to common terms for it. This problem stems, in part, from
differences in agenda. Plato and Aristotle, for example, discuss mimesis in opposition to
diegesis, distinguishing the act of showing from the act of telling. This pairing stresses
the presentational quality of mimesis over the descriptive or expressive quality of
language. Others instead speak of concrete representation vs. abstract representation,
stressing the “tangibility” of sensory imitation over other, more ethereal modes of
signification and addressing the manner of thought (concrete or abstract) associated
with each. Some discuss realism vs. abstraction (or realism vs. formalism), fixating on
mimesis’ illusion of “realness” when compared with styles that are further detached from
the real. And still others discuss representation vs. abstraction, focusing on the mimetic
act of “presenting again” (re-presenting) contrasted with expression by other means.
This last pairing has become the preferred terminology amongst videogame
scholars,

largely thanks to Mark J.P. Wolf’s influential essay “Abstraction in the Video
Game,” one of the first accounts to comprehensively deploy the term in relation to
videogames:
To abstract something is to simplify it, reducing it to a few essentials and basic
forms instead of trying reproduce it. Representation, which seeks to create
resemblances and reproduce something, is the polar opposite of abstraction (and
is sometimes conflated with realism).
73

Keep in mind that the deployment of these terms come from a tradition that uses
“representation” in a more restrictive sense than today’s common usage, which is often
synonymous with “signification” and tends to include any instance which points to a
signified referent (by this definition, we would have to say that abstraction is, in itself, a
form of representation). I’ll use representation in a more literal manner, as a re-

73
Mark J.P. Wolf, “Abstraction in the Videogame,” The Video Game Theory Reader, eds. Mark
J.P. Wolf and Bernard Perron (New York: Routledge, 2003), 48.

76
presenting of the referent; “depiction” or “portrayal” may be a little too specific, but these
are probably as close a synonym as we might find.
Despite a wide variance on the representational side of the binary, we see a
commonality in the oppositional term of “abstraction” (even the narratological pairing of
mimesis and diegesis fits to some degree, as I’ll explain shortly). Abstraction doesn’t
offer the esteem of mimesis’ philosophical origins, but its etymology still proves telling—
from the Latin abstractus, it means to “draw away” (ab=away, trahere=draw), to separate
or remove. Stephen Park notes:
Any definition of abstraction will necessarily be a binary one for we must address
what is being moved away from. Abstraction in painting and abstraction in
thought will obviously be different because they work within different contexts;
they are removals from different locations.
74

It is for this reason that abstraction fits so well in a wide variety of binaries—regardless
of the opposed term, an abstraction is simply a movement away from it, whether from
realism, from concrete thought, from representation, or anything else. In a semiotic
context, abstraction begins as a representation that removes characteristics of the
signified referent, most commonly by simplification, and at its most extreme moves away
from depictional means altogether.
75
In this sense, abstraction is a move away from
mimesis, a move away from pure sensory imitation to some other mode of signification,
and we can trace this movement through our five knight signifiers, each demonstrating a
different degree of abstraction at work.
If we take Oblivion’s knight as a mimetic one—the least abstracted of the five—
the next step down the line would be Ghosts ‘N Goblins’ knight, a cartoonishly pixilated

74
Stephen Park, “Abstraction,” The University of Chicago Theories of Media Keywords Glossary,
http://csmt.uchicago.edu/glossary2004/abstraction.htm (accessed November 2007).
75
We can see this most clearly in the history of written language, beginning with pictorial
representations, then (as those pictures become familiar and conventional), moving to more
abstracted characters (such as hieroglyphics), and eventually becoming so abstracted that they
no longer have any connection to their signified referent at all.

77
rendition that we might term an iconic signifier.
76
An icon (as I will use the term) is not
entirely mimetic, and not entirely abstracted. Instead, it is an abstract imitation, a
simplified imitation that bears resemblance to its referent yet would never be confused
with the thing itself. Icons have some perceptible relation to the thing they signify, yet
make no effort toward realistic sensory imitation—that is, no effort to pass for the thing
itself. Their mode of reference is not entirely literal, yet not entirely conceptual. Icons
provide us with the clearest example that we cannot simply divide signifiers between the
mimetic and the abstract, as these signifiers are unquestionably both.
Cartoons are the most obvious example of icons—animations which mimic the
fundamental details of the thing they mean to depict, yet lacking in many peripheral
details. Ghosts ‘N Goblins’ avatar can be easily recognized as a knight—it has the basic
shape of a human body and some semblance of a bearded face, as well as silver armor
and a helm, allowing a sensory recognition of what the thing is meant to be. Yet
simultaneously, it lacks many of the details that we see in Oblivion’s photorealistic
knight; while we might credit Oblivion with providing a legitimate attempt to give us a
“real” knight, no such claim would be made for Ghosts ‘N Goblins. Yet we can hardly
call this cartoonish knight an arbitrary signifier; it is chosen due to resemblance, even if
no one expects it to pass for the thing itself, so while it may not be an entirely “natural”
signifier, neither is it entirely conventional.
Symbols, on the other hand, are arbitrary, signifying only by convention, and it is
here that we would place Adventure’s knight. Flags, peace signs, pink ribbons, wedding
bands, and the Nike swoosh are all symbolic signifiers, signs so abstract that their

76
Peirce uses “icon” to indicate anything which bears a resemblance to the thing it is meant to
signify—including mimetic depictions—and resultantly, this has become the common semiotic
deployment of the term. However, I’m going to use a variation on the term, one more in line with
our contemporary usage of the word (particularly when used in relation to videogames and
computers).

78
meaning can only be determined by context or prior knowledge. If the receiver is not
already familiar with a symbol, and not provided systemic cues from which to infer its
significance, they would have no means of interpreting it properly. Texts may use
cultural symbols (those with which a particular audience may already be familiar), or they
may establish their own symbols (by creating such associations on their own), so the
conventions which define a symbol may come from within the system in which it exists or
from without. In the case of Adventure, the text establishes its own symbol for “knight,”
providing an entirely arbitrary visual indicator—a square—and defining that signifier by
the context of its usage. It is for this reason that you most likely did not recognize the
earlier square as a knight until viewing it in context; unlike a flag or peace sign, a square
carries no cultural or extra-textual significance, thus, as a symbol, must be defined within
the system of the videogame.
Language—whether written or verbal—is also composed of symbols. Barring the
rare exceptions of onomatopoeia (iconic words like “buzz” or “meow”), words have no
deliberate correlation to the things they signify, and equally represent only by
convention. This is precisely why different languages use different words—their signified
referents have been designated by custom rather than any natural reasoning. Linguistic
signifiers—such as Brimstone’s knight—are a particular manner of symbol, however.
Practiced within a unique set of conventions with which we become familiar at a very
early age, language is the primary means by which we communicate with others,
exchange knowledge and information, and express our ideas and feelings. In fact,
language is such a prevalent part of the manner by which we transmit and process
information that it is an inherent part of our consciousness, and negotiates much of our
thinking, making these particular symbols unique by being so implicit to our
understanding of the world. Further, language marks a departure from our previous

79
three in that it can no longer “stand for” in a perceptive sense. Even in Adventure, using
an arbitrary signifier with no sensory connection to its signified, our square was still able
to offer a concrete instance of some sort—we see it occupying space, and can still point
to it as the knight in question. With the word “knight,” however, we’ve left any sense of
concrete reference and moved entirely into the realm of concepts.
Here, we’ve crossed a particularly complicated threshold—that between
representational space and abstract thought. Adventure’s square is entirely arbitrary,
and thus not an icon by any means, yet unlike symbols such as peace signs, pink
ribbons, or words, our square still operates within representational space. Much like the
symbolic “X” on a treasure map, we can point to our square and say “there it is!” Despite
its arbitrary nature, the square provides a referent that can be located in space, that
invokes some degree of concrete existence. When we see the word “knight” in this very
sentence, however, we would not say, “ah, there is the knight!” nor do we locate “peace”
where its sign is drawn. Here, we’ve left spatial reference and moved into the
abstraction of ideas alone, essentially crossing the gap between mimesis and diegesis,
between showing and telling. In truth, Adventure’s square and a treasure map’s X both
provide the absolute minimal degree of mimesis possible by imitating their referents only
in that they “exist somewhere,” even though common sense would tell us that these
signifiers have much more in common with arbitrary symbols like peace signs and flags
than they do with imitations or icons. Should we, then, delineate spatial symbols—those
that “show”—from intangible or conceptual ones that merely “tell”? Indeed we could, yet
where would we draw this line?
If we play Rogue, an ASCII dungeon-crawler which uses letters and typographic
symbols to create a rudimentary map, we might see a “T” indicating a troll positioned
next to the “@” that indicates our avatar. While this instance certainly indicates a spatial

80
relationship, like any map would, the game also relies on stats and other contextual
indicators to explain what is occurring, requiring the player to “fill in the gaps” by
imagining that a troll is attacking the game’s adventuring hero. As such, should we call
the T a spatial symbol because there is some sense of concrete space depicted,
allowing us to see where it is in relation to the environment, or a conceptual symbol
because we need to imagine the troll, much like Brimstone’s text-based gameplay would
require? While it may seem that there should be a definitive distinction between
showing and telling, between signifiers in space and in thought, the truth is that many
instances will problematize such a distinction. Peace signs and flags might fall on one
side of this threshold, while a treasure map’s X or an ASCII T fall on the other, but I’m
not going to worry over finite distinctions. For our purposes, they’re all symbols—
arbitrary signifiers that represent only by convention.
The unique symbols of language, however, always remain thought-based—must
always tell. Even when we see the word “Florida” written on a map of the United States,
we do not think of the word as delineating any sort of spatiality; rather, the “L” shaped
land mass positioned in the southeastern corner of the map is the spatial signifier—
“Florida” merely names that land mass, “tells us” what to call it. Similarly, when we see
the phrase “you are here” on a shopping mall directory, that phrase rarely indicates
“here” on its own; instead, it is typically accompanied by a separate spatial indicator (a
circle or triangle) rather than allowing the words themselves to designate a point in
space.
Language has become such an innate part of our everyday lives that we rarely
give it a second thought, and often assume its conventions to lack the signifying
flexibility which characterized our simple square; we figure that “knight” simply means
knight and that’s that. Yet even words retain flexibility, specifically because they are

81
assigned meaning by convention. In this chapter alone, I’ve found it necessary to
delineate my usage of such terms as system, text, game, discourse, mimesis,
abstraction, representation, and icon for this very reason. Despite using a code of
conventions agreed upon by both author and reader—the English language—and
despite many of these words being moderately common in that language, different
people still deploy these words in different manners. Because words are arbitrary
signifiers, their meanings can evolve, or be repurposed altogether. Language even
allows us the capacity to coin entirely new words, to which we can assign whatever
meanings we wish. As such, words only have the meanings we give them.
When Brimstone uses the letters K-N-I-G-H-T, we still must consider the context
of that system and delineate the precise usage involved. After all, “Launcelot, knight of
the Round Table,” “Batman, the Dark Knight,” and Knight Rider’s “Michael Knight” all use
that particular signifier to different ends—the first points to its traditional medieval
signified, the second to a more generic signified retaining only its broad characteristics of
nobility and adventuring, and the last repurposes the word as the name of an individual,
providing an entirely new signified. But in each case, we bring our existing familiarity
with that signifier’s conventional usage(s) and consider them in relation to the signifier’s
current context.
Yet when we repurpose such a signifier (whether by broadening it, as with
Batman, or redeploying it altogether, as in Knight Rider), what happens to that original
meaning? Rarely does it disappear entirely, and it is here that we can address our final
signifier, Gabriel Knight, protagonist of Gabriel Knight: Sins of the Fathers and its two
sequels.
Gabriel Knight is a bookstore owner and aspiring author who embarks on a
series of supernatural adventures in modern-day New Orleans. In this instance, “Knight”

82
is merely a surname—a linguistic signifier that represents an individual, just as “Buerkle”
indicates the author of this dissertation. The word in no way points to a medieval
crusader in the manner that Oblivion’s, Ghosts ‘N Goblins’, Adventure’s, or Brimstone’s
knight signifiers do. Yet we’re still aware of the word’s original meanings, even if we’re
consciously aware that in this context, the word merely points to some fellow named Mr.
Knight. The mental archetype of a medieval crusader still somehow lingers about, a
signifying ghost that continues to haunt its new usage. As a hero embarking on some
manner of adventure, our association of Gabriel Knight with such figures becomes all the
stronger. How, then, do we address this lingering, associative trace? How does this
character still point to the same referent that we saw in our other four examples?
Here, we have a new manner of signifier, one even further abstracted than
symbols or language—a connotative signifier. Connotative signifiers are the most
abstracted of our five types, as far removed from mimetic representation as we might
reach, signifying only via an implicit or uncertain association. Stereotypes, emotional or
affective associations, cultural associations, intertextuality, metaphors, impressions,
codes of taste, and codes of style are all invoked by connotative signifiers—instances
which create an implicit, ambiguous, or “unspoken” association rather than the explicit
meaning of imitations, icons, symbols, or language.
Numerous theorists have discussed such meaning in semiotic terms, but none
more famously than Roland Barthes. Barthes notes that Saussure’s model of signs
deals only with literal meaning—denotation, in Barthes’ terms—at the expense of
connotative meaning. He extends Saussure’s model by saying that while a sign is
indeed comprised of a signifier and signified, that sign (in its entirety) then becomes

83
another signifier of further, more ambiguous meanings—connotations.
77
To make this
leap, Barthes borrows the notion of “orders of signification” from Louis Hjemslev,
discussing denotation as a first-order of signification, and connotation as a second-order
that attaches the first, explicit sign to a second, implicit one.
78
When we see the name
Gabriel Knight, we know it to signify our modern-day hero. Yet beyond this explicit
reference—the name’s denotative meaning—the name also evokes notions of nobility,
adventuring, righteousness, and good conquering evil, all connotations brought by the
cultural baggage of the word “knight” (and we’d have to combine these connotations with
those of “Gabriel,” invoking the archangel of the same name to further connote
righteousness, in addition to a plethora of biblical associations).
Videogames are chock full of connotations, and they are by no means limited to
those surrounding words. Score music, for example, typically connotes a particular
affect—the theme from Halo connotes excitement and adventure, the trumpets of Medal
of Honor connote patriotism and “the just war”, and the use of John Williams’ “Imperial
March” in Star Wars: Battlefront creates an implicit sense of menace and danger (in
addition to more explicitly signifying Darth Vader and the Empire, thanks to the
signification assigned by the films). Racial stereotypes, such as those in the Grand
Theft Auto franchise, are lambasted precisely for their connotations—while each
stereotyped Mafioso in Grand Theft Auto III may only denote an individual character in
the game’s fictional universe, these characters could be seen as making implicit
suggestions about the Italian-American community at large, an ambiguous and

77
Roland Barthes, Mythologies, trans. Annette Lavers (New York: Hill and Wang, 1972. Barthes
also revisits these ideas in much of his other writing—for example, discussing photographic
images in terms of both objective signs (the “studium”) as well as subjective associations (the
“punctum”) in Camera Lucida. Roland Barthes, Camera Lucida, trans. Richard Howard (New
York: Hill and Wang, 1981), 25-7.
78
Louis Hjelmslev, Prolegomena to a Theory of Language, trans. Francis J Whitfield (Madison:
University of Wisconsin Press, 1961).

84
unspoken signification that escapes the game’s fiction to characterize people in the
world outside. Connotations are at the heart of many such controversies (in videogames
and in other media texts), precisely due to their ambiguous nature—while denotation can
be readily discussed by providing fairly concrete meaning, laying all their proverbial
cards on the table, the abstract nature of connotations make them much more difficult to
examine or annunciate. Resultantly, they can seem all the more dangerous (“they’re like
subliminal messages!”), are difficult to discuss (“don’t you see what I see?”), and can
often invoke further, undue paranoia (“who knows what other messages are in there?”).
With these categories in mind, we can distinguish our knight signifiers in the
following order, from most representational to most abstract:

Keep in mind that these examples are delineated only in how they point to one particular
signified—the archtypal notion of a medieval crusader. After all, the Gabriel Knight
example iconically points to a blond-haired male, and linguistically points to a specific
person, yet our concern here is only in how we invoke the particular signified that all five
examples have in common. When looking at the entire range of signification bound up
in a particular instance, we might see it operating in any or all of these categories.
As Barthes noted, texts offer different “orders of signification”—any given signifier
might point to numerous signifieds, each operating on a different level of meaning.
Barthes speaks of only two such levels—denotation and connotation—yet we might go

85
further beyond these. If we’re willing to take Ghosts ‘N Goblins’ world at face value as a
fictional universe complete with cartoonish characters (in the same manner that we
might say Mickey Mouse simply is Mickey Mouse, that the animation is exactly what
Mickey Mouse is supposed to look like), then we could say its avatar is a mimetic
indicator of a character, an icon of a knight, a symbol of the player, and a connotative
indicator of chivalry, nobility, and heroism. That gives us at least four different
contexts—four different levels of signification—in which this same signifier points to
different signifieds in different manners: denotatively and connotatively in relation to the
world at large (pointing to the mental archetype of a knight and to our associations with
that archetype), as well as in a narrative context (pointing to the character) and in the
context of the formal game system (pointing to the player).
These categories are by no means a definitive list, and the distinctions between
them are hazy at best. While this scale provides a nice sampling of the variety in modes
of signification, we’d be hard-pressed to drop any one signifier in any one category.
We’ve already seen ambiguity between icons and symbols (symbols that “exist
somewhere”), and between symbols and language (Rogue’s troll). We might further
note that Oblivion’s knight is not a perfect imitation, so would this mean that it is
technically iconic (if not, where do we place the division between the two)? And what do
we do with words that have “built-in” connotations, such as the negative implications in
the word “notorious?” Are these linguistic or connotative?
Rather than thinking of these categorically, we ought to instead use a spectrum
approach, thinking of signification along a sliding scale with each category demarking a
certain range along it:

86

Yet even this needs further tweaking, as this model neglects to distinguish the
fundamental distinction between showing and telling, and the threshold between the two.

Only our first three categories are capable of addressing concrete instances within a
representative space—this is why our first three examples are the only ones that we
point to as “the knight itself,” where we can conceptualize the formal instance that we
see onscreen as somehow being the thing in question. We might discuss the formal
properties of the letters k-n-i-g-h-t in Brimstone or the pixilated nature of Gabriel Knight,
yet this would not be addressing the manner in which these point to the mental
archetype of a knight—this would instead address the manner in which a particular set of
lines indicates a word of the English language (which in turn signifies a knight), or the
manner in which a pixilated sprite signifies a fictional bookstore owner (which in turn
connotes knightly associations). If we’re limiting ourselves solely to concrete
sensations—images, sounds, and tactile sensations—and the manner in which they are
read within representative space, we can only deploy these first three categories.

87

Equally, only the last three can escape concrete instances to indicate abstract notions;
mimetic and iconic signifiers must always offer a sensory perception which is, to some
degree, taken as the referent itself, therefore can only address that which has a
perceptible quality. Symbols, language, and connotations, on the other hand, can
escape representative space and signify solely in the realm of ideas, referencing
abstract and intangible concepts such as love, freedom, evil, infinity, or time, concepts
which cannot be deployed mimetically or iconically. Even when signifying actual
“things”—as in referencing a knight—such signifiers do so conceptually, asking the
receiver to imagine the thing in question rather than providing a sensory frame of
reference. Symbols, then, provide the middle-ground which can go either way, arbitrarily
assigned as a concrete thing or a conceptual notion, and we can see how Adventure’s
knight signifier operates both ways. Adventure gives us a concrete thing to point to, so
that we can reference the square and say, “that’s the knight.” But at the same time, any
sense of a knight must be imagined by the player, as—unlike Oblivion or Ghosts ‘N
Goblins—we don’t really believe that square as the literal actant. Adventure’s square,
then, demonstrates a simultaneous showing and telling.

88

It is for this reason that some videogames seem more reasonably discussed as
representational objects, offering tangible and concrete signifiers much like a Hollywood
movie or a classical painting, while others seem more closely related to abstract media,
offering ambiguous and conceptual signifiers much like music or modern art. To one
degree or another, however, every videogame must ultimately be some combination of
the two.

Plasticity and the “In-Between” Arts
All videogames are representative to some degree—if they were not, we would
have no means of entrance, no means of understanding what is occurring. Without
some imitation of some activity that we can recognize, even in the most abstract sense,
players would not be able to make sense of the sensory stimuli they are provided.
Essentially, the game would have no meaning. Equally, all videogames are abstracted
to some degree—in one way or another, the player will always be filling in details,
making sense of what they perceive and participating in the textual process, as no
videogame can include “everything.” While we might posit hypothetical extremes of pure
representation and pure abstraction, all signifiers are going to lie somewhere in between,
in a technical sense—after all, even the most realistic imitation will need to be somehow

89
abstracted (a photograph, for example, abstracts by eliminating the third-dimension),
while the most abstract connotation must have meaning that comes from somewhere
(otherwise it would have no meaning at all, and thus not be a signifier).
This variety is largely due to the animated nature of videogames. Many
contemporary forms of representational media have a certain degree of mimesis or
abstraction fixed by their very nature—the ontological nature of cinema and
photography, for example, automatically brings with it a high degree of mimesis, as does
the live performance of theater. On the other hand, literature, bound by language, is
innately abstract. Aristotle and Plato distinguished mimesis form diegesis for this very
reason—they saw that while the “showing” of theater offers a mimetic presentation, the
“telling” of literature and oration provide the ultimate abstraction, choosing to represent
through description (a conceptual activity) rather than depiction. This is also why
mimesis vs. diegesis was our only pairing that did not reference abstraction—Plato and
Aristotle were only dealing with performance and narration, a distinction that only offers
an either/or proposition. If to abstract is to simplify or remove details, telling and
showing seem to offer no leeway—you either show (represent) or tell (abstract); there is
no in-between. “In-between” only occurs in the plastic arts.
Painting, sculpture, mosaic, ceramics, woodworking, metalworking, and other
arts that use pliant materials are the “in-between” arts.
79
Such media are characterized
by their ability to be shaped and modulated, offering a more extensive flexibility than the
strict mimesis of photography or the stark abstraction of language. These can employ
deliberate mimesis or abstraction, but frequently offer some variation between.

79
Traditional accounts of the plastic arts will often exclude painting, using the term to distinguish
three-dimensional arts. Yet because paint offers just as much plasticity as these other materials,
I’m including it in this list.

90
The classically realist paintings of the Renaissance—a period before
photography had arose to make realistic painting irrelevant—strove to offer pure
mimesis (or at least, as close to it as the skill of the artist would allow). With no media
better suited to capture realistic depiction, the purpose of painting was assumed to be a
mimetic one, used to capture the visual representation of people, places, objects, and
individual moments in time, much like we now use photography. A portrait such as the
Mona Lisa demonstrates, serving to “show” much like any other mimetic art.
Centuries later, modern artists of the 20
th
century explored abstraction as a
deliberate transgression of such technique, perhaps best exemplified by the works of
Jackson Pollack—drips and splatters of paint entirely removed from mimetic
representation, seeking not to depict, but rather to express—to conceptualize emotion
abstractly, essentially “telling” much like language-based arts, though lacking in the
convention-based specificity of language. These are connotative works—they signify
emotions, ideas, and other ethereal concepts which lack any perceptible form that could
be imitated via the senses, or even annunciated by precise wording, and they outwardly
oppose the mimetic ideals championed by Aristotelian and Renaissance tradition.
Yet in addition to these extremes, painting has seen many degrees in-between.
The works of impressionism and early expressionism, such as that of Claude Monet or
Vincent van Gogh, combine mimesis with abstraction, and cannot be adequately
couched by one or the other. Van Gogh’s Starry Night remains mimetic by visually
depicting the rooftops of a city and the night sky above it, yet that image is also severely
abstracted by simplifying many details and adding the expressive detail of van Gogh’s
prominent brushstrokes, features that connote a wide range of emotional and affective
meaning.

91
The digitally animated nature of videogames—as well as the illustration of
cartoons and comic books—allows the same degree of plasticity. Much like painting, the
signifying power of animation and illustration is in its ability to offer more than merely
showing or telling—more than simply perceptual or conceptual signification—instead
offering a manner of presentation that combines the two based on the artist’s use of the
pliant materials at hand. Yet few hold animated media in the same esteem. In fact,
many refuse to even categorize them as “arts” at all (as illustrated by Roger Ebert’s
quote in my introduction). Despite a long history of “legitimate” art that combines
representation and abstraction, there seems a continuing aesthetic prejudice against
such a hybrid when used in temporal media. Although we regularly accept the abstract
representations of Monet and van Gogh as esteemed works of art, as well as those used
in sculpture, mosaic, and other brands of “museum art,” we do not associate
animation—of any sort—with such work.
As we saw Seymour Chatman distinguish earlier in his definition of texts, painting
and sculpture differ from discursive modes such as cinema, theater, literature, comics,
and videogames in that we receive them “all at once.” They are artifacts in the strictest
sense, without a temporal dimension to their reception, and resultantly, we hold them to
different standards than temporal media.
80
Why the difference in standards? Perhaps
the temporal dimension is so striking that we are unable to compare the two. Perhaps
the relatively late stage in history in which we finally achieved moving illustrations—long
after the plastic and temporal arts had been established as separate domains—has left
animation as a bastard child of sorts, not belonging in either family and thus left out in
the cold. Or perhaps we simply don’t know what to do with such hybrid work unless a

80
Even novels and comics, while not providing as fixed a pace as cinema or theater, remain
temporal as they are received “in time” rather than “all at once.”

92
learned professional has put it somewhere clearly demarcated as a space of art—a
gallery or museum—so that we can be assured of its legitimacy. Whatever the reason,
there is a long-standing tradition that expects mimesis and diegesis—both narratological
terms, and thus temporal by nature—to be kept separate, maintaining some sense of
aesthetic “purity.” Perceptual signification via realist depiction and conceptual
signification via language are seen as disparate, hallowed modes which become
tarnished when they begin to intermingle.
Part of the problem is that we often think of attempts to create a one-to-one
correspondence between the object and the representation—to attempt a strict imitation
rather than a stylized or abstracted re-presentation—as “straight” representation. To
depict something literally is to depict it “in the normal way,” while to depict it abstractly or
in some formalized manner is something else. Keep in mind that language is a manner
of communication with which we are so innately accustomed (as few of us can even
remember the time before we communicated through language) that we generally don’t
think of it in terms of representation or abstraction at all. Instead, we think of language
as something else entirely, a manner of conveying information that has nothing to do
with what we perceive via the senses, while simultaneously remaining just as strict as
mimesis in its signification (as we assume words to simply “mean what they mean”).
When we remove language from the equation, abstraction suddenly becomes a form
that is largely alien to the average adult audience, instead reserved for educated
aesthetes (those “in-the-know”) or children and the ignorant (those who don’t know any
better).
This bias is not always limited to the temporal arts—the Renaissance was
infatuated with realism (a prominent reason that we associate it with “classical”
representation) in a way similar to Aristotle’s own enamor with the mimetic tradition, and

93
even today, we can still observe the somewhat stereotyped reaction to modern art, when
frustrated museum-goers stare incredulously at inexplicable lines, shapes, and colors
and exclaim, “This garbage is art?” Even in the art world—amongst those “in the
know”—abstracted styles have often taken years, if not generations, to gain widespread
acceptance. Certainly, we should glorify the mimetic skill of photorealistic paintings, but
what aesthetic worth could nonsensical shapes and brushstrokes carry? This trend is all
the more prominent in temporal media, however, most evinced in the relegation of
animation, comic books, and videogames to the cultural basement of “children’s media.”
Adults, of course, partake of the skilled realism of cinema, theater, and the like (or the
purely conceptual discourse of literature), but who would bother with the visual simplicity
of animation other than children and the ignorant? Videogame culture even shares in
the bias, as “hardcore” fanboys fetishize the improved realism of this month’s new
releases while disparaging the lackluster graphics of “last-gen” systems. Serious
gamers only bother with the superior photorealism of the latest technology, because
after all, who would bother with the rudimentary (read: more abstracted) visuals, physics,
and A.I. of last year’s titles? Everywhere we look, mimesis seems to carry favor over
abstract representations, with mimesis repeatedly associated with maturity, skill, and
“serious” concerns, while abstraction is linked to children and childishness.
Meanwhile, literary art has gone in the opposite direction by insisting on the pure
abstraction of language alone, a trend lamented by comics champion Scott McCloud.
As children, our first books had pictures galore and very few words because that
was ”easier.” Then, as we grew, we were expected to graduate to books with
much more text and only occasional pictures, and finally to arrive at “real”
books—those with no pictures at all.
81

81
Scott McCloud, Understanding Comics (New York: HarperCollins Publishers, 1993), 140.

94
McCloud further notes that “this widespread feeling that the combination [of words and
images] is somehow base or simplistic has become a self-fulfilling prophecy.”
82
With the
assumption that mimesis loses legitimacy when it becomes abstracted (as in animation),
and that diegesis loses legitimacy when it becomes mimetic (as in comic books), in-
between media has become largely catered toward children, and alas, the stereotype
continues.
Why, then, would one choose to utilize abstraction if mimetic means are
available? What value could abstraction carry, in the face of such stiff prejudice?
McCloud suggests a couple answers. In Understanding Comics, his insightful
examination of comics as a legitimate art, McCloud discusses abstraction as a process
of “amplification through simplification.”
When we abstract an image through cartooning, we’re not so much eliminating
details as we are focusing on specific details. By stripping down an image to its
essential “meaning,” an artist can amplify that meaning in a way that realistic art
can’t. …The ability of cartoons to focus our attention on an idea is, I think, an
important part of their special power.
83

In other words: a picture may speak a thousand words, but what if we only want to speak
ten? Those other 990 words would only serve to distract from the point, so abstraction
pares away those frivolous details. Similarly, symbols such as a peace sign or the
American flag hold their value precisely because they isolate a clear and singular
referent—as opposed to, say, a photo of war protesters or of the continental United
States, both of which could signify the same ideas, but would also convey a wide range
of superfluous information that might muddy the point. Language only serves to further
isolate the details that an author or speaker wishes to present.
This vast range of signifying potential holds unlimited value to a designer
conscious of its possibilities, but we’ll use just one as example. Consider the difference

82
Ibid, 141.
83
Ibid, 30-31.

95
in negotiating a cartoonishly abstracted videogame world, such as that of Super Mario
World, as opposed to a more mimetic one, such as Half-Life. Due to the severely
reductive nature of Super Mario World, we can readily distinguish power-ups such as the
mega-mushroom. With a relatively limited number of perceptual details vying for the
player’s attention, useable objects stand out and are nearly impossible to miss, allowing
players to focus solely on the actions at hand, knowing that vital objects will always grab
their attention. In Half-Life, on the other hand, useable objects are made to easily blend
into the mimetic environment, forcing the player to pay closer attention to the details of
their surroundings; there may be some ammunition on a shelf, but it’s not going to jump
out and announce itself. Videogames need not choose one over the other, mind you—
Metal Gear Solid demonstrates a merger of these techniques by providing a highly
mimetic world, yet making useable items, such as the game’s ammo boxes, relatively
abstract (they float and spin in mid-air) so that they stand out and remain easy to find.
McCloud also notes universality as a further value to abstraction, explaining that
a photorealistic face could be said to describe one and only one person, while an
extremely abstracted face—two simple dots for eyes and a line for a mouth—could be
said to describe virtually anyone. The reductive features of cartoon figures allow them to
escape the specificity of representing one unique human being, and as a result, can
make them easier for us to identify with. The value of such a technique becomes
especially obvious with avatars—rather than offer a highly detailed, photorealistic figure
that the player can readily identify as someone other than themself, videogames might
choose to provide a more abstracted avatar that readily accommodates a wide audience
of players—would players accept the identity of Super Mario if, rather than a simplistic
cartoon of a plumber, the game offered a more photorealistic fellow? Yet such
universalizing is hardly limited to avatars—a racing game might abstract city streets to

96
embody “generic beach town” rather than a more specific San Diego, CA; a war game
might abstract its battlefield and enemy soldiers so as not to paint any particular nation
as “the bad guys”; or a sports game might abstract its teams and players so as to remain
relevant beyond one single sports season.
A further reason for abstraction flows from our tendency to associate mimesis
with realism. While significantly different terms, these concepts cannot be divorced—
mimetic painting is applauded specifically for its relation to the real world; mimetic
narratives carry their power because we associate their characters with real people; and
mimesis is one of the primary factors in nearly every mode of realism. If the mimetic
style aims to imitate reality, we can hardly divide it from the expectations of realism or
realistic-ness. One value to abstraction, then, is in its ability to move away from such
expectations.
This is perhaps most obvious in comparing live-action filmmaking with animation.
While live-action generally demands some adherence to logic (even in the most
ludicrous, slapstick, or fantastic of contexts), animation can get away with an exceptional
range of illogicality—audiences simply don’t hold the same expectations of abstract
drawings as they do mimetic photography. Consider any of the absurdist plots that have
been featured on the animated series The Simpsons or South Park and ask: would
these situations work in any live-action sitcom? This has often been the problem that
plagues live-action adaptations of cartoons, comic books, and indeed, videogames—the
fantastic premises of such properties often beg for abstract representation, and
resultantly flounder when forced into mimetic contexts (not surprisingly, such films are
often degraded as “cartoonish”). The outright “creepiness” that many find in
photorealistic cartoons (the live-action/CGI adaptations of Scooby Doo, Garfield, Alvin
and the Chipmunks, and countless others) equally results from this confounding of

97
expectations, placing animated characters and narratives in photorealistic contexts that
bring with them significantly different expectations.
84
Similarly, abstract videogames
carry with them different expectations than overtly mimetic ones—the cartoonish Mario
Kart 64 affords a different manner of gameplay than the heavily realist Gran Turismo,
despite both offering automotive racing, and similarly, American McGee’s Bad Day L.A.
provides a drastically different experience than fellow shooters Call of Duty or Half-Life.
EA Sports has recognized this very phenomenon in the “EA Big” lineup of games, using
titles like NFL Street and NBA Street to showcase outlandish gameplay that would not fit
within the realist demands of the Madden NFL and NBA Live franchises.
85

Of course, none of these techniques—sensory imitation, icons, symbols,
language, nor connotations—are anything new. Innumerable other media have utilized
such strategies to assorted ends and in varying degrees, and certainly videogames can
learn a lot from that history. Yet this is a work in medium-specificity, so our concern is
not only for what videogames can learn from the past. And while videogames do indeed
provide much of their meaning via the representations we see, hear, and feel, they are
also able to deploy such signifiers in a new manner that has very little history in the arts,
and it is here that we ought to address the value and significance of the medium.

84
This is especially evident in the phenomenon of the “uncanny valley,” a hypothesis originally
developed by roboticist Masahiro Mori. Mori noted that our emotional attachment to human
imitations (he addressed robots, but game designers frequently apply this to digital animation)
typically becomes stronger as the imitation becomes more life-like, and is strongest in those
imitations that are entirely life-like, however he noted that there is a point at which the imitation
becomes almost-but-not-quite human that suddenly shifts to repulsion. He named this sudden
dip in positive response the uncanny valley. Masahiro Mori, “The Uncanny Valley,” Trans. K. F.
MacDorman & T. Minato, Energy, 7:4 (1970), 33–35.
85
This is not to say that abstraction cannot address reality—The Simpsons and South Park (both
simplistically animated programs) have tackled a wide range of “real world” issues in extremely
abstracted contexts during their long-standing popularity. These shows find their success in
satirizing contemporary society, yet their animated styles allow them to do so in manners that few
other programs are able to achieve. In literary narratives, fairy tales have accomplished a similar
feat by commenting on real world problems, issues, and moral dilemmas via abstracted,
metaphorical contexts (and not surprisingly, we associate these stories with children as well).
Such texts still address the world, they just do so in a different fashion, as “less realistic” does not
necessarily equate with “less relevant to the real world.”

98
Actions as Signifiers
Imagine that I tell you about a particular creative work that has been engaged by
millions upon millions of people all over the world. Many people interpret it differently—
as a metaphor for life, as an inaccessible fantasy, as emblematic of contemporary
existence—yet although inviting a wide variety of interpretations, in truth, none of these
are right nor wrong. It is a work that haunts people’s dreams and that changes the way
they look at the world, a work that people find themselves coming back to again and
again. And though different people have different opinions or reactions to it, there
seems some universal quality about this work that speaks to everyone familiar with it in
some powerful way. Sounds like we ought to put it in a museum, doesn’t it? I’m talking,
of course, about Tetris.
Over the years, a variety of players and critics have offered their own “reading” of
Tetris. Some say that Tetris is a metaphor for life and death: the player works and
strives to stay alive, continually struggling to avoid the death that occurs once the blocks
reach the top of the screen, yet as time goes by, staying alive only becomes more and
more difficult, as the inevitability of death creeps closer and closer, since Tetris offers no
win-condition. Playing Tetris, then, is merely a struggle to stave off death for as long as
possible, where the point is not winning or losing, but rather the experience that
precedes the end. Others instead read Tetris as an obsessive-compulsive fantasy
unattainable in our real lives, offering a satisfying perfection where everything fits
together, as suggested by James Paul Gee:
Tetris…models one of our deepest human desires: to solve problems by finding
patterns inside a safe world in which there is a clear and comforting underlying
order. …Tetris is an escape into the very desire for order, control, and workable
solutions that we have all the time in our lives, a desire often frustrated in life, but
never in Tetris.
86

86
James Paul Gee, Why Video Games are Good for Your Soul: Pleasure and Learning (Urbana,
IL: Common Ground: 2005), 15.

99

Our real lives may remain messy, imperfect, and haphazard, yet Tetris allows the
satisfaction of seeing everything fall perfectly into place. Perhaps most famous,
however, is Janet Murray’s oft-quoted reading of Tetris as a metaphor for workaday
monotony:
Instead of keeping what you build, …in Tetris everything you bring to a shapely
completion is swept away from you. Success means just being able to keep up
with the flow. The game is a perfect enactment of the over-tasked lives of
Americans in the 1990s—of the constant bombardment of tasks that demand our
attention and that we must somehow fit into our overcrowded schedules and
clear off our desks in order to make room for the next onslaught.
87

Of course, none of these readings are the “correct” one, and designer Alexey Pajitnov
quite likely did not intend any particular meaning to his most famous work, yet even
those who do not annunciate their own metaphorical reading of Tetris still find that it
speaks to them in some powerful way. Many have had the experience of playing Tetris
for hours on end, absorbed into repeating the simplistic task of maneuvering blocks into
place over and over and over again and later wondering where those lost hours went,
yet even more telling is the phenomenon of “Tetris dreams,” an occurrence with which
many of us are likely familiar, and which has been reported on by psychologists at
Harvard Medical School.
Robert Stickgold asked people to play Tetris for 7 hours over 3 days. On the
nights after playing Tetris they reported seeing Tetris shapes floating around.
This even happened in people with severe amnesia who could not remember
playing Tetris. “I see images that are turned on their side. I don’t know where
they are from. I wish I could remember, but they are like blocks.”
88

This is not a minority phenomenon, mind you—Stickgold found that 75% of those in his
study had Tetris dreams.
89
Many others report waking Tetris moments, as they look at

87
Janet Murray, Hamlet on the Holodeck (New York: The Free Press, 1997), 144.
88
Christopher D. Frith, Making Up the Mind: How the Brain Creates Our Mental World (Boston:
Blackwell Publishing: 2007), 54.
89
Mike Cardwell and Cara Flanigan, Psychology A2: The Complete Companion (Cheltenham,
UK: Nelson Thornes: 2003), 81.

100
geometrical elements in the world around them and suddenly envision a Tetris
landscape that they want to arrange in some Tetris-like manner.
In other words: in some way or another, despite its mode of signification seeming
incomparable to our traditional notions of art and media, Tetris undoubtedly means
something to the people who play it. It is a powerful and deeply affecting work, one that
seems timeless and transcendent in its ability to captivate audiences. The aesthetic
power of this work, however, does not lie in its stunning visual qualities (its aesthetics
are fairly rudimentary, in fact). Its power is not in its ability to weave an intricate story,
develop characters, depict institutions, or speak to moral dilemmas (as the game
undoubtedly does none of these). Its power is not tied to its cultural origins (coming out
of the late-Cold War Soviet Union) or the haunting melody of its ubiquitous theme song
(though you’re probably humming it to yourself right now). Instead, Tetris’ power lies in
a manner of signification that it shares with very few other forms of art, and it is the very
thing that distinguishes videogames from other popular forms: what you do.
In no other medium is the activity of the audience so vital to the experience of the
text. A scant few “old media” texts have employed this notion: Alfred Hitchcock famously
utilized the spectator situation—what the viewer is “doing” in a film—in Rear Window and
Vertigo, while Italo Calvino references the reader’s act of reading in his book If on a
Winter’s Night a Traveler. But even these instances are rather minimal when compared
with the “doing” involved in a videogame. Brenda Laurel observes that the computer’s
most interesting capability lies “in its capacity to represent action in which humans can
participate,” affording the audience an ability to act as agents within a representational
context.
90
And I doubt this comes as an earth-shattering revelation to anyone—the

90
Brenda Laurel, Computers as Theater (Reading, MA: Addison-Wesley Publishing Company,
1993), 1, 21.

101
audience’s activity and participation is the very thing that separates videogames from
most of its media brethren.
Actions most certainly have signifying power, and we can divide them amongst
our five categories of signifiers just as readily as images, sounds, sensations, or
language. Actions can be mimetic, imitating real world actions, such as those performed
in Wii Sports. Using Nintendo’s Wiimote controller, the player of Wii Sports imitates the
swinging of a golf club, tennis racket, or baseball bat, the rolling of a bowling ball, or the
punches of a boxing match, all actions that are translated on screen in a relatively
mimetic fashion. Actions can be iconic, patterned after real world actions in a more
abstracted manner, such as the control scheme of Skate. Here, the player toggles their
controller’s analog sticks in a manner that mimics the footing of skateboarding, and while
it may not be as mimetic as Wii Sports, numerous skateboarders have commented on
the control scheme’s striking faithfulness when compared with other “button-mashing”
skateboarding games. Actions can be symbolic, as the actions in Puzzle Quest.
Against a Tolkeinesque role-playing context and donning the guise of a knight, wizard,
warrior, or druid, the player of Puzzle Quest moves two-dimensional jewels around an
8x8 gameboard to signify one-on-one combat with ogres, zombies, and dragons, making
the actions of gameplay entirely arbitrary to those it signifies. Actions can be linguistic,
as in text-based games like Zork, or point-and-click games like Maniac Mansion. In the
former, players merely type the actions they wish to perform (“use key,” “open door”),
while in the latter, players click on words like “use” or “open” to indicate their desired
actions, both requiring the player to annunciate their actions—to say “this is what I’m
going to do”—rather than perform them in any sense. And lastly, actions can be
connotative, as demonstrated in the powerful yet ambiguous manner in which the
actions of Tetris speak to the game’s players. Much like the shapes and colors of

102
modern art or the melodies of instrumental music speak to their audience via some
implicit yet uncertain correlation to emotions, ideas, and affect, so too do the actions of
many videogames.
Discussing actions as signifiers is not nearly this simple, however. For one, we
need look at both the literal activity of the player as well as the conceptual activity of the
player. The literal pressing of the X button on a Playstation controller to make an avatar
jump is an entirely symbolic action—pressing X has absolutely no relation to jumping,
thus the connection between the two is arbitrary.
91
When compared with such imitative
controls as pulling an Xbox controller’s trigger button to fire a weapon, swinging a
Wiimote to swing a golf club, or toggling Skate’s analog sticks to imitate skateboard
footing, pressing X seems completely abstracted. Yet the player learns this arbitrary
signifier in much the same way as learning language—through usage, it becomes an
intuitive behavior that the player can deploy without conscious separation between the
signifying action (pressing X) and the signified action (jumping), and once such symbolic
actions are learned, players can deploy them in complex combinations. At this point, the
player conceives of their activity as jumping, even though they are merely pressing
buttons, just as we often conflate words with their signified meanings. Once the
language is learned, ideally, the player’s communication with the system becomes a
seamless and intuitive act, and resultantly the player conceives of their activity as the
very act they perceive on screen. Compare this with the combat system of Puzzle
Quest, where the player perceives their activity as manipulating jewels and other icons
around an abstract gameboard as a symbolic representation of combat—the action seen
on screen is entirely abstracted, and combat itself is referenced only in a conceptual

91
Though, in much the same manner that Adventure’s square is mimetic only in that it “exists
somewhere,” we might say this action is mimetic only in that the player “does something.”

103
manner. Yet when played on the Nintendo DS, the literal action of Puzzle Quest is
actually highly mimetic—the player uses the DS touchscreen to move game pieces with
the console’s stylus, simply dragging icons where they want them to move, directly
imitating the very action they wish to perform. Thus while a Playstation game like Tomb
Raider provides actions which are literally symbolic (pressing X) but conceptually
mimetic (I’m jumping!), the actions of Puzzle Quest are literally mimetic (of one activity:
moving jewels) but conceptually symbolic (of a different activity: fighting).
We are often disinclined to consider actions in a semiotic or discursive manner,
however, because they do not fit our traditional notions of “content.” When thinking of
what a text means, our proclivity is to look right to its concrete elements—those things
that we can see or hear, and that outwardly and explicitly denote some signified referent.
It is for this reason that so many videogames get a bad rap—those who wish to criticize
the medium rarely have any interest in playing them, and thus have no familiarity with
the core gameplay structures from which the text’s purpose and significance arises.
Instead, many prefer to judge and label videogames based solely on surface
representation—the immediate and obvious characteristics that appear upon cursory
examination. Thus, they see cartoonish images and label the text childish; they see
violent images and label the text sadistic; they see popular culture images and label the
text trivial; or they see abstract images and label the text meaningless.
92

We don’t think of videogame actions as textual properties, primarily because the
audience’s activity is irrelevant in our traditional notion of a text (or at least, irrelevant to
the difference between texts—every film involves spectating, every book involves
reading, every song involves listening, etc.). Instead, we customarily associate actions

92
This is by no means to suggest that there are not childish, violent, trivial, or meaningless
videogames out there—as in any other medium, there is indeed plenty of assembly-line trash
being cranked out for a quick buck.

104
with another manner of recreation and entertainment: games. Which provides us a
convenient segue to the last part of this chapter.
Semiotics mark the beginning of discourse—all formal concerns of the text
emerge from here. There are plenty of other concerns to consider in a text—style,
aesthetics, conventions, and loads of medium-specific properties—yet these all continue
the semiotic concern of signification, of conveying information to the audience so that the
text means something. Regardless of medium or form, regardless of intent, regardless
of artist or time period, the core of any text, at its most fundamental, is to convey
information. It may express, it may argue, it may narrate, it may document, it may
explain, it may depict, or it may it may educate, and it may do so in any number of ways
via any number of conventional or creative decisions, but in one way or another, every
text signifies—it conveys information by making reference to something outside of itself.
Yet where does the art of a game lie? If the art of a text lies in the conveyance of
information, then where do we look to discuss the value and purpose of a game?
Games of all sorts quite often do convey external meaning—but certainly not all. What
does tic-tac-toe signify outside of itself? How about Soccer? Or Connect Four? While
signification does often occur in games, it remains only a secondary concern, one that is
by no means requisite. If we want to consider where the art of a game lies, we must
instead look to its own fundamental purpose: play.

Systems of Rules and Spaces of Possibility: Play and the Aesthetics of Emergent
Space
In the beginning, Nolan Bushnell created Pong, the second coin-operated video game
(after creating Computer Space, the first coin-operated video game), and placed it in
Andy Capp’s Tavern in Sunnyvale, California. It grew gravid with quarters in less than a
week, and Nolan Bushnell saw that it was good.
93

93
D.B. Weiss, Lucky Wander Boy (New York: Plume, 2003), 132.

105

As this excerpt from D.B. Weiss’ Lucky Wander Boy suggests, the origins of Pong—and
by proxy, the videogame industry—are by now the stuff of legend. And as in all legends,
the details of the true story are sketchy at best. But as the traditional account goes,
Nolan Bushnell founded a fledgling company named Atari, hired a young programmer
named Al Alcorn, and—as a test of his programming chops, before assigning a more
complicated chore—asked Alcorn to devise a rudimentary ping-pong game. They
named it Pong, and after finding little interest in it from pinball manufacturers, decided to
give their prototype a test-run in a Sunnyvale, California tavern.
One day, soon after Pong’s arrival, the tavern’s manager arrived to open the bar.
He found several patrons waiting in line outside. They came for Pong.
Shortly thereafter, the game stopped working. The disgruntled manager gave
Alcorn a late-night call, and Alcorn arrived to find treasure inside. Not gold, mind you,
but silver: patrons had pumped so many quarters into the game that the jury-rigged coin
receptacle had overflowed, jamming the machine’s inner workings. Pong was a hit.
People liked videogames. So designers started making more, and an industry was born.
Now, Pong is incredibly abstract—aside from a very basic rendition of deflection
physics, it doesn’t represent much. And Pong is incredibly simple, entailing only two
paddles bouncing a ball. And in the end, Pong is—by most definitions, anyway—
incredibly meaningless. Or at least, it does not mean in the way that we typically expect
an object of entertainment to be compelling. So why is it so compelling? Why did
players line up outside Andy Capp’s Tavern to willingly part with their quarters, and why,
thirty-five years later, did they do the same with $600 for the Playstation 3? Which is to
ask: why do we play games?

106
There is extensive research on why we play—it expends excess physical and
mental energy, it contributes to both learning and leisure, and it arguably sparks much of
art and culture. But why play games? What is it about formalized play—play structured
by a system of rules, play given formal organization—that we find so engaging? And
how does such play manifest in videogames?
We need look first at play itself. But as it ends up, “play” is a rather complicated
word.
"Play nice, now."
"Jim plays the saxophone."
"How do you think this story will play in the city?"
"The show plays at 9:00."
"Just play along."
"Stop playing for time!"
"Let's play it by ear."
"I love the way the light plays off the water."
"She always plays the fool."
"Wanna play cops and robbers?"
"Give the rope a little more play."
“You've played right into my hands!"
“What a boneheaded play.”
"I can't wait to see how this plays out."
“He lost $200 playing the ponies.”
"I'm playing a hunch."
"If only she knew he's been playing around…"
"We should play up the political angle."
"I used to play fast and loose."
"Here is how the game is played…"

Webster’s Unabridged offers sixty-two different definitions to the word—twenty-one
intransitive verbs, twenty-four transitive ones, and seventeen nouns
94
—Dictionary.com
offers ninety-four,
95
and The American Heritage Dictionary provides a whopping one
hundred and nineteen different meanings.
96
A hundred and nineteen! And that’s not

94
Jean L. McKechnie ed., Webster’s New Universal Unabridged Dictionary, Deluxe Second
Edition (New York: Simon & Schuster, 1983), 1377.
95
"play," Dictionary.com Unabridged (v 1.1), Random House, Inc,
http://dictionary.reference.com/browse/play (accessed January 22, 2008).
96
"play," The American Heritage Dictionary of the English Language, Fourth Edition (Boston:
Houghton Mifflin Company, 2004), http://dictionary.reference.com/browse/play (accessed January
22, 2008).

107
even counting derivations like “playful” or compounds like “wordplay.” Yet among these
myriad variations on the word’s usage, we can discern four commonalities:
(1) An activity of recreation
(2) The quality of being fun, and not serious.
(3) An enactment or performance.
(4) Having freedom of movement, or looseness.
If we assume sports and gambling to be some manner of recreation (rather than a job or
a compulsion), then the vast majority of these definitions of play—roughly 90%, by my
count—fit at least one of these four categories, and even the few exceptions seem to
descend from metaphors and game analogies (“to cooperate” as in “she refused to play,”
or “an enterprise or venture” as in “an oil and drilling play”).
The first two traits are somewhat intertwined—one is the generic activity of play,
the other the attitude or spirit that generally accompanies it; one is an abstract state of
being, the other its manifestation. In the first sense, play is essentially “not work,” an
activity performed for the experience and enjoyment it produces rather than for profit or
productivity, an expenditure of physical and/or mental energy directed toward pleasure
or amusement. In the second, play is “not serious,” a state of being that is associated
with fun, glee, jest, triviality, or silliness, rather than one’s everyday concerns. In both,
play is consciously and defiantly outside of “real life” and separate from things that
matter—not surprisingly, we typically associate both with escapism and diversion. And
these are the two traits that we most readily associate with the playing of a game—
unless a professional athlete or compulsive gambler, we generally assume the playing of
a game to be an activity of recreation and to have some quality of fun.
The latter two, on the other hand, we may be more inclined to consider as
applying to other contexts. As an enactment or performance, play is most often used in

108
the context of drama and music—to play a role, to play a melody, to play an instrument,
or to put on a play. Alternately, we might use the term to reference the taking on of a
persona (“to play the role of big brother”) or to don false characteristics (“to play dumb”).
As freedom of movement or looseness, we generally attribute play as a particular quality
at work in machinery (the wiggle-room between parts, as in the play of a gear), a
physical property of slack or flexibility (the play of a rope), or a characteristic of
haphazard or organic motion (the play of light or water).
Yet we should not simply dismiss the latter two. Indeed, much of play involves
some manner of enactment or performance, from the play of make-believe (“playing
pretend”) to showing off (a playful manner of performance). Equally, freedom of
movement is essential to play—when children run about the yard, explore a creekbed, or
fiddle with toys, such freedom is the dominant characteristic of their play. It is the
looseness with which they operate that distinguishes their play from other, more
formalized activities such as chores, school, or ceremony.
And yet, games are, by definition, a formalized activity. This is the very
characteristic that distinguishes freeform play from gameplay. When we address
videogames as games, we are addressing them as systems of rules, and considering
the manner in which their formal properties shape, structure, and organize a particular
type of activity: play. Yet how do rules give rise to play? What does it mean to “enact” a
system of rules, and how do we address a “freedom of movement” in such rigidly
structured activities? For one reason or another, play seems infinitely more compelling
when it is contained by rules, when it is couched within the formal structure of a game.
Bouncing a ball is fun; playing basketball is much more fun. There is a strange and
fundamental relationship between rules and play, and we are here concerned with why
exactly that is.

109
To find out, we once again return to the organizing principle of this chapter: the
system of a videogame. Rather than considering gameplay culturally, sociologically,
psychologically, or aesthetically, we can find the answers by considering it
systemically—as the product of a formal system, the result of interactions occurring
within a systemic context. Our answers lie in a concept that originates far outside of any
concern for play or games, instead coming to us from systems theory. Our answers lie
in emergence.
A concept that descends from both philosophy and the sciences, emergence
results when complex behavioral patterns result from relatively simple interactions
(which, in and of themselves, are not directed toward such patterns). Ant colonies,
economic systems, social organization, urban growth, genetic evolution, weather
systems, cellular behavior, collective intelligence, and the human brain—not to mention
certain types of computer software—are all common examples of systems that provide
emergent phenomena. Each exhibits “bottom-up” behavior, as higher-order organization
“emerges” from lower-order processes; resultantly, emergence is a classic case in which
the proverbial whole is greater than the sum of its parts. Per emergence theorist John
Holland:
Emergence is above all a product of coupled, context-dependent interactions.
Technically these interactions, and the resulting system, are nonlinear: The
behavior of the overall system cannot be obtained by summing the behaviors of
its constitutive parts.
97

Emergent systems provide behaviors which are at once organized and yet lacking a
comprehensive organization; they lack an authorial hand or a universal intelligence,
instead manifesting from the complex web of interactions between their component
parts, without an organizing principle. While we can account for the components that
generate these behaviors (ants in a colony, citizens in a city, lines of codes in a

97
John Holland, Emergence: From Chaos To Order (New York: Basic Books, 1998), 121-2.

110
program), these individual parts cannot explain the entirety that they combine to create.
Neither the behavior of an ant colony, the workings of the brain, nor the patterns of a
weather system can be fully explained via their individual elements, and similarly, neither
basketball, chess, nor Civilization can be reduced to their respective rules.
The 2007-2008 Official Rulebook of the National Basketball Association is a
lengthy 62-page document, organized by topic and indexed for quick reference, that
diagrams and explains the features of the court, defines terminology and hand signals,
clarifies the roles of players, coaches, and officials, designates the procedures of play,
elaborates on all violations and penalties, and even describes such trivial details as
proper uniform standards. Now imagine that you read this entire rulebook, having never
played basketball, seen a game of basketball, or even heard of this thing called
basketball. After reading the rules, do you think you’d have a very good impression of
what basketball is? The rules, after all, are the very system of the game, but are these
enough to create a familiarity with it? The rules, for example, clearly delineate all the
procedures for handling the ball and scoring, yet make no mention of the numerous
strategies that teams often employ—the pick-and-roll, screen passes, zone defense, and
such. The rules also clarify that each team must have five players on the court, yet
never describes the five positions that are now standard to the game—a point guard,
shooting guard, small forward, power forward, and center. While generally considered
part-and-parcel to basketball, neither these plays nor these positions are actually
inherent to the game as a formal rule-set.
The rules are the closest thing to authorial control that we can find in a game, the
only place in which we can locate a unified organizing principle—and yet, the rules alone
don’t really dictate the nature of the game nor the behaviors that occur within it. Just as
understanding the individual behaviors of ants will not give us an understanding of an

111
ant colony and understanding brain cells will not give us an understanding of the human
brain, understanding the rules of a game—the procedures which players must follow—
will not give us a proper picture of the game. It is only by observing the system in
operation, by seeing players interacting and the game at work (or better, at play) that we
get a sense for the system itself. We can only come to know the system of rules by
seeing those rules enacted.
Now imagine that you play one game of basketball—would you now have a
complete picture of what basketball is? You’d certainly have a better familiarity with it,
but would hardly be a veteran of the game. If you now play a second game against a
different team, that game would hardly be predictable based on your prior experience, as
different teams employ different tactics to utilize different strengths. So you play another
game, then another, and with each playing, your familiarity increases, and you get a
better sense of what basketball is. Yet no matter how many times you play the game, or
watch others play, you’ll never be able to make more than an educated guess as to what
will occur, as any individual match—any one derivative of the rule-set—will manifest
from the unpredictable, and yet rule-based, interactions of ten players on the court.
It is for this reason that I described games as a derivative form—we cannot come
to know a game simply by knowing its rules, but instead glean an understanding of the
system by seeing instances of how the rules play out. And each iteration of the game
emerges from players interacting within that system of rules. The game as a whole—the
thing itself—becomes an abstract sense of all these iterations combined, the large-scale
patterns that emerge from repeated playings (the pick-and-roll, the seven-foot-tall
center) as well as the unique aberrations that surprise even the most experienced
veteran (the last-second three-pointer from midcourt, the five-foot runt with a wicked

112
jump shot). In short, the game of basketball is much more than merely the prescriptions
of its rules, and it is in this quality that the game finds its emergence.
Writing in the inaugural issue of the journal Emergence, Jeffrey Goldstein
recognizes several commonalities of emergent behavior, and in the chart below, we see
that these six commonalities (as encapsulated by Peter Corning) can be readily applied
to basketball, or any other game:
“(1) radical novelty (features not
previously observed in the
system);…”
Any playing will feature a unique
configuration of actions, and remain
highly unpredictable, so that each
match is a novel one.

“(2) coherence or correlation
(meaning integrated wholes that
maintain themselves over some
period of time);…”
Each playing provides a coherent
whole—all actions are a part of the
game, so that no occurrence is
superfluous (if a player stopped to make
a phone call, on the other hand, this
would clearly be outside the game, and
not part of the playing).

“(3) A global or macro “level” (i.e.,
there is some property of
“wholeness”);…”
Every iteration becomes part of the
game as a whole, so that we can
observe the overall behavior of the
game proper beyond individual
instances.

“(4) it is the product of a dynamical
process (it evolves);…”
Every game is a continuous evolving
process, as any one moment in the
game is the result of all the interactions
to that point (that is, we never reach a
point where “everything that’s happened
so far doesn’t matter” unless we literally
start over).

“(5) it is “ostensive”—it can be
perceived;…”
We can observe individual matches as
example of the game.

(6) “[and] for good measure, …
supervenience—downward
causation.”
98

Resulting from the interactions of
individual players (rather than one
controlling intelligence), every game
provides bottom-up behavior.

98
Peter A. Corning, “The Re-Emergence of ‘Emergence’: A Venerable Concept in Search of a
Theory,” Complexity 7:6 (2002): 18-30,
http://www.complexsystems.org/publications/pdf/emergence3.pdf (accessed January 18, 2008).
Corning is paraphrasing: Jeffrey Goldstein, “Emergence as a Construct: History and Issues,”
Emergence 11 (1999), 49-72.

113
Emergence is a vast and immensely complicated notion—entire books have
been dedicated to its understanding—yet for our purposes, we might consider it best as
a manner of “organized complexity.” Between predictable order and disorganized
chaos, we find emergence—behaviors which are heterogeneous and nondeliberate, and
yet “show the essential features of organization.”
99
Steven Johnson positions
emergence as that which exists between the behavior of simple systems (those with one
or two variables) and disorganized complexity (those with billions of variables), and
demonstrates the notion with what he terms “The Myth of the Ant Queen.” The myth is
simple: despite her title, the queen of an ant colony exhibits no authority. The colony
has no central intelligence, and no hierarchy of decision-making. Instead, the vastly
complex organization of the colony emerges from thousands of ants interacting with one
another in relatively simple ways.
100

Yet it is a different example that Johnson employs which is of most interest to us:
gameplay.
[O]ne of the things that make all games so engaging to us is that they have rules.
In traditional games like Monopoly or go or chess, the fun of the game—the
play—is what happens when you explore the space of possibilities defined by the
rules. Without rules, you have something closer to pure improv theater, where
anything can happen at any time. Rules give games their structure, and without
that structure, there’s no game. …

This emphasis on rules might seem like the antithesis of…open-ended, organic
systems…, but nothing could be further from the truth. Emergent systems too
are rule-governed systems: their capacity for learning and growth and
experimentation derives from their adherence to low-level rules. …If any of these
systems—or, to put it more precisely, the agents that make up these systems—
suddenly started following their own rules, or doing away with rules altogether,
the system would stop working: there’d be no global intelligence, just a teeming
anarchy of isolated agents, a swarm without logic. Emergent behaviors, like
games, are all about living within the boundaries defined by rules, but also using
that space to create something greater than the sum of its parts.
101

99
Warren Weaver, “Science and Complexity,” American Scientist 36 (1948): 536. As cited by
Steven Johnson, Emergence: The Connected Lives of Ants, Brains, Cities, and Software (New
York: Scribner, 2001), 47.
100
Johnson, 47.
101
Ibid, 181.

114

Truth be told, Johnson’s exposition on gameplay is largely a metaphor—he’s attempting
to put the concept in laymen’s terms by using games as an analog for emergent
systems. Yet while demonstrating the ideas behind emergence, Johnson actually gives
us a fairly compelling account of the activity of gameplay and just what the rules of a
game provide. Using this account, we can identify at least five essential characteristics
of rules that shed light on their unique properties as formal systems, and the manner in
which they give rise to gameplay: (1) emergence results from agents operating within the
system of rules; (2) the system of rules creates a space of possibilities; (3) gameplay
appears paradoxical, (4) gameplay provides a capacity for learning, growth, and
experimentation, and lastly—an implicit deduction—that (5) gameplay is, in itself, an
emergent behavior. Let’s address each of these in turn.
(1) Emergence results from agents operating within the system of rules. There’s
a certain degree of ambiguity that often characterizes discussions of emergence in
games. Because emergence scholars often use gameplay as an analog for emergent
systems, while also addressing emergent software such as SimCity and The Sims—
programs that happen to be videogames—it becomes easy to confound these two, and
as a result, many accounts of emergence in games tend to be somewhat vague as to
where exactly the emergence comes from. Chess is a standard example: the game is
comprised of a mere ten or twelve rules, a relatively simple rule-set, yet within those
rules, we see a seemingly infinite number of possible configurations, while
simultaneously finding common patterns, strategies, and other regularities that emerge
in game after game (along with new strategies that continue to emerge after hundreds of
years of gameplay). Many credit the rule-set with such emergent behavior, suggesting
that emergence results from the rules working together in complex ways; this assumes

115
the rules as the “agents” that are interacting to create bottom-up characteristics. Yet
let’s rethink this notion. I don’t want to say that this attitude is wrong per se, but it does
sidestep a fundamental part of the system—the engine that drives it.
Nearly every instance of emergent gameplay hinges on one powerful
commonality—the human factor. By injecting a decision-making human—a massively
complex thinking machine—into the system, it immediately becomes infinitely more
dynamic. It’s easy to assume in Johnson’s example that he's using games to
demonstrate emergence, yet this is a mistake. He's not using games; he's using
gameplay. And this is an enormous difference. After all, if we simply programmed a
computer to make legal chess moves at random, those regularities would disappear, and
the game’s “organized complexity” would quickly fall back into predictable order (pawns
moving slowly forward) and utter chaos (queens haphazardly moving in every direction).
Despite the vast number of discussions on complexity in games, surprisingly few
explicitly address the human factor as allowing for its organization, instead vaguely
focusing only on the rules as generating such patterns. Yet Salen and Zimmerman, for
one, clarify the situation: “In games, emergence arises through the interaction of the
formal game system and decisions made by players.”
102
If we wish to pursue Johnson’s
metaphor literally, we need look beyond only the rules as creating emergence, and
equally consider the agents operating within those boundaries.
Emergent behavior is always driven by independent agents of some sort—ants in
an ant colony, citizens in a city, cells in a body, and micro-programs in artificial
intelligence all drive their respective systems, and those systems become emergent
based on their bottom-up behaviors. The component engines don’t deliberately create
the overall patterns; rather, unpredictable behaviors emerge from the interaction of these

102
Salen and Zimmerman, Rules of Play, 164. Italics are mine.

116
component parts that each have their own agendas. When we describe games as
emergent systems, what we’re really saying is that games are a space of possibility
negotiated by an intelligent agent. We call them emergent, then, because behaviors,
strategies, and other patterns of play emerge from the system that are unintended and
unanticipated by the designer.
(2) The system of rules creates a space of possibilities. It is precisely because a
game’s designers cannot intend or anticipate such behaviors that Salen and Zimmerman
describe game design as a “second-order design problem”:
As a game designer, you are never directly designing the behavior of your
players. Instead, you are only designing the rules of a system, …[as] it is not
always possible to anticipate how the rules will play out. …As a game designer,
you can never directly design play. You can only design the rules that give rise
to it. Game designers create experience, but only indirectly.
103

Here we see the conundrum of emergent design—of crafting a system which will
manifest in emergent iterations rather than a fixed instance of discourse. We already
noted that as players, we can never directly access the game itself, instead only
gleaning it through its derivative matches. The flip side of that postulate is that
designers can never have direct control over these derivatives, instead only shaping the
system of rules which give rise to them. Rather than designing behavior or experience,
a game’s designers instead create a bounded arena of sorts, what Johnson terms the
“space of possibilities defined by the rules.”
A game designer’s job is to create a space of possibility in which agents can
operate in some meaningful way—to provide a particular range of actions, while also
delimiting the entirety of all actions. If the player has little or no choice, we’d get
predictable behavior; if the player has too much choice, we’d get chaos. Instead,
emergence—organized complexity—occurs within a formal system when thinking agents

103
Salen and Zimmerman, Rules of Play, 168.

117
(players) operate within a particular range of possibilities (a rule-based context) while
pursuing their own agendas (an objective). All we need do is make that context artificial,
and we’d have the very definition of a game.
Johnson describes game designers as “control artists”: “they have a feel for that
middle ground between free will and the nursing home, for the thin line between too
much order and too little.” Such artists provide their audience a goal, then shape their
means for attaining that goal. “Anything goes” will not provide us a game; one linear
path will not provide us a game. Instead, it the happy medium between these two
extremes that mark the aesthetics of game design. It is this space of possibility in which
games find their art—and in which we can observe the contradiction of rule-based play.
(3) Gameplay appears paradoxical. When we address a game as a system of
rules, we’re discussing its formal identity—the formal structure that makes up the game
itself. But in describing rules as the “formal” properties of a game, we ought remember
this term’s numerous meanings: structural (providing the form), conventional (following
standard procedures or formalities), and official (not informal; not casual or flexible), as
well as logical (methodical: as in formal proof) and precise (strict and meticulous: a
formal report). And yet, the content of rules is “play,” an activity that we assume as
defiantly informal and a term whose meanings run contrary to formality: fun, free, silly,
not serious, and allowing for movement, flexibility, and experimentation. Johnson notes
that the rigidity of rules “might seem like the antithesis of open-ended, organic systems”
such as those which exhibit emergence, and similarly, Johan Huizinga describes games
as a somewhat incongruent affair: on the one hand, absolute, binding, fixed,

118
unambiguous, undeviating, and orderly, yet equally peculiar, uncertain, arbitrary, and to
some extent absurd.
104

Yet while the formal structure of rules may seem contradictory to the freedom
and creativity of play, Huizinga describes a “profound affinity between play and order,”
and suggests several reasons why rules are essential to the magic of play. For one,
rules are all that bound play from the outside world. A game is a closed system, a space
shut off from reality, and it is these absolute and explicit boundaries that offer such clear
separation from the rest of life (thus the scorn placed on those who break the rules, who
shatter these boundaries and allow the world back in). Equally, he sees the supreme
order of rules as a provisional utopia: “Into an imperfect world and into the confusion of
life it brings a temporary, a limited perfection.” And lastly, it is in this tension between
freedom and restraint, between creativity and control, that Huizinga identifies the
aesthetics of play. “It may be that this aesthetic factor is identical with the impulse to
create orderly form,” Huizinga claims, noting the terminology with which we might
describe rule-based play: “tension, poise, balance, contrast, variation, solution,
resolution, etc.”—all terms commonly applied to the arts.
105

Brenda Laurel observes a “mysterious and symbiotic” relationship between
creativity and constraints, and it is in this symbiosis that we may resolve our seeming
contradiction. Citing psychologist Rollo May, Laurel insists that limitations are not only
helpful to the creative act, but in fact necessary:
Creativity arises out of the tension between spontaneity and limitations, the latter
(like river banks) forcing the spontaneity into the various forms which are
essential to the work of art. …The significance of limits in art is seen most clearly
when we consider the question of form. Form provides the essential boundaries
and structure for the creative act.
106

104
Huizinga, 7-13. All these adjectives are culled from Huizinga’s description, aside from the
paraphrased “absurd.”
105
Ibid, 10.
106
Rollo May, The Courage to Create (New York: W.W. Norton, 1975). As cited by: Laurel, 101.

119

The painter stands in front of a canvas and is constrained by a rectangular frame, a two-
dimensional surface, and the paints and brushes at their disposal; the playwright
struggles against the restrictions of a stage and set housed under a theater’s roof, of the
limited practicalities when facing a live audience on one side of a proscenium arch. And
yet within these boundaries—within these spaces of possibility—such artists are able to
produce works of astounding imagination. Creativity comes not from their ability to do
anything they please; rather, it comes from their ability to do something inventive within
these constraints. When all is said and done, then, creativity is an emergent behavior.
Just as emergence results from the tension between order and chaos, so too creativity
results from the tension between spontaneity and limitations. The rules of a game
provide a similar dynamic.
Applying such rationale to digital environments (not just videogames, but any
such environment), Laurel insists that while users may think they want complete freedom
and infinite functionality in navigating an interactive environment, such freedom is never
a good idea (not even to the limits practically available). In part, this is because users
would then have no idea what to do:
A system in which people are encouraged to do whatever they want will probably
not produce pleasant experiences. When a person is asked to “be creative” with
no direction or constraints whatsoever, the result is…often a sense of
powerlessness or even complete paralysis of imagination. Limitations—
constraints that focus our creative efforts—paradoxically increase our imaginative
power by reducing the number of possibilities open to us.
107

By this reasoning, limitation becomes vital to a game by (counter-intuitively) enabling the
player to act. By limiting the means available, the game allows the player to recognize
the means available, to know what can and can’t be done. Further, a game’s constraints
allow the player to better identify causal relations within the system. While Huizinga

107
Laurel, 101.

120
values the closed systems of games for idyllic perfection, Brenda Laurel addresses this
dynamic in more practical terms: the bounded and ordered nature of such systems
prevent the causal relationships between events from being obscured “by the
randomness and noise characteristic of open systems,” and thus allow an “increased
potential for effective agency.”
108
Because games are closed off from the world outside
and rigidly structured by rules, we’re able to understand what goes on inside the system
of a game, and are resultantly better equipped to function within it. The benefits of
constraints, then, are actually twofold: they enable the player to act as well as to
understand their context.
(4) Gameplay provides a capacity for learning, growth, and experimentation.
These constraints do not only provide creativity and artistry, however—they also afford
learning via experimentation and discovery. Both games and emergent systems are “all
about living within the boundaries defined by rules,” Johnson insists, and “their capacity
for learning and growth and experimentation derive from their adherence to [such] rules.”
John Holland also stresses learning as a key component to emergence, and equally,
psychologists and sociologists have long recognized play as having intimate connections
to learning.
109
When we see children doing somersaults down a hillside, throwing rocks
in a puddle, or finding bugs under a rock, would we say they are playing with gravity,
water, and insects, or are that they learning about gravity, water, and insects? Is there
any way to separate the two? Play is an activity associated with experimentation,
improvisation, and exploration, and this comes from having movement within
boundaries. We don’t simply “explore” in play—we explore the edges, the limits, the

108
Ibid.
109
For a plethora of examples, see Play = Learning, a collection of thirteen essays on child
development and learning that detail the vital role that play serves in this regard. Dorothy G.
Singer, Roberta Michnic Golinkoff, and Kathy Hirsh-Pasek, Play = Learning: How Play Motivates
and Enhances Children's Cognitive and Social-Emotional Growth (Oxford University Press:
2006).

121
boundaries of the system. Children learn by exploring the world around them—they
discover its formal structure by playing within its “rules,” learn about the nature of things
by testing out their properties. Equally, the players of a game learn by exploring its
space of possibilities—by surveying its boundaries, by trying new techniques and
strategies, by a ludic process of trial and error.
We see this most clearly as players gain experience in playing a game. Upon
first playing, the player will likely not fare well, perhaps play again and fail, but as the
player explores the space of possibility defined by the rules, they begin to identify
strategies, tactics, and the allowances that the rules afford—they learn not only to play
the game, but to play the game well. And it is for just such reasons that the strategies
and positions of basketball have evolved over the years, despite existing nowhere within
the game’s rules. As players and coaches have explored the space of possibility offered
by basketball’s rules, they’ve collectively come to realize the pick-and-roll as an effective
offensive tactic, along with the benefits of a tall and powerful center positioned near the
basket and a quick and agile point guard operating on the periphery. It is here in which
we see the “global intelligence” to which Johnson refers: the individual agents all explore
and negotiate the space of possibilities in the pursuit of their own agendas, yet all
contributing to a larger pattern of gameplay that manifests on a macro-level. We see
similar patterns evolve in videogames—the best strategy for defeating a boss, the
quickest route for negotiating a gauntlet, or inventive tricks for avoiding an obstacle.
Further, James Paul Gee, a specialist in issues of learning and literacy, has argued that
learning is not only an inherent quality to videogames, but one of their fundamental
pleasures.
110

110
James Paul Gee, What Video Games Have to Teach Us About Learning and Literacy (New
York: Palgrave/Macmillan, 2003).

122
(5) Gameplay is, in itself, an emergent behavior. Ultimately, then—and implicit to
Johnson’s explication on game rules and emergence—all gameplay is, to one degree or
another, an emergent behavior. The phrase “emergent gameplay” has a certain amount
of redundancy to it—play that is overtly ordered would not be a game (such as
performing a stageplay), and play that is overtly chaotic would neither be a game (such
as running freely in a field).
111
Gameplay is in that space between—an activity that
involves the enacting of rules, yet also free movement within those boundaries. We do
indeed see varying degrees of emergence in videogames—some retain more deliberate
structure than others—yet all games (theoretically) have some degree of emergence in
the end.
It is for this reason that many are wont to include puzzles within gameplay.
Salen and Zimmerman provide an example of such an argument:
According to puzzle and game designer Scott Kim, puzzles are different from
games because puzzles have a correct answer or outcome. Think of a
crossword puzzle: the puzzle designer creates the correct answer, and the
player’s activity consists of trying to reconstruct that answer. This is a very
different situation than a game of Poker, for example, in which there is no fixed
“right answer” posed by the creator of the game. Instead, in Poker, players make
complex decisions at every moment, taking into account the evolving dynamics
of the game.
112

Puzzles typically provide little room for emergent behavior, as the player’s activity merely
consists of a process of deduction. It for similar reasons that some refer to hypertext
adventures—those in which one deliberate sequence of commands are necessary for
completion—as “interactive fiction” rather than games; they do not offer the pliability and
space of possibilities necessary of a game. Such works are essentially narrative
puzzles, and merely ask the player to figure out their solution. We see a similar

111
In comparison to the overt order of a stageplay, it’s telling that improvisational troops often
refer to their sketches—loosely structured by a context, rules, and audience suggestions but
lacking a predetermined script—as “games.”
112
Salen and Zimmerman, Rules of Play, 80.

123
uncertainty over linear games like Halo or Metal Gear Solid—while few dispute them as
games, many note something particularly “un-game-like” about the fixity of their design
and the linear progress necessary for victory. Jeffrey Goldstein counted a capacity for
evolution as one of the requisites of an emergent system, and equally, Salen and
Zimmerman cite the “evolving dynamics” of a game and the “complex decisions” made
by players as lying at the heart of this distinction. Whether puzzles or linear
videogames, the discounting of these examples are a reaction to their lack of emergent
potential—the activity is too ordered to allow for the freedom of movement requisite of
play, nor the evolving nature requisite of emergence.
While acknowledging the argument against them, Salen and Zimmerman
ultimately accept puzzles as games, and equally, I’m willing to accept linear videogames
as games nonetheless. Even in a crossword puzzle, we see some room for movement
in how players choose to solve the puzzle: start with the easy words first, or start with a
few big words to get the ball rolling, or start working on one corner of the puzzle and
work out from there. There may only be one correct end-state, but one can arrive at that
end-state via many routes. Equally, while a videogame may have a single correct
solution, there’s always room for experimentation en route to finding it; players can still
explore the boundaries, learn about the system, and experiment with plenty of incorrect
approaches. In distinguishing “linear” or “fixed” videogames, we’re addressing the
components of the design that are predictable—when playing Halo, we know that the
first level will always end with the Master Chief being jettisoned in an escape pod that
crash-lands on the alien space station of Halo, and that the Flood will always appear in
the same location at the same moment. Yet while we may be able to predict such
embedded elements, we cannot predict what players will do between and around these
fixed components, and how they will negotiate (or fail to negotiate) the obstacles

124
presented. Regardless, the argument against puzzles remains telling—puzzles and
linear videogames are, at the very least, edge-cases that push the boundaries of what
we might include in the category of games, and this distinction demonstrates the
necessity of emergence—variable iterations based on the interactions of players—to the
fabric of what comprises a game. This is, after all, the very quality that distinguishes
games from other formally structured leisure activities (such as watching a movie).
To polish off this digression on emergence and play, I’d like to conclude by
distinguishing three different manners in which the concept is typically applied to
videogames specifically. The term “emergence” gets bandied about somewhat willy-nilly
around games, and is resultantly applied in at least three different manners which are
often conflated. We can consider these as three general ways in which videogames
produce emergence.
(1) Unpredictable player behavior. Salen and Zimmerman have already made
this point for us—as a “second-order design problem,” games do not directly shape a
player’s behavior; the system of rules merely shapes the space of possibilities that gives
rise to gameplay, thus we can never accurately predict how a game will play out.
Certainly, the designers will constrain the potential ways that it may play out—delimit the
space of possibilities—yet the player will always have room to move within that space (if
they do not, we likely do not have a game on our hands). At its most mundane, this
unpredictability may merely lie in how a player navigates the linear gauntlet of Halo—
whether they charge into battle with guns blazing, carefully and deliberately eliminate
enemies one at a time, or try to avoid battle whenever possible. Or perhaps it will simply
be a matter of whether they win or lose a round of Mortal Kombat. In more interesting
examples, it may be in creative ways of utilizing or combining a game’s resources—
Civilization offers a wide variety of resources and actions, as well as numerous ways to

125
win, so that players can choose from countless strategies. Yet even in games using a
strictly linear design or a relatively simple game mechanic, players often find drastically
unpredictable ways to subvert the game’s intentions, and these are the behaviors
which—at the level of the individual player—receive the most attention as emergent.
Exploits are one such phenomenon: when players discover unforeseen
strategies by manipulating the game’s physics or utilizing technical glitches. Among the
most famous is the technique of “rocket-jumping,” pioneered in Quake but used in
numerous other shooters as well. By exploiting the concussion physics of the game’s
explosions, players of Quake have repurposed the game’s rocket-launcher as a “turbo-
jump” of sorts—fire a rocket into the ground, then ride the resultant explosion to jump
higher than otherwise possible. Sequence-breaking is a close derivative of exploits:
reaching areas or obtaining items out of the intended linear order of the game (typically
via an exploit). Players of both Metroid and Super Metroid have discovered ways of
obtaining objects earlier than intended and bypassing entire areas by manipulating the
morph ball’s bomb-hop (like rocket-jumping, a concussive boost into the air), wall-
jumping (jumping against a wall, then jumping off the wall for a further boost), the
“shinespark” (a super-charged jump), the ice beam (a weapon that freezes enemies,
which can then be used as platforms), and other devices. Each of these techniques, in
and of themselves, were intended by the games’ designers, yet through meticulous
timing, precise coordination, and/or clever combinations, players have managed to use
them to reach areas meant to be inaccessible. Perhaps most intriguing, though, is
alternative gameplay—when players make their own rules to play their own game within
the game, as with “Cat and Mouse” played in Project Gotham Racing 2. Making teams
of two, consisting of an extremely slow car (usually the Mini Cooper) and an extremely
fast car (Ferrari, Porsche, etc.), players attempt to get their team’s slow car over the

126
finish line first, whether by pushing it with their fast car, or by attacking the opposing slow
car—a racing mode that exists nowhere within the rules of PGR2. In all these examples,
players still operate within their games’ predetermined constraints, yet their behavior
forms part of a larger pattern of gameplay which cannot be anticipated by the designers,
behaviors rooted in players finding creative techniques and solutions within the
constraints of the game system. Exploits, sequence-breaking, and alternative gameplay
are just samples of the unpredictable results that occur when autonomous agents are
given room to move within a system of rules.
(2) Multi-player interaction. When we compound this unpredictability by having
multiple players interacting with one another—each thinking, autonomous agents, each
pursuing their own agendas, and each continuously reacting to the others’ activities—the
results become even less predictable. This is the model example of emergence—
autonomous agents performing context-dependent interactions that combine to form an
organized and yet non-deliberate whole. When two teams of eight (sixteen players in
all) play “Capture the Flag” in Team Fortress Classic, we see the very definition of
organized complexity. No one organizing intelligence is dictating the patterns of play;
there is no overriding control or authoring presence. And even amongst those playing,
no one has any idea how the game will end, who will win, or what the other team is
going to do (the strategy of such play relies on this ambiguity). And yet the overall
activity seems very organized. If we were to watch the game, it would likely have some
manner of dramatic content—perhaps one team jumps out to an early lead, but the other
team pulls from behind to win—it would display identifiable patterns on a macro-level,
and would have a clear beginning, middle, and end. The game would evince many of
the expectations of scripted entertainment, yet without any deliberate organization

127
dictating those features. Rather, such patterns emerge from multiple players interacting
with one another under the context and constraints of the rule system.
These are the two manners of emergence that we associate with traditional
gameplay—just as in chess, basketball, or Monopoly, unpredictable behaviors and
strategies emerge from players, separate from the rules themselves, exploring the space
of possibilities, making creative decisions, and interacting with one another in dynamic
ways. This is why we play games, after all—because we don’t know how they’ll play out,
we don’t know what the other player will do, and we don’t know who will win. Put simply,
players contribute to the emergence of videogames both in their individual behaviors and
their group behavior. Our third type, however, is the one which remains exclusive to
videogames.
Traditional games simply provide a system of rules; they are, in a very literal
sense, a formal structure. When we address their formal properties, we’re addressing
that which defines the space of possibilities and the conditions therein. And there are
some videogames which we could potentially describe in similar terms. Super Mario
Bros. provides one fixed world that never changes, where every enemy follows set
patterns (or reacts to the player in highly predictable manners), so much so that players
have been able to memorize the timing and sequence of button-pressing to the point
where they can literally conquer a level with their eyes closed. The emergent qualities of
such a game come solely from the unpredictability of player behavior. Many
videogames, however, provide much more than this.
(3) Artificial Intelligence (and player-AI interaction). When playing the single-
player campaign of Halo 2, we see much less variety than in a multi-player setting like
Team Fortress. Rather than sixteen autonomous players interacting within the
gamespace, a plethora of nonplayer characters (both friends and foes) have been

128
placed in predetermined locations by the game’s design, and only a single player
navigates the space. Yet despite such fixity of design, unpredictable results may still
occur, even beyond the player’s activity. Consider this small instance: a player operates
the turret on the back of a Warthog (an armored ATV) being driven by an AI-controlled
NPC (non-player character), who navigates the vehicle down a winding pathway into an
enemy encampment. Playing this short sequence the first time, the NPC driver, reacting
to incoming fire, swerves too close to the edge, and the Warthog tumbles down a
hillside, ejecting the player into a swarm of enemies. Playing the same sequence again,
the driver successfully navigates the Warthog down the path, and the player begins
taking out foe after foe with the aid of the turret. Playing again, the driver is killed by
incoming fire, so that the player must take the wheel and resultantly cannot fire their
weapon. None of these three sequences are scripted by the game’s designers, yet
neither are they directly motivated by the player’s decision-making nor the complex
interactions of multiple players; instead, they result from the dynamic interaction of the
player and autonomous agents sourced within the game itself.
Here we see a drastic shift in game emergence. Gameplay has long been the
product of players driving the iterative—the playing-out of the game occurs from players
operating within the rules. Yet videogames can provide a much more dynamic
experience. Thanks to their computational nature, videogames have the potential to
equally participate within the system of the game, to not only provide a space of
possibilities, but to provide agents within that space as well, and in a sense, to itself
operate as a thinking agent that interacts with the player. In such cases, the game not
only provides the constraints and boundaries of the system, but participates in it as well.
Chris Butcher, Engineering Lead for Halo 2, provides a telling description of how
Halo 2’s artificial intelligence operates: “Halo is a simulation engine. The engine creates

129
the world, then puts the player and the AI into it. …[The] characters and their code are
isolated from the world.”
113
Functioning separately from the world around them, the non-
player characters in Halo 2 each have multiple impulses and decision-making
processes—that is, their own independent rule-sets—that interact to drive their behavior.
Simplified, the process goes something like this (in Butcher’s terms):
• The character uses its AI “senses” to perceive the world—to detect what’s
going on around it.
• The AI takes the raw information that it gets based on its perception and
interprets the data.
• The AI turns that interpreted data into more processed information.
• The AI makes decisions about what actions [it should perform].
• Then the AI figures out how it can best perform those actions to achieve the
desired result based on the physical state of the world around it.
114

Able to make thousands of decisions per second, the AI then continually adjusts its
course of action in response to its interactions with the world and the other agents
around it. Rather than following scripted procedures, or even a series of “if/then”
procedures (if event X occurs, then perform action Y), the game’s artificial agents
produce spontaneous and unpredictable behaviors based on complex interactions with
the surrounding environment and its own drives and knowledge—much like a player.
The AI is itself part of the game’s formal set of rules, yet because each agent functions
autonomously (rather than functioning as one cohesive top-down unit), they are able to
generate emergent gameplay in a manner similar to multiplayer interaction.
It is in such instances that we see emergence move into the design of a
videogame, when the system of rules becomes part of the gameplay, rather than merely
providing the space of possibilities that will allow for it. The AI of Halo 2 offers a minor
example; SimCity provides a much more explicit one.

113
Robert Valdes, “The Artificial Intelligence of Halo 2,” How Stuff Works,
http://electronics.howstuffworks.com/halo2-ai.htm (accessed February 12, 2008). Ellipsis
included in original quote.
114
Ibid.

130
One of the most popular early examples of emergent design, SimCity provides a
vivid microcosm of simulated life through bottom-up behaviors much like those of a real
city. The world itself—the titular city—teems with emergent behavior, and reacts in
dynamic ways to every decision the player makes. As Steven Johnson explains, the
game achieves this by creating a meshwork of autonomous units that continually alter
their behavior in response to their neighboring units and to decisions made by players.
A given city block in SimCity possesses a number of values—the price of the
land, say, or its pollution level. As in a real-world city, these values change in
response to the values of neighboring blocks; if the block to the west drops in
value, and the eastern neighbor develops a higher crime rate, then the current
block may well grow a little less valuable. …the algorithms themselves are
relatively simple—look at your neighbors’ state, and change your state
accordingly—but the magic of the simulation occurs because the computer
makes thousands of these calculations per second. Because each cell is
influencing the behavior of other cells, changes appear to ripple through the
entire system with a fluidity and definition that can only be described as lifelike.
115

While we’re typically inclined to think of the player as the only agent operating within
SimCity, and the city merely as the world in which they are playing, a more accurate
model would be to consider each city block as its own autonomous agent, much like the
NPCs in Halo 2. The emergence of a thriving metropolis in SimCity comes from
hundreds of separate, autonomous agents interacting with one another within the
game’s system of rules—and the player is only one of them. The city blocks play
different roles and follow a different set of rules than the player, yet they are agents
within the system nonetheless. And like ant colonies, weather systems, and the human
brain, the city operates in a bottom-up fashion—each block follows its own limited set of
rules, and makes autonomous “decisions” based on its interactions with neighboring
blocks and with the decisions of the player, making the system so unfathomably complex
that we could never accurately predict exactly what will occur (though experienced
players will make educated guesses, just as meteorologists do with the weather).

115
Johnson, 88.

131
Further, we could never play the same game twice, resulting in a generative design
rooted in emergent programming—a series of micro-programs that interact in a bottom-
up fashion, much like any other emergent system.
The computer itself is, in a sense, a massive thinking agent that obeys an
inordinately long series of rules—the lines of code in the game-text itself. Yet the
computer is also able to run multiple processes simultaneously, and keep those
processes independent from one another, allowing it to play countless roles at once.
The computer can thus act as dungeon master (the narrating presence who maintains
the world of the game), as referee (the arbiter of the game’s rules), and as countless
participants (artificially intelligent agents within the game). The emergence of a
videogame remains unique, then, because players no longer only interact within the
system, but instead interact with the system.

Discursive Systems of Rules: New Media, Iterative Discourse, and the
Combination of Signification and Play
The emergent nature of a traditional game like basketball makes particular sense
because in basketball, technically, there is no actual “thing”—basketball is merely a
concept, an idea that we hold in our head, so just like warmth, freedom, love, or fear, we
can only know the concept by example. And because no one match can be called a
“definitive entirety” of the game (a “quintessential” playing of basketball), we only get a
thorough understanding of the game through multiple iterations of it. Yet while each
successive match gives us a better sense of what “basketball” is, we’ll never have a
finite picture of “the thing itself” in the way we can concrete objects. The game always
remains an abstract concept that we glean through each version we experience, an

132
ever-changing whole that we construct in our minds through each playing out of the
rules.
Texts, on the other hand, do indeed offer a concrete instance, thus traditionally
allow for a definitive entirety. Because of the fixity and tangibility that characterize texts,
we expect that every word, every image or sound, every brushstroke, everything that the
artist has put into the text will be made available to us by the time the experience is
finished, providing an undisputable discursive instance. We may not process all of it—
we rarely take note of every detail or subtext—and we may interpret it differently than
others, but we assume there to be no ambiguity as to what’s “in” the text. After all, we all
know what’s in Citizen Kane, the Mona Lisa, and Beethoven’s 9
th
. But what’s “in” a
game like basketball? Or for that matter, a videogame like Civilization III?
To check, I boot up my PC and revisit Civilization III, choosing all the same
introductory options as my last playing—same map size, same difficulty level, same
nation. The game starts up—“Dawn of Civilization…4000 BC...”—and after clicking
“OK,” I once again see my tiny Settler standing in the middle of an isometric map. But
much has changed since I last played. The mountains to the north have been replaced
by open grasslands. The ocean to the west is now a forest. I roam eastward, toward
where I once encountered the Roman Empire, but Rome is nowhere to be found. A
Japanese Scout, however, soon arrives over the northern grasslands. This seems a
very different game, indeed.
For comparison, I turn on my Xbox and return to Halo. Unlike Civilization, this
game does indeed seem quite familiar to my last playing: characters provide all the
same exposition, enemies appear just where I remember them, and I once again crash-
land on the Halo space station at the end of level one. Yet this iteration soon starts to
vary as well—whereas last time I attacked the Covenant patrols searching the surface of

133
Halo for human survivors, this time I keep my head low and evade the enemies
whenever possible.
So which account of Civilization do I cite as “how the game unfolds?” Which
version of Halo relays the true story of the Master Chief? Why, both, of course. And
neither.
Much like basketball, videogames also lack the definitive entirety we typically
expect of texts. When we play, we only get one possible version—one derivative
iteration. We may play again and again and again, and we may feel as though we’ve
experienced every possible iteration available, but we’ll never get “the whole thing” at
once, never experience a definitive playing of the game. Playing a game of Civilization
or Halo from beginning to end does not mean we’ve experienced the entirety of those
games—even though we often feel as though we have. In linear, story-driven
videogames like Halo, it is especially tempting to confuse the derivative for the text itself,
however even a linear game with a specific idealized solution, where we can follow a
“correct” series of events from beginning to end, won’t allow one definitive playing. At
the very least, a loss state—typically death—is always one possibility of the text, and if
we win, we don’t get that possibility.
116

The disappearance of a definitive entirety is one of the crucial ramifications of
games—as an abstract rule-set, games can only manifest as derivative iterations. Yet
texts exist as definitive objects, finite and complete works that allows for one formal

116
It’s worth noting that some people don’t consider such linear games—particularly those of the
text-based variety—as videogames at all, but rather as “interactive fiction.” Such a stance is
made for similar reasons as those who don’t consider puzzles or riddles to be games—if there is
one “correct” solution (the argument would go), we do not have a game on our hands, as a game
is an activity with a variable outcome. I’m willing to include such strictly linear videogames within
the medium, however I agree that there is a strong case in making such a distinction; they are, at
the very least, an edge-case that pushes the boundaries of what we might include in the category
of videogames, and this distinction demonstrates the necessity of variable iterations to the very
fabric of what comprises a game.

134
totality. Game-texts, then, are unique in that while an artist does create one formal
totality, that object does not provide one discursive totality (nor can the audience ever
have direct access to that object). Instead, a game-text allows for a potentially endless
number of discursive variations. Resultantly, while a videogame’s designer does create
a definitive object that we can point to as the thing itself, much like a text…

…we never have direct access to it, and only come to know the thing itself through its
derivatives, much like a game. While a book’s readers consume its words directly from
the page, a painting’s viewers look directly at the brushstrokes on the canvas, and a
film’s spectators, while not looking directly at the celluloid itself, still receives its sounds
and projected images in an orderly and undeviating fashion, a videogame’s players
never have this manner of access to the entirety of the text itself.
Many attribute this shift to videogames’ position within “new media” (i.e.,
computational media). Lev Manovich, for example, cites this capacity as one of the
fundamental principles of new media:
A new media object is not something fixed once and for all, but something that
can exist in different, potentially infinite versions. …Old media involved a human
creator who manually assembled textual, visual, and/or audio elements into a
particular composition or sequence…, its order determined once and for all.
…New media, in contrast, is characterized by variability.
117

117
Lev Manovich, The Language of New Media, 2
nd
edition (Cambridge: The MIT Press, 2001),
36.

135
Janet Murray instead discusses “procedural authorship” in computer-based texts,
118

Marie-Laure Ryan points to the algorithmic nature and “interrupt structure” of electronic
texts,
119
and Marsha Kinder credits the database functions of new media,
120
and each
includes videogames in their discussions. As “new media” becomes a more prevalent
buzzword in academia, the emergent discourse of videogames is more frequently
attributed to their computational functions—and this is true! Videogames indeed offer a
computational property that “old media” cannot afford us, allowing information to be
dispersed in a non-linear, generative fashion rather than in the traditional mode of pre-
ordained reception, and we cannot divorce this discursive shift from the computer
hardware that allows it. Yet while new media scholars think of these in computational
terms, we can address them in a much more low-tech manner: as rule-sets.
Manovich clarifies that the world according to a computer is comprised of two
basic “software objects”: data structures and algorithms.
121
Data structures are sets of
instructions for organizing and accessing stored information; algorithms are complex
formulas that define step-by-step computational procedures in a precise order and
based on specified conditions. Yet each of these can also be readily understood as
rules—absolute and precise instructions for “how to play.” It’s not surprising that
videogames are among the most common examples of new media—if any object is
particularly suited to computational application, it’s a game—a rule-based, nonlinear,
generative activity (the only thing that separates a game’s rules from those of any other
computational instance is the inclusion of specified, achievable goals). Videogames and

118
Murray, 185.
119
Marie-Laure Ryan, Narrative as Virtual Reality (Baltimore: The Johns Hopkins University
Press, 2001), 215-6.
120
Marsha Kinder, “Narrative Equivocations Between Movies and Games,” The New Media Book,
ed. Dan Harries (London: The British Film Institute, 2002), 119-32.
121
Manovich, 223.

136
other new media may allow for an unfathomably more complicated rule-set than
traditional games, yet they are systems of rules nonetheless.
The shift from “old” to “new” media, then, is from a discursive system based on
fixed content to one based on underlying rules. By providing a generative space of
interaction (via a database of possibilities and/or algorithmically generated possibilities),
game-texts become abstracted generalities that lacks any one tangible manifestation. It
is for this reason that such discursive artifacts must be experienced as derivatives rather
than a definitive instance of any sort—at every turn, we are offered a variety of different
choices (generated by the rules), but in navigating these choices, we can choose only
one at any given time.
122
The thing itself, then—the game-text—becomes an elusive
ambiguity that we may come to know inside and out via repeated playing, but that we
can never attain in its completion the way we can other texts (we can watch Citizen
Kane and get all of it in one pass; we can never do the same in a videogame).
To discuss the emergent, generative, algorithmic, nonlinear, procedural, or any
other “new medial” qualities of videogames, then, is to discuss their “game-ness”—their
being structured by a formal system of rules rather than a definitive instance of fixed
content. To discuss their expressive, descriptive, narrative, expository, argumentative,
or any other discursive qualities is to discuss their “text-ness”—their being structured in
order to convey information from a sender to a receiver. To answer my earlier question
of where the “art” of a videogame lies, then, we’d have to answer using the properties of
both texts and games—it is in their role as game-texts, in both play and semiotics: in
emergent discourse.

122
Hence Marsha Kinder’s discussion of “database narrative,” where she compares the process
of selection and combination that storytellers make in constructing a novel or film to the choices
players make in navigating a game. Kinder, 126-7.

137

Chapter Two: WORLD

Redguard

The most naturally talented warriors in Tamriel. In addition to their cultural
affinities for many weapon and armor styles, they also have a hardy constitution
and a natural resistance to disease and poison.

Skill Bonuses: Athletics +10, Blade +10, Blunt +10, Light Armor +5, Heavy Armor
+5, Mercantile +5

Specials: Adrenaline Rush, Resist Poison, Resist Disease

Sounds like a solid character. I choose a male Redguard from the ten races
offered by the character creation screen, then go about shaping his face to my liking.
Using a range of sliding scales, I mold a prominent jaw, sharp nose and cheeks, and a
heavy brow, topped by a thick head of stark white, windblown hair that handsomely
compliments the character’s dark skin and complexion, and I choose Crusader for my
character class, a vocation that offers proficiencies in the use of heavy armor and all
weapons, as well as destruction and restoration magic (combat and healing spells). I
give my Redguard Crusader a name, then click the X button to finish the character
creation process.
After a short load-time on my Xbox 360, I enter the world of The Elder Scrolls IV:
Oblivion, finding myself in a cramped prison cell. A Dark Elf in the cell across the

138
corridor makes some crude comments, yet I ignore his remarks, attending instead to the
voices I hear hurriedly approaching. Apprehending the world from the point-of-view of
my Redguard Crusader, I watch as some guards approach the door to my cell and
escort an elderly man in regal attire into the room. As he enters, the old man appears to
recognize me: “Let me see your face,” he says. “You are the one from my dreams.”
The game offers me some dialogue choices to guide the conversation, and I
learn that the man before me is Emperor Uriel Septim, ruler of all Tamriel. Assassins
have murdered his sons, and his Blades—elite guards assigned to protect the bloodline
of the Septim family—are leading him out of the city along a secret escape route that
happens to pass through my cell. The Blades interrupt, insisting that they must keep
moving, and one of them activates a secret door that rumbles open, revealing a dark,
cavernous passageway beyond. The guards and the emperor descend down the tunnel,
and Baurus, one of the Blades, calls back, “Looks like this is your lucky day. Just stay
out of our way,” before I am once again left alone in my cell.
I cautiously descend down the passageway and emerge in what appears to be
some long-forgotten ruins beneath the prison. Cobwebs fill the corners and crumbled
stones lie about where walls and pillars have given way to the years. I hear swords
being unsheathed up ahead, along with a cry of “protect the emperor!” and as I hurry to
catch up, I come upon a melee between the Blades and a trio of assassins. With no
weapon nor armor to defend myself, I stand clear of the fracas, watching as the guards
dispatch the interlopers. The Blades lose one of their own in the fray, yet dutifully
continue their escort nonetheless, and I watch as they disappear with the emperor
through a heavy wooden gate. Yet I am not alone for long.
As the door locks shut, one of the nearby walls suddenly crumbles, and two
enormous rats, each a good two meters in length, pounce upon me. I defend myself

139
with some weak blows of my fists, and manage to subdue the beasts. Alone and empty-
handed, I search the bodies of the fallen assassins, finding a few simple potions and an
unlit torch amongst their robes, then turn to the body of the slain Captain Renault, which
carries an Akaviri katana. Relived to get hold of a weapon, I hastily take the sword, and
with no other means of exit, I hesitantly peer into the cavernous passageways beyond
the newly-crumbled wall…
Though taking a much more circuitous route than the Emperor—through dark,
cavernous tunnels burrowed through the earth, filled with zombies, goblins, and other
foul beasties—I somehow manage to rejoin Septim and his Blades after thirty minutes of
adventuring, just in time for another attempt on the emperor’s life. “A trap!” Baurus
exclaims, and instructs me to guard the emperor with my life before retreating to battle
the assassins who have maneuvered behind us. As Baurus and his companion valiantly
fight, Septim confides in me the existence of a surviving heir to the throne: “I can go no
further. You alone must stand against the Prince of Destruction and his mortal servants.
He must not have the Amulet of Kings! Take the Amulet. Give it to Jauffre. He alone
knows where to find my last son. Find him, and close shut the jaws of Oblivion.”
Septim then gives me the amulet, just seconds before succumbing to an
assassin’s blade. Baurus returns, weeping, “We’ve failed. We the Blades are sworn to
protect the Emperor, and now he and all his heirs are dead.” I inform Baurus of Septim’s
instructions to visit Jauffre, the Grandmaster of the Blades, and deliver the Amulet so
that it might reach the last heir to the throne. Despite his misgivings, Baurus trusts in the
Emperor’s final wishes, and instructs me to seek out Jauffre at Weynon Priory, where he
now lives as a monk. He gives me the key to the sewers so that I might escape with the
Amulet, and I depart into the bowels of the prison before finally emerging into the
daylight of Cyrodil.

140
And it is here that the game truly begins.
To call the province of Cyrodil vast would be an understatement. Whereas the
caverns and sewers had offered a linear gauntlet of challenges, deliberately routing me
through a pre-designed path of narrative events and tutorial information, Cyrodil offers
quite the opposite: open vistas in every direction, allowing unhindered movement and
endless exploration. An expansive landscape containing nine cities, hundreds of unique
locations (caves, forts, ruins, camps, villages, cottages, inns, shrines, and more), and
roughly 16 square miles of terrain, the gameworld of Oblivion is simply enormous. As I
stand outside the sewer gates, I gaze upon a lush lake shore and spot some white stone
ruins on the far side, preceding grassy hills that expand upwards into mountains that
tower in the distance. Behind me stands the glorious Imperial City, towering atop its
rocky island perch, the spire of the city’s central tower reaching upwards into the
heavens. To my left and right, I see the lands stretch off to the horizon—lands which are
all readily explorable, offering seemingly limitless opportunities.
In the coming weeks and months (both in-game and in my own life), I undertake
countless adventures across the expanse of Cyrodil. I find Jauffre, locate Septim’s heir,
and escort him to safety at the Blades’ sanctuary high in the mountains of Bruma. I save
the sieged city of Kvatch from ruin by closing a dimensional portal to the literal hell of
Oblivion, and come to be known across the kingdom as “the hero of Kvatch.” I infiltrate
the Mythic Dawn, the cult behind the emperor’s assassination, and learn the insidious
intentions of their leader, Mankar Camoran. And as I traverse the kingdom, I find that
nearly every citizen I meet seems to have some task needing completed. Local
merchants in the Imperial City ask me to investigate a newcomer who is undercutting
their prices; a distraught Argonian in Chorrol implores me to locate her missing daughter;
the Countess of Bruma requests that I recover an ancient relic from the ruins of a nearby

141
fort; the proprietor of Faregyl Inn wants me to discover who’s been stealing her potatoes.
Everywhere I turn, there is another quest to be fulfilled, and as I navigate the land, I
come to a significant realization: I know the world of Oblivion better than I know its story.
Though curious to see how the plot unfolds, I find myself loading up the game
time and again not for a concern of defeating Mankar Camoran, but rather wanting to go
back to Cyrodil. Make no mistake, Oblivion is a heavily narrative game—there’s lots of
story to be found here, and as the opening exposition establishes, a very deliberate plot
binds the game together. Yet that story is mapped onto the landscape in such a manner
that the diegetic world of the game becomes the story. Cyrodil lives and breathes more
than any character in the game—a vast and elaborate world of countless places and
infinite happenings, Cyrodil is constructed such that simply being there, running across
the landscape, becomes just as powerful as what happens. And it is while running
across that land that I discover a rather curious phenomenon in this virtual world.
Having joined the Fighter’s Guild—an order of benevolent mercenaries—I’m
instructed to investigate some fellow guild-members who have been causing trouble in
the town of Leyawin. I open my map to find the city nestled away in the southeast
corner of the province, and begin my long trek across the continent. Traveling across
the countryside, I encounter several minor adversaries—a pair of trolls lingering about
the entrance to a cave, a wolf that pounces forth from the underbrush—yet nothing I
can’t handle. I move through the hills and valleys, traipse over a grassy ridge, and see
below me a blue stream winding west to the sea. I hop down from a rocky outcropping
and continue striding ahead, when suddenly, and without plausible explanation, I hit
what appears to be an invisible wall.
“You cannot continue this way. Please turn back” I’m told by white text that
appears at the top of my screen. Defying this request, I attempt to press my avatar

142
ahead, yet my movement continues to be hindered by this invisible barrier. I try to run
around the impenetrable blockade, yet everywhere I go, the invisible wall blocks my
progress. I can see the open hillside continuing in front of me, the stream still winding
below, yet try as I might, I cannot continue any further. My Redguard Crusader
ineffectually runs in place, impeded by an invisible barrier for which the game offers no
explanation. For a brief moment, all semblance of the game’s internal reality subsides;
I’ve reached the border of the gameworld.
Every game, and equally every fiction, invokes a world of some sort, a world
delineated as separate from our own. Whether a chessboard or the Wild West, a
basketball court or “a long time ago in a galaxy far, far away,” both games and fiction
take place in a time and space apart from the real world. In this regard, we might
consider Cyrodil in one of two ways: as a diegesis, a fictional world of story and
character, or as a play-ground, the space of possibilities inside the rules of the game.
And it is in the incongruities between these that we encounter such peculiarities as its
invisible wall, resulting from the discrepancy between the game’s systemic logic and its
fictional logic. Fictional logic would dictate that Cyrodil expands far beyond the wall, as
evinced by the field, stream, and mountains I can see in the distance. Yet the game’s
system of rules dictates that the world ends here; beyond this point is “out of bounds,”
for while the fictional world may be limitless, the gameworld is decidedly finite.
If we wish to look, we could find countless examples of such discrepancies.
Spatial boundaries are easy—every game has to end somewhere. Designers typically
attempt to veil these boundaries—as with the tunnels beneath the prison, where walls
and locked doors delimit my movement—though we can find many cracks in the façade
if we wish to look. Why do the villainous creatures of Metroid reappear every time I
return to an area, regardless of how many times I defeat them? Why does time stand

143
still in Resident Evil while I manipulate my inventory? Why can’t I jump in The Legend of
Zelda? Why does my avatar reincarnate within second of dying in Team Fortress
Classic? Why do so many monumental events in Metal Gear Solid coincidentally
happen just as I arrive at a particular spot? The answer: because the rules of the game
say so. These quirks result from disjunctions between the rules of the game and the
fictional logic of the text. As we’ll see, videogame worlds need serve both functions, and
while game design typically attempts to legitimize the rules of the game within the logic
of the fiction—avoiding exactly these sorts of discrepancies—these quirks help us to
delineate the two.
As we saw in chapter one, semiotics—the domain of texts—is concerned with
signification, conveying information via the representation and abstraction of concepts
and referents. Gameplay, meanwhile—the domain of games—results from operating
within a delimited space of possibilities. As game-texts, then, the domain of videogames
is in the representation of a space of possibilities, and the player’s ability to operate
therein. The primary function of a videogame is in creating—or, more accurately,
signifying—a world of play.
To put it another way, all videogames represent something, and the infinite list of
things that a videogame might represent always starts with the same thing: space. No
matter what game we play, we’re always confronted with some manner of space: a vast
narrative world, an illogical maze, a two-dimensional gauntlet, or an abstract arena.
Even text-based games still invoke a conceptual space to be navigated; they merely
“tell” of that space rather than show it. In every videogame, we encounter some world
with its own parameters of time, space, and logic, a space of possibility begging
interaction, one in which we imagine ourselves to be active. And yet, in reality, there is
no space—it’s all just pixels of light on a screen.

144
When I play Pong, I see two paddles deflecting a ball between one another…yet
there are no paddles! There is no ball! And even if there were, there’s no space in
which to perform this action! It’s all an illusion—a representation of space. Similarly, if
we step back from our Oblivion example, we see that indeed there is no wall, but then
again, there is no “me,” either. I’m not running through a field. Neither is my Redguard
Crusader, for that matter, as there is no Redguard Crusader. None of this exists—
they’re all representations. All videogames are merely comprised of signifiers—a vast
series of representations and abstractions, each conveying some manner of information
to the player. So while there may be no wall, neither is there a space in which that wall
might exist. All that appears on my screen is an intricate web of signification, and the
vast land of Cyrodil is but one of many concepts that I read from those signifiers.
So while chapter one addressed the formal structure that comprises a
videogame—the system of rules and aesthetics which provide its formal identity—this
chapter is instead concerned with the “content” that system delivers: the world of the
videogame.

145
Every videogame offers us a world of some sort—a land in which to live, a space in
which to play—and whether we view it as a play-ground or a diegetic space, that world is
what’s “in” the game-text, what is transmitted by videogame discourse. This chapter,
then, is on videogames as world simulators—objects invested in spatiotemporal
projection.

Temporary Worlds: The Play-ground and the Diegetic Space
To address the world of a videogame, we begin this chapter where we left off in
the last—with play. Roger Caillois notes:
All play presupposes the temporary acceptance, if not of an illusion (indeed this
last word means nothing less than beginning a game: in-lusio), then at least of a
closed, conventional, and in certain respects, imaginary universe.
123

This imaginary universe is the object of our attention—the play-world, the ludic domain
that is so persuasively delineated from the world outside. Certainly, every videogame
must project some manner of space—a fictional arena in which the game may be
played, an imaginary world on the other side of the screen. In discussing the emergence
that results within a game’s system of rules, we saw numerous references to the
figurative space created by play—the game’s “space of possibilities,” the room for
movement within the game’s formal structure. In this regard, the world of a videogame
is the space inside the rules—bound by formal constraints, gameplay results as players
discover avenues of opportunity within the game’s formal rule-set. Yet videogames must
equally entail some manner of actual space—a spatiotemporal domain in which the
player may be active.
Writing in Homo Ludens—by most accounts, the canonical work on games and
play—Johan Huizinga describes this domain as the play-ground, using the term only

123
Caillois, 19

146
partly for its literal definition. “Play is distinct from ‘ordinary’ life both as to locality and
duration,” Huizinga tells us. “It is “’played out’ within certain limits of time and place [and]
contains its own course and meanings.” Temporally, “play begins, and then at a certain
moment it is ‘over,’” allowing play to be repeatable and to occur at fixed intervals.
124
Yet
even more striking than its temporal dimensions, he says, are its spatial parameters.
All play moves and has its being within a play-ground marked off beforehand
either materially or ideally, deliberately or as a matter of course. …The arena,
the card-table, the magic circle, the temple, the stage, the screen, the tennis
court, the court of justice, etc., are all in form and function play-grounds, i.e.
forbidden spots, isolated, hedged round, hallowed, within which special rules
obtain. All are temporary worlds within the ordinary world, dedicated to the
performance of an act apart.
125

As Johan Huizinga describes it, the play-ground is, at least in concept if not materiality,
the time and space delimited by the rules of play, the arena in which play occurs. While
Huizinga lists a bevy of literal spaces as example, he is more concerned with their formal
and conventional nature—the domain of their rules—than their concrete boundaries.
Rules state that a particular logic applies here, a logic different from the world at
large, creating a whimsical and impermanent elseworld that is, in one way or another, an
alternative to reality possessing its own curious dimensions.
All play has its rules. They determine what “holds” in the temporary world
circumscribed by play. The rules of a game are absolutely binding and allow no
doubt. …[A]s soon as the rules are transgressed, the whole play world
collapses. The game is over.
126

For Huizinga, rules are the fragile membrane that separates gameplay from the laws of
the real world, establishing a conceptual space free of outside concerns and barricaded
from reality—a space of possibilities, yes, but more importantly a space that is overtly
separate from the outside world. Roger Caillois echoes these sentiments, addressing
the “agreed boundaries” of a game which isolate it from the rest of the world, replacing

124
Huizinga, 9-10.
125
Huizinga, 10.
126
Huizinga, 11.

147
“the confused and intricate laws of ordinary life” with “precise, arbitrary, [and]
unexceptional rules” that govern a fixed space and time.
127
Whether playing basketball,
Scrabble, or Oblivion, when we engage the rules of a game, we step outside the
demands of our everyday lives (jobs, bills, relationships, politics) and enter a space
where a jump shot matters, a triple word score matters, and whether our Redguard
Crusader defeats Mankar Camoran matters. Both writers thus address rules in similar
fashions: as delimiting a space—sometimes literally, but always conceptually—in which
the “real world” subsides and a fanciful act of make-believe stands in its place, but doing
so in strictly organized terms. Much like emergence toes the line between order and
chaos, both writers see games as spaces that negotiate the order of rules with the
freedom of play.
Lest we take Huizinga’s metaphor of the playground too far, we should
acknowledge that while literal playgrounds exist as spaces for generalized play,
Huizinga’s version is actually much more limited. Though outwardly addressing play of
all sorts, Huizinga’s account privileges formalized play over freeform play by his
repeated mention of rules and boundaries.
128
This is the distinction that Roger Caillois
addresses as paidia and ludus, between spontaneous, “instinctive” play and play by
organized convention (i.e., the rules of games, the conventions of drama, etc.).
At one extreme an almost indivisible principle, common to diversion, turbulence,
free improvisation, and carefree gaiety is dominant. It manifests a kind of
uncontrolled fantasy that can be designated by the term paidia. At the opposite
extreme, this frolicsome and impulsive exuberance is almost entirely absorbed or
disciplined by a complementary, and in some respects inverse, tendency to its
archaic and capricious nature: there is a growing tendency to bind it with
arbitrary, imperative, and purposely tedious conventions, to oppose it still more
by ceaselessy practicing the most embarrassing chicanery upon it… This latter

127
Caillois, 6.
128
This seems largely the result—as addressed in chapter two—of his native Dutch lacking a
distinction between “play” and “game” (as both translate to the Dutch spel), making the vital
distinction between generalized play and rule-based play the one notable oversight in his
otherwise brilliant account.

148
principle is completely impractical, even though it requires an ever greater
amount of effort, patience, skill, or ingenuity. I call this second component
ludus.
129

Caillois chooses these terms based on their roots in “child” and “game,” respectively,
indicating raw, innate play on one hand and the sophistication of organized play on the
other. In addressing the rules, boundaries, order, and conventions of play, Huizinga
privileges ludus, organized play, and it is here that we need distinguish Huizinga’s “play-
ground” from the traditional use of that term.
Freeform play is rarely bound from within—while some authority (parents,
teachers, society) may demarcate spaces appropriate for free play by asserting
boundaries from outside (playgrounds, game rooms, recess yards), freeform play does
not necessitate such bounds. These spaces do not manifest from the play itself, but are
instead asserted from outside (the reason that such authoritatively demarcated spaces
often have a difficult time containing free play, as children always seem to be playing
“where they’re not supposed to”). Games, however, create deliberate bounds by their
system of rules, thus the very activity of the game asserts the world in which they occur.
“Out-of-bounds” only occurs within ruled play, and while not every game places tangible
limitations on time or space, every game does allow us to distinguish who is “in” the
game and who is not, when the game is in progress and when it is over. Even if the
boundaries of time and space are not formally delimited, we can still distinguish the
game’s domain and its duration.
The “imaginary universe” of play occurs beyond the distinct formality of games
and rules, of course. Outside of games and other formal activities, however—in the
realm of paidia—that imaginary universe manifests in much more ambiguous fashions,
as described by Katie Salen and Eric Zimmerman:

129
Caillois, 13.

149
A child approaching a doll, for example, can slowly and gradually enter into a
play relationship with the doll. The child might look at the doll from across the
room and shoot it a playful glance. Later, the child might pick it up and hold it,
then put it down and leave it for a time. The child might carelessly drag the doll
around the room, sometimes talking to it and acknowledging it, at other times
forgetting it is there.

The boundary between the act of playing with the doll and not playing with the
doll is fuzzy and permeable. Within this scenario, we can identify concrete play
behaviors, such as making the doll move like a puppet. But there are just as
many ambiguous behaviors, which might not be play, such as kneading its head
while watching TV. There may be a frame between playing and not playing, but
its boundaries are indistinct.
130

By this example, there is a still a clear difference between playing and not, but the
boundary between the two is uncertain. The child playing with the doll certainly seems
to participate in play’s imaginary universe, but without formal constraints—without
rules—it becomes difficult to delineate where play ends and the real world returns. Yet
such is hardly the case in gameplay. When we play a game, any game, there is little
doubt to the domain of play. Whether chess, baseball, Tetris, or Oblivion, the distinction
between playing and not playing is without ambiguity. As formal systems, then, games
concretize the imaginary universe, both in their frequent spatial parameters and their
explicit strictures on behavior. Rules formalize the domain of play, making the
boundaries of a game much less ambiguous than those around the child playing with the
doll. This is the play-ground, as we’ll use the term—the play-world as concretized by the
formal system of rules, the demarcated boundaries of time, space, and behavior that
govern the play of a game.
As Huizinga and Caillois both acknowledge, the temporary world of a game is
first and foremost a psychological construct, one reliant on the player’s willingness to
pretend. It is for this reason that both begin by stressing play’s voluntary nature, even
before addressing its separateness from the rest of life and reality or the conventions
that mark it (as Huizinga notes, obligatory play would be, at best, a “forcible imitation”

130
Katie Salen and Eric Zimmerman, Rules of Play, 94.

150
thereof).
131
Though the play-ground frequently entails some physical manifestation, the
space of play is ultimately nothing more than an idea, a state of mind. Resultantly, play
can only occur by willing submission to its world—both the literal rules of the game and
the whimsical logic of its make-believe. It is for this reason that Huizinga warns against
trivializing play based on its make-believe quality, pointing to the passionate seriousness
which frequently accompanies it:
[T]he consciousness of play being “only a pretend” does not by any means
prevent it from proceeding with the utmost seriousness, with an absorption, a
devotion that passes into rapture and, temporarily at least, completely abolishes
that troublesome “only” feeling. …The inferiority of play is continually being
offset by the corresponding superiority of its seriousness.
132

While play is easily dismissed as trifling and inconsequential, we must equally
acknowledge the vital earnestness and sincerity of play, as play’s illusion relies on the
participant’s willingness to take the act seriously, to subscribe to its reality despite its
very unreality.
In chapter one, we largely treated “play” as a manner of activity, focusing on play
only insofar as it offers some manner of concrete demonstration. In considering the
term, we focused primarily on three of our four principal definitions of the word: (1) an
activity of recreation, (3) an enactment or performance, and (4) having freedom of
movement, or looseness. Yet we paid little heed to its second definition—(2) the quality
of being fun, and not serious—addressing this simply as the spirit which generally
accompanies the first. In this sense, play is the abstract mindset that promotes such
recreation, the “idea” of play rather than its outward activity, and here that spirit is our
foremost concern, as the “temporary world” to which Huizinga refers is the manifestation
of that spirit. In this sense, play is not so much as an act, but a state of mind—what
Bernard Suits terms “the lusory attitude.”

131
Huizinga, 7; Caillois, 9.
132
Huizinga, 8.

151
Suits describes the lusory attitude as “the acceptance of constitutive rules just so
the activity made possible by such acceptance can occur.”
133
Though his account
proceeds in a much more rhetorical manner, avoiding abstractions which cannot by
objectively demonstrated, Suits echoes Huizinga and Caillois’ emphasis on the spirit of
play behind the activity. The evidence, he says, is in the player’s willingness to submit to
arbitrary constraints in the pursuit of a goal.
The attitude of the game player must be an element in game playing because
there has to be an explanation of that curious state of affairs wherein one adopts
rules which require one to employ worse rather than better means for reaching
an end.
134

As example, he asks: if a golf player wishes to put a ball in a hole in the ground, why
would they choose to walk several hundred yards from the hole, proceeding to knock the
ball toward the hole with a metal stick, when simply using one’s hands to place the ball
in the hole would be a much more efficient means? He answers, ultimately, by
acknowledging that game actions—putting a ball in a hole, crossing a finish line,
positioning game pieces in a given configuration—have no value in and of themselves,
but rather gain their value in the context of the game (where they become a hole-in-one,
a victory, or a checkmate). The lusory attitude is the acceptance of that context—the
willing consent to the unreality of the game.
“To play a game is in many ways an act of ‘faith’ that invests the game with its
special meaning,” say Salen and Zimmerman. “To decide to play a game is to create—
out of thin air—an arbitrary authority that serves to guide and direct the play of the
game.”
135
To describe that act of faith, they deploy their concept of the “magic circle,” a

133
Suits, 40.
134
Ibid, 38.
135
Katie Salen and Eric Zimmerman, Rules of Play, 98. This metaphor of “faith” may actually be
very telling. Throughout his book, Huizinga connects play to the origins of ritual and ceremony,
stressing their gravity and sanctity—the very “belief” in the activity—as having its basis in the
seriousness of play.

152
phrase borrowed from Huizinga “as short-hand for the idea of a special place in time and
space created by a game.”
As a closed circle, the space it circumscribes is enclosed and separate from the
real world. As a marker of time, the magic circle is like a clock: it simultaneously
represents a path with a beginning and end, but one without beginning and end.
The magic circle inscribes a space that is repeatable, a space both limited and
limitless. In short, a finite space with infinite possibility.
136

Expanding on Huizinga, Salen and Zimmerman acknowledge a formally delimited space
and time bound by the unique logic of the system of rules, and their description further
echoes the discussion of play and emergence from chapter one, describing a world both
delimited and infinite. Yet equally important, they describe the magic circle as a contract
into which the player enters, regardless of deliberate physical components. For them,
the magic circle is a “cognitive frame” more than anything, offering a particular context
through which to understand the game’s meanings, and it is the playful willingness to
accept the very unreality of the game, despite its arbitrary and technically meaningless
nature, that defines its domain. “As a player steps in and out of a game,” they say, “he
or she is crossing that boundary, or frame, that defines the game in time and space.”
137

The world of play, then—whether Huizinga’s play-ground, Caillois’ imaginary
universe, or Salen and Zimmerman’s magic circle—is a psychological re-
contextualization of one’s activity. “Within the magic circle, special meanings accrue and
cluster around objects and behaviors,” Salen and Zimmerman explain, re-casting
elements from the real world to have new significations. “In effect, a new reality is
created, defined by the rules of the game and inhabited by its players.”
138
Psychologist
Michael Apter similarly addresses play as establishing a “protective frame” standing
between the player and the real world, creating an “enchanted zone” that is valuable

136
Ibid, 95.
137
Ibid, 370.
138
Ibid, 96.

153
precisely due to its separateness from life and from consequence,
139
while anthropologist
Gregory Bateson describes play as entailing a “psychological frame” that provides a
special manner of signification.
140

Bateson ascribes several qualities to this psychological frame, but we can boil
these down to three primary characteristics. First, it is both inclusive and exclusive,
clarifying certain components as meaningful within it while delimiting others that do not
belong to the play context. Bateson uses a picture frame as metaphor in that it tells the
viewer, “attend to what is within and do not attend to what is outside.” The play-ground
does not merely frame the play-world inside, but equally “frames off” the world outside,
foregrounding some components while delimiting others. Second, this frame asserts a
“premise,” or what I am here calling a “context”—it tells the participant that they should
not use the same thinking when interpreting representations inside the frame that they
would otherwise. The play-ground contains its own logic and meanings, and its framing
of play asserts that new context. But most important to Bateson’s model is the final trait,
that this frame does not merely communicate meaning, but equally entails a
metacommunication:
Any message, which either explicitly or implicitly defines a frame, ipso facto gives
the receiver instructions or aids in his attempt to understand the messages within
the frame. …[Conversely,] every metacommunicative or metalinguistic message
defines, either explicitly or implicitly, the set of messages about which it
communicates, i.e., every metacommunicative message is or defines a
psychological frame.
141

By Bateson’s thinking, this metacommunication is the play context, providing the very
frame that contextualizes the activity; for him, play is fundamentally a manner of
signification that entails an additional lusory signification.

139
Michael J. Apter, “A Structural-Phenomenology of Play,” in Adult Play: A Reversal Theory
Approach, eds. J.H. Kerr and Michael J. Pater (Amsterdam: Swets and Zeitlinger, 1991) 15.
Cited by Salen and Zimmerman, Rules of Play, 94.
140
Bateson, 187.
141
Ibid, 188.

154
As a signifying activity, Bateson argues, play establishes meanings and
communicates various manners of content, yet it simultaneously communicates the
artificiality of that content so that all participants are aware that they are “only playing,”
indicating not only specific meanings, but equally the attitude we need take when
interpreting that meaning. The play context thus establishes representations while
equally reminding us that they are only representations—they mean something within
the protective frame of the game, yet do not mean the same things outside that frame.
Bateson points to such communication not only in human behavior, but in animals as
well, using such examples as monkeys play-fighting and dogs nipping at one another. In
such instances, he says, even those observing the playing animals recognize their
behavior to be “only play,” not “the real thing.”
Now, this phenomenon, play, could only occur if the participant organisms were
capable of some degree of metacommunication, i.e., of exchanging signals which
would carry the message “this is play. …Expanded, the statement “This is play”
looks something like this: “These actions in which we now engage do not denote
what those actions for which they stand would denote.” …The playful nip
denotes the bite, but it does not denote what would be denoted by the bite.
142

Bateson terms this a metacommunication because it is a communication about the
communication itself—while the explicit message makes a statement (“I’m biting you!”),
the metacommunication delivers an implicit message about that statement (“…but I’m
not really biting you—it’s just pretend!”).
Playing re-situates our activity under an alternative reality. When a child
encounters a stuffed animal, or an athlete picks up a baseball bat, or a gamer plays
Oblivion on their Xbox 360, those activities are re-contextualized to have some vital,
alternative meaning that is separate from the literal act itself—playing with an elephant
rather than pieces of woven fabric stitched together and stuffed with synthetic filler;
hitting a home run rather than arbitrarily swinging a wooden stick at a cowhide ball;

142
Ibid, 179-80.

155
traversing the landscape of Cyrodil in an epic quest to defend its throne rather than
clicking buttons and staring at a screen of light while sitting on the living room sofa. In all
these cases, the participant performs an act of make-believe, pretending to be in one
situation when in fact existing in another, all the while conscious of that situation’s
artificiality. Play, then, is a willing delusion, and the “frame” of play is both the boundary
of that make-believe and the context that allows for it. Whether psychologically
transporting the player or simply re-contextualizing their actions, the temporary world of
the play-ground results from a context that we simultaneously accept as real but unreal,
and the metacommunication of play is the implicit acknowledgement of that artificiality.
Salen and Zimmerman describe this as a “double consciousness” in which the
player is cognizant of their activity’s meaning both inside and outside of the magic circle,
and we see this double consciousness echoed in the work of numerous critics, using a
variety of terms to describe it:
[H]ow can it be both the case that words and other elements…have their ordinary
meanings and yet the rules that attach to those words and other elements and
determine their meanings are not complied with?
143

[O]ur ability to understand…is distinct from our beliefs as to its truth,
appropriateness, plausibility, rightness, or realism.
144

[W]e have here one of those intrinsically ambivalent behaviours in which a single
action, with double roots, expresses simultaneously virtually opposite
tendencies.
145

All of these accounts acknowledge the participant’s comprehension occurring in two
separate domains simultaneously, understanding a communication while equally
understanding it not to mean what it is meant to mean. Yet these statements were not
intended to reference games, nor even play in any traditional sense, instead discussing

143
John Searle, Expression and Meaning: Studies in the Theory of Speech Acts (Cambridge:
Cambridge University Press, 1979), 58
144
Branigan, 192.
145
Christian Metz, The Imaginary Signifier: Psychoanalysis and the Cinema, trans. Celia Britton,
Annwyl Williams, Ben Brewster, and Alfred Guzzetti (Bloomington: Indiana University Press,
1982), 102.

156
a manner of metacommunication that generally goes by a very different public moniker:
fiction (the above quotes are from John Searle, Edward Branigan, and Christian Metz,
respectively, all writing on the nature of fiction).
We shouldn’t give games more credit than they deserve, as they are hardly the
only formalized manifestation of the “magic circle.” Anytime that a spectator sits down in
a movie theater, a reader sits down with a novel, or a theatergoer sits down to enjoy a
stageplay, they are entering a playful make-believe and engaging the
metacommunication of play (Huizinga did, after all, include the stage and screen as
examples of the play-ground). Such audiences enter into the world of a story and allow
themselves to be affected by events that have never occurred, all the while fully aware
that they have never occurred, necessitating an act of play—of agreeing to pretend that
“this is the case” despite knowing otherwise—just as readily as a childhood game of
cops and robbers. Though cultural convention leads us to associate “play” with light
entertainment and childish pastimes, even the most serious, solemn, intellectual, or
adult-like of stories—Citizen Kane, The Grapes of Wrath, A Portrait of the Artist as a
Young Man, or Death of a Salesman—all utilize the magic circle. They occur in the
realm of fantasy, that place in which we “suspend our disbelief” and allow ourselves to
be affected by that which is not real, thus even the most gut-wrenching of tear-jerkers or
the most somber of tragedies require an act of make-believe.
The work of Christian Metz regarding the fiction film is especially telling of its
connection to the metacommunication of play. Comparing fiction with dreaming, Metz
notes that unlike dreams, which provide a genuine illusion of reality (in that the dreamer

157
is unaware that it is only a dream), fiction instead offers “a certain impression of reality,”
maintaining an implicit awareness of the gap between fiction and actuality.
146

It is understood that the audience is not duped by the diegetic illusion, it ‘knows’
that the screen presents no more than a fiction. And yet, it is of vital importance
for the correct unfolding of the spectacle that this make-believe be scrupulously
respected…, that everything is set to work to make the deception effective and to
give it an air of truth.
147

Much like Huizinga’s account of the seriousness of play, fiction still requires that we
“take it seriously,” that we willfully commit to its unreality while under its spell, despite an
outward knowledge if its artificiality. Just as the baseball player must commit to the
belief that a home run matters and the player of Adventure must commit to the belief that
“that square is me,” so too the cinematic spectator must be willing to accept fictional
characters and events within the terms of their own reality. Metz even cites this willful
seriousness as resulting in a third domain of the fiction.
I shall say that behind any fiction there is a second fiction: the diegetic events are
fictional, that is the first; but everyone pretends to believe that they are true, and
that is the second; there is even a third: the general refusal to admit that
somewhere in oneself one believes they are genuinely true.
148

Metz’ account ultimately reiterates much of Huizinga, Bateson, and Salen and
Zimmerman’s claims regarding play, demonstrating that fiction is but a particular
manifestation of play—the narrative variety.
Both enabled by a psychological frame that couches objects and behaviors under
a guise of make-believe signification, play and fiction are linked in their status as
implicitly-acknowledged artificialities. Games and stories are simply different
formalizations of the play context, each invoking some framed, imaginary universe with
its own dimensions of time, space, and logic. These formalizations, however, don’t
necessarily entail the same manner of imaginary universe.

146
Ibid, 101.
147
Ibid, 72.
148
Ibid.

158
Games and traditional play re-contextualize the world, creating their imaginary
universe by re-casting the world in a new light. The play-ground is a space where
objects and behaviors are given a new gravity, perhaps even an entirely new
signification, yet they still exist in the world at large, temporary worlds within the ordinary
world. While the act of fiction often entails a play-ground in Huizinga’s sense—the
stage, the screen, the site of narration—it is much more concerned with the world
projected by that narration: the diegetic world. This is fiction’s imaginary universe, the
world of the story, and while equally a formalized manifestation of play, this world
proceeds by a significantly different design.
Though we saw its original usage in chapter one, used to convey a verbal
“telling” rather than an imitative “showing,” the term “diegesis” has evolved in more
recent decades to become the accepted name for the fictional world inside a story.
149
Its
most frequent application is in distinguishing the “diegetic” components of a narrative
from the non-diegetic, separating those elements which exist within the fictional world
from discursive components experienced only by the audience (such as subtitles or
underscoring). In a broader sense, however, the diegesis is the entirety of time and
space that exists within the fiction, the universe in which the story takes place and
everything that happens therein. Per Edward Branigan:
We understand [narrative] events as occurring in a “world” governed by a
particular set of laws. I will refer to that imagined world as the diegesis. The
spectator presumes that the laws of such a world allow many events to occur
(whether or not we see them), contains many objects and characters, contains
other stories about other persons… The diegetic world extends far beyond what
is seen in a given shot and beyond even what is seen in the entire film, for we do
not imagine that a character may only see and hear what we observe him or her
seeing or hearing. The diegesis, then, is the implied spatial, temporal, and

149
The term arrived here by way of Etienne Souriau, who re-deployed the term to reference the
“recounted story” of a film, and although Souriau’s usage was originally applied to cinema, the
term has since been adopted for discussing other media as well. David Bordwell, Narration in the
Fiction Film (Madison, WI: The University of Wisconsin Press, 1985), 16, citing Etienne Souriau,
Preface, in L’univers filmique, ed. Etienne Souriau (Paris: Flammarion, 1953).

159
causal system of a character – a collection of sense data which is represented as
being at least potentially accessible to a character.
150

The fundamental difference we need acknowledge between gameworlds and fictional
worlds is in their boundaries: games clearly assert an “out-of-bounds,” limits where the
playspace ends (in this regard, a game provides a universe similar to the pre-Socratic
view of a flat Earth, where the world simply “stops” somewhere). A diegetic world,
however, continues infinitely, existing, if not in our own universe, then at least in one as
equally large. We do not assume the diegetic world to end at the border of the cinema’s
screen or the stage’s proscenium arch, nor beyond the scope of the literary narrator;
instead, fiction provides a small segment of what we assume to be a much larger world
(hence the possibility for endless sequels, prequels, spin-offs, and tie-ins, other stories
that occur within the same fictional universe). While Salen and Zimmerman described
the world of a game as “a finite space with infinite possibility,” we might conversely
describe the diegetic world as an infinite space with finite coverage.
Like the play-ground, the diegetic world is also an imaginary universe created by
a formal system—in this case, a discursive system of narration rather than a system of
rules—yet unlike the arena, the tennis court, the chessboard, or the stage, the diegetic
world does not actually “exist” anywhere, occurring only in the mind of the audience.
Fiction creates an illusion of time and space—we might point to the images on a screen
or the verbal descriptions in a book for reference, yet the world of fiction is a wholly
imaginary one, cued in the mind of the audience by the representations that comprise
the text. Yet a diegesis does create a temporary world with its own “spatial, temporal,
and causal system” (in Branigan’s words), and in this regard, fictional worlds bear great
resemblance to the play-ground. When we engage a fictional world—whether the
fantastic universe of the Star Wars franchise or the relatively close-to-life rendition of

150
Branigan, 35.

160
James Joyce’s Dublin—we step outside our own reality, suspend our disbelief, and
temporarily commit to the internal reality of that world. When we engage a time-past in
the works of Shakespeare or a time-yet-to-come in the works of Philip K. Dick, we leave
the present day and visit an era that is clearly separate from our own. And when we
enjoy J.K. Rowling’s Harry Potter universe or J.R.R. Tolkien’s Middle Earth, we accept a
world that allows for mysticism, magic, and monsters, regardless of whether we’d accept
the logic of those possibilities in our own world. Diegetic worlds may seem vastly
abstracted from our own, or may appear very much like the world we know, yet
regardless of correlation, their worlds proceed by their own time, space, and logic—three
fundamental components that comprise a gamespace and that equally define the
dimensions of a fictional world.
151

The play-ground and the diegetic world are actually two different manners of
space which may occur—and frequently do—at one and the same time. Both are
manifestations of the magic circle, worlds that materialize for the sake of, and equally as
a result of, an act of play, yet they perform different services to that act. Our best
example might be in the case of theater: when an actor stands in front of an audience
performing a role in Hamlet, is that actor on the stage, or “in Denmark”? Are they in the
play-ground (playing pretend) or in the diegesis (transported by the make-believe)? By
all accounts, we’d have to say that, to one degree or another, the actor is in both places
at once. While the play-ground is a formalized space delimited for the sake of that
activity, the diegetic world is the space that emerges from within it. Enacted fiction thus
entails both manners of space—one in which to perform the act of play, another
projected by that act. Such is equally the case in many traditional games, from

151
Note that while some narrative models will say “causality” rather than logic—as does
Branigan—this is merely to focus on a particular aspect of diegetic logic: that which connects two
events.

161
Monopoly’s gameboard projecting an Atlantic City vied over by real estate barons to the
Tolkienesque worlds of Dungeons & Dragons that are evoked by the dice-throwing and
graph-paper-drawing of its table-top gameplay. And this is also the case in a
videogame, and this is precisely why we encounter such peculiarities as the invisible
wall of Oblivion. When we encounter Oblivion’s wall, we see the slip between these two
worlds—the play-ground ends, yet the diegetic world continues, much as Hamlet’s set
ends at the stage’s wings, while “Denmark” expands far beyond.
All videogame worlds must act as both spaces: as a play-ground because all
games offer a space of possibilities, a formally delimited space in which to play, and as a
diegesis because all videogames entail a projected space, occurring not “here” but
rather “there,” on the other side of the screen, in a space that exists only in our mind’s
eye. As players within a play-ground, we act within a context and environment whose
very reality is defined by a formalized system of boundaries and affordances, defining
the very structure of the game. In this manner Cyrodil is no different from the basketball
court or the chessboard, a delimited arena in which we recognize a strict and peculiar
logic which dictates our behavior. Equally, however, we witness a tangible world that
does not exist where we sit, nor even “inside” the screen on which it seems to appear—a
projected spatiotemporal realm that exists only in our imagination, an implied idea of
spacetime projected by the discursive object. In this manner Cyrodil is no different from
Singin’ in the Rain’s fictionalized Hollywood or 1984’s dystopian future, an infinite world
that we can imagine to expand far beyond the scant bit that we actually encounter
through the text. One world may be privileged over the other (the diegesis in Oblivion
and Halo, the play-ground in Tetris and Pac-Man), yet every videogame world is
comprised of both of these spaces to one degree or another.

162
Narrating Space: A Diegetic Model of Videogames
As evidence of these parallel worlds, consider Simon. Remember Simon? Co-
developed by Ralph Baer (father of the Magnavox Odyssey!) and distributed by Milton
Bradley, the original Simon was an electronic device that featured four primary-colored
buttons that would light up in sequence, requiring its player to memorize and repeat the
sequence. So Simon is a game. And it’s driven by a computer.
152
It’s a “computer
game.” But is this a videogame?
Similarly, think of pinball—not the old, clockwork pinball machines, but rather the
modern electronic variety. Also computerized games, these are even featured in
arcades, environments that are now largely dedicated to videogames. But are these
videogames?
In both cases, I think we can most assuredly answer “no.” But why? Before
getting caught up in the “video” part of the word, keep in mind that videogames have not
been played on the raster video equipment for which they were named in decades, so
we can’t rely on this semantic distinction. Rather, these fail to qualify because they do
not entail a diegetic space—the player’s activity does not occur anywhere other than
“right here.” Videogames, on the other hand, entail virtual spaces, worlds that exist on
the other side of the screen, cued into the mind of the player by the representations of
the text.
153

We might ask: why do games like Dance Dance Revolution or Guitar Hero need
to create diegetic spaces? Unlike videogames played with traditional controllers, these

152
As stressed by its slogan: “Simon's a computer, Simon has a brain, you either do what Simon
says or else go down the drain.”
153
Though it is not the screen itself that is requisite. I suspect that if future generations were to
see holographic technology or some manner of direct neural input replace the videogame screen,
we would still call such experiences videogames. An early ad campaign for the Playstation 2
even suggested as much, depicting a futuristic Playstation 9 which players jack directly into their
brains (the ad then introduces the PS2 as “the beginning”).

163
require players to mimic the very actions they are supposed to be performing—dancing
or guitar playing—and the only information that players need in order to play is the
sequential button combinations that they must hit in rhythm. So why not simply allow the
action to occur in the “real” space in which the player is standing on a dancepad or
holding a guitar-shaped controller? Why do these games still evince a need to merge
the player’s activity outside the screen with an imaginary space “inside” the videogame?
Perhaps the better question would be: if they did not, would they still seem like
videogames? I suspect that we’d instead call them “videogame-like,” something closely
related to videogames—like pinball or Simon—and yet ostensibly “something else.”
Videogames, after all, are not just “digital games,” games that happen to be
computerized. If such were the case, objects like Simon and pinball would be included
in the category. This difference, I suspect, is the lack of an internal world—such games
do not project the player “into” their domain, do not entail a spatiotemporal realm beyond
the screen. They are games that happen to be digital, rather than videogames.
154

Videogames, on the other hand, project a virtual space in which a player participates;
they entail not just a play-ground, but a diegetic play-ground. This is precisely why the
avatar becomes a vital component—to allow the player presence in that world beyond
screen.

154
It is for the same reasons that I would not include several manners of game that are played on
videogame consoles, such as Sudoku or The New York Times Crossword played on the Nintendo
DS. I hesitate to call these videogames in much the same way that I would not call a stand-up
comedian’s performance “theater.” The comedian’s performance may entail many of the tropes
we associate with theater (a performance, a stage and set, a live audience), and by the same
token, The New York Times Crossword entails many of the tropes we associate with videogames
(interactivity, a system of rules, videogame hardware), yet each are clearly “something else,”
something closely related yet intuitively different. Nor do I include wholly idea-based games,
such as webpages that simply generate trivia questions, within my definition. These are exactly
the sort of texts that necessitate the catch-all of “interactive entertainment,” as they don’t seem to
fit what we mean when we discuss “videogames” in particular.

164
Projected spaces, then, are part of the very fabric of the videogame medium. All
videogames ultimately seek to transport the player “somewhere else,” entailing activity
that occurs in some separate spatiotemporal context from the player’s own. Simon and
pinball are games that we play with; videogames are games that we play in. In this
regard, videogames are world simulators—discursive objects invested in spatiotemporal
projection. They tell of spaces “somewhere else,” imaginary spaces that exist nowhere
but in our minds, as even the most abstract of videogames project some space beyond
the literal surface of the screen. Tetris, Pong, Qix, Pac-Man, Hexic, flOw, Amplitude,
Lumines Live!—regardless of how abstract or conceptual the game may be, every
videogame projects some manner of spatiality beyond the dimensions of the screen
itself.
In other words: videogames “narrate” space.
Now, the dispute against narrative models was one of the first grand battles in
videogame studies, so we will certainly need to test this hypothesis, as it is indeed a
controversial one. Oblivion obviously fits the bill—loaded with pre-designed narrative
components, Oblivion clearly tells a story. But can we say this of all videogames? Can
we argue this…

…as a narrative space?

165
Most of us recognize that some videogames are narrative ones, yet most would
equally agree that not all videogames “tell stories.” Super Breakout seems to be entirely
free of narrativity. Manipulating the paddle at the bottom of the screen, the player
bounces the “ball” against the colored bars at the top of the screen, eliminating blocks
each time the ball strikes them. The player’s goal is to eliminate all the blocks, and the
game ends when the player has missed the ball a specified number of times. We see
none of the common tropes of narrative—no characters, no discernable plot, nothing
recognizably concerned with “storytelling” of any sort. Referring to Super Breakout as
narrative seems just as silly as describing Connect Four, charades, baseball, or
blackjack as such. Similarly, puzzle games, sports games, racing games, party games,
and countless others rely not on plot or character, but rather rule-based mechanics that
offer little correlation to storytelling.
We would hardly call a basketball game an act of narrative. So why would we
describe NBA Live as such? Neither would we attribute narration to Scrabble, so why
Scrabulous? Poker certainly doesn’t fit the bill, so why think otherwise of Xbox Live
Arcade’s Texas Hold ‘Em? Because in all of these cases, a basic ontological change
occurs in the translation between the original games and their digital counterparts—the
play-ground disappears, moving their action into a virtual space.
When I play basketball, the game does not need to tell me that I’m bouncing a
ball because I am, in fact, bouncing a ball. No one need “project” me onto the court,
because I am on a basketball court. We describe such spaces as “imaginary” because
the game’s system of rules recasts the meaning of that space (and the actions therein),
thus signifying a new context, yet I am still in a literal space performing a literal act. As
Huizinga describes, a game provides a temporary world within the ordinary world, a
world bounded from real life yet still rooted in our own reality. When I play NBA Live,

166
however, I’m not bouncing the ball, because there is no ball. There is no court on which
to bounce the ball. There’s not even a “me” to do the bouncing. In the move from game
to game-text, the activity loses its footing in the real world, moving to a virtual space that
exists nowhere but in our minds. Everything in the game must be projected by the
videogame, and as such, the play-ground becomes a fictional space not only in its ludic
context, but in its very existence.
Much like the worlds of literature and cinema, a videogame world is an imaginary
universe completely detached from the player’s own time and space—not a temporary
world within the ordinary world, but rather one apart from the ordinary, entirely
independent of the world outside. If we were to stop a game of basketball, move it off
the court, and continue it in a different location, at a different time, in an environment
with different physics (say, on the moon), it would hardly be the same game, for while we
address the play-ground as a space delimited from the world, its dimensions of time,
space, and logic are still bound to our own in a fundamental way. Yet if I pause NBA
Live, move my Playstation to the moon and continue, the game’s internal reality remains
constant. Unlike basketball, NBA Live’s play-ground exists independent of the real
world. Despite the unreality of its rules, basketball still occurs—still exists—in a basic
ontological sense; NBA Live, however, is nothing but patterns of light and a series of
changing ones and zeroes.
As narratologist Edward Branigan describes it, narrative is “a fundamental way of
organizing data” that we encounter “not just in novels and conversation but also as we
look about the room, wonder about an event, or think about what to do next week.”
155

When deploying the term, we tend to focus on a particular brand of narrative—that
involving a deliberate, linear chain of events (plot) organized around special individuals

155
Branigan, 1.

167
(characters) and following culturally-sanctioned conventions and formulas (structure,
point of view, genre)—yet narrative is actually much broader than this. Branigan
describes narrative as a human practice that offers a special “psychological use value” in
its ability to model spatiotemporal data into perceivable experience:
Narrative is a perceptual activity that organizes data into a special pattern which
represents and explains experience. More specifically, narrative is a way of
organizing spatial and temporal data into a cause-effect chain of events with a
beginning, middle, and end.
156

By Branigan’s perspective, narrative conceptually organizes time and space to project a
series of events linked by causal logic, and we can apply this description to Super
Breakout just as readily as Citizen Kane or Moby Dick. Super Breakout offers a
consistent and coherent world defined in both space and time (the two-dimensional
space of the screen spanning the continuous time between beginning and end), a chain
of events (paddle deflects ball; ball hits block; block disappears…) dictated by the
internal logic of that space (the game’s physics), containing a clear beginning, middle,
and end (game start to game over). It may not seem a particularly compelling narrative
by conventional standards, but it clearly demonstrates all the requisites that Branigan
lays out.
My claim, then, is not that videogames inherently “tell stories” (at least, not in the
colloquial sense of that phrase). Some do (Oblivion, Metal Gear Solid, Half-Life), while
others don’t (Super Breakout, Tetris, NBA Live). Rather, my claim is that videogames
are a narrative medium in that they are patently reliant on narration for discursively
creating a space of play. All videogames entail the projection of a consistent
spatiotemporal domain wherein events occur by their own internal logic, and only
narration provides the means for such an act.

156
Ibid, 3. Italics are Branigan’s.

168
The problem is not with the concepts of narrative; the standing work on
narratology can readily accommodate even the most abstract of videogames. The
problem, instead, is in the unnecessary baggage that the term itself carries and the
connotations that follow its colloquial usage. The most reductive definitions of narrative
describe a manner of discourse in which information is organized in such a way as to
assert a trajectory of events, occurring within a conceptual world and connected by
causal logic, and such a definition can accommodate even the utter abstraction of Super
Breakout. Yet when we deploy the term “narrative” or “narration” in everyday use, we
usually mean something much more specific. As such, we need to rehabilitate (for lack
of a better word) narrative terminology that is applicable to videogame play—and more
specifically, to videogames’ manner of spatiotemporal presentation—yet get away from
the tacit assumptions that tend to accompany it.
There’s an unfortunate ambiguity in the word “narrative” itself, a term which can
appropriately be used to describe a discursive practice (“the act of narrative”) as well as
a particular product of that practice (“a narrative”), and equally as a descriptor for things
related to that practice (adjectival uses like “narrative filmmaking”). This leads to a
number of terminological faux pas, such as the mistaken assumption that to describe an
object as “narrative” is to say that it is “a narrative.” Further, we tend to conflate such
words as “narrative,” “story,” and “plot,” terms that point to three different referents yet
are often used interchangeably. When we discuss narrative, what we frequently mean is
“conveying a plot” (and equally, the phrase “narrative videogames” usually means
“videogames with a plot”), yet plot is very different from story or narrative. So to better
deal with videogame discourse, we’ll make a pivotal terminological distinction, dividing
“narration” from “storytelling.”

169
Narration, as we’ll use the term, is a discursive practice of organizing
spatiotemporal data to posit a trajectory of events occurring within the internal logic of a
diegetic spacetime.
157
Or, for a more compact version: narration is the act of
spatiotemporal projection, of creating a diegesis. This is the broader use of narration,
not bound by colloquial assumptions of plot, character, and other details to which we’ve
become accustomed yet are by no means vital to the narrative act. Storytelling, on the
other hand, we’ll use to describe the narrower, more conventional practice of conveying
a character-oriented plot: a deliberate, causal chain of events organized around
protagonists. By this distinction, narration is a broader activity than storytelling, and
storytelling a particular subset of narration.
158
Rather than get bogged down in the
vagaries of “narrative,” then, we will instead focus on a particular mode of discourse—
narration—along with the content of that discursive act—story.

Narration and Story
What does it mean to “narrate?” Numerous narratologists suggest that we may
find our best initial understanding of narration by considering what it is not.
159
When
imagining the vast multitude of ways in which we might organize data and deploy
information, how does narration compare with everything else? In chapter one, we
noted that as discursive objects, all texts are concerned with some conveyance of
meaning, involving information (content) that is transmitted via a means of transmission
(the medium) between a sender (author/artist/designer) and a receiver (audience).

157
“Spacetime” comes to us from physics, used to reference a mathematical model that
combines space and time in a single coordinate system, merging the three dimensions of space
(height, width, and depth) with the fourth dimension of time. Though we frequently reference
these as separate concepts, the spacetime continuum model explains space and time to be
indivisible.
158
By the distinction that some ludologists wish to make, we could say that simulation is another
subset.
159
For example: Branigan, 1; Chatman, Coming to Terms, 6.

170

Yet this is merely the broadest model of communication; within this, we can identify
numerous subtypes of discourse, each with their own agenda—argument, description,
expression, exposition, classification, evaluation, instruction, documentation, and many
more. Each of these serves a different discursive purpose, referring to a different
manner of content—argument attempts to convince; exposition is meant to explain;
description relays characteristics; expression conveys emotion and abstract thought.
We needn’t concern ourselves with the entirety of all discourse, however. For us,
narration’s value is specifically in its ability to frame spatiotemporal data—to posit a
world—and as such, we need not compare narration with any and all forms of organizing
data. Modes such as argument and expression can certainly manifest in videogames,
yet these do not convey time nor space, instead, dealing in more abstract data (thought
and emotion). Modes like instruction and exposition occur as well, yet these deal in
process—how or why things occur. Our concern is instead in distinguishing narration
from other modes capable of conveying spatiotemporal data.
Let me clarify straightaway that I am by no means arguing that videogames only
employ narration. As Seymour Chatman observes, discursive types are not exclusive
and “routinely operate at each other’s service,” thus nearly any text will employ multiple
types.
160
We might use description while relating a narrative, or a narrative to further an
argument, or an argument to aid instruction. Oblivion’s narration is persistently
accompanied by description and exposition—the sounds and images of the game

160
Chatman, Coming to Terms, 10.

171
provide a vivid depiction of the fictional world of Cyrodil and everything within it, while
information attained from a variety of sources help explain why current events are
occurring and how the logic of this fantasy world operates. The game employs a
number of expressive qualities, from the music used to convey emotional states to the
eerie blue glow featured in Cyrodil’s Ayleid ruins, used to convey a sense of ethereal
mysticism. We can even find mild arguments throughout the game, both in general
ideological attitudes (the consequences of criminal activity) and more deliberate parallels
to the contemporary world (the relationships between different races). And throughout
the game, instruction is used to bolster the player’s ability to navigate the world and
operate through the game’s interface. That’s easily five manners of discourse at work
beyond narration alone. So videogames can indeed argue, express, teach, explain, and
perform innumerable other discursive functions. Yet these are not our concern here. In
projecting gameworlds, we’re only concerned with the videogame’s ability to create a
viable impression of a self-contained internal reality, detached from the player’s own
reality. In this regard, we primarily need distinguish narration from its closest sibling,
description.
Could we argue that Super Breakout (or Oblivion, for that matter) is merely a
descriptive text, rather than a narrative one? As audio-visual media, all videogames
deploy description—their sensory output essentially states “this is what things look
and/or sound like.” By offering sensory imitations, videogames inherently detail, just as
painting or photography; such media passively assert details by showing them rather
than telling of them, yet they detail nonetheless. Though relatively few and far between
nowadays, even text-based games are chock full of description—verbal description, but
description all the same—providing details of the gameworld so that the player can

172
adequately understand its nature. Without description of the world and of what exists
and is occurring inside of it, the player would be incapable of operating within that space.
Yet this is precisely an example of one manner of discourse operating in the
service of another. Videogames do describe; yet if they were merely to describe, they
would not create a fully dimensioned spacetime. What, then, is the difference between
narration and description? What does it mean to narrate as opposed to merely
describe? Chatman suggests we can distinguish these by their respective logics,
dividing the “metonymic logic” of description from the “chrono-logic” of narration.
A term Chatman borrows from Phillipe Hamon, metonymy literally means to
substitute the name of one thing for that of another.
161
Description occurs by relaying
details in place of the thing itself: to verbally describe my Redguard Crusader, I
accounted of a “prominent jaw,” “sharp nose and cheeks,” “heavy brow,” “stark white
hair,” and “dark skin,” all named attributes that I compound to create a mental
impression that stands in place of a particular character. To sensorily describe my
Redguard Crusader, Oblivion illustrates a prominent jaw, sharp nose and cheeks, heavy
brow, stark white hair, and dark skin, which compounded also stand in place of that
particular character. In either case, details become proxy for the thing itself, a sort of
metaphor in which characteristics “stand for” the signified content. To describe, then, is
to not merely name the referent, but to assert abstracted details which further stand for
it.
Troublesome is that nearly all narration includes such details. Without
descriptive signifiers deployed in place of the signified content, one could only gain the
most rudimentary sense of character, event, or setting. In defining narrative, Branigan

161
Chatman, Coming to Terms, 24. Term taken from Phillipe Hamon, “Rhetorical Status of the
Descriptive,” trans. Patricia Baudoin, Yale French Studies 61 (1981), 1-26.

173
acknowledges it as a pattern that both “represents” and “explains” experience, terms that
invoke description and exposition, suggesting the sympathy between these three modes.
Not only is narration inherently descriptive, but expository as well. To explain the
process of narration, then, we need be able to account for details and logic, while
equally distinguishing narration as a separate mode of discourse. This distinction lies in
the framing of such details.
While to describe is merely to assert characteristics, to narrate is to frame
descriptive elements (via expository logic) to assert a trajectory of events. As Branigan
indicated, narration does not only represent and explain experience, but more
specifically organizes spatiotemporal data “into a cause-effect chain.” The following
statements are all descriptive: “The dog was brown with black paws.” “The dog ran
clumsily.” “It was a bright and sunny day.” “I got a dull ache in my stomach.” Yet if I
frame those statements in a manner which posits events connected by some causal
logic, they become narrational: “My neighbors had a brown dog with black paws.
Everyday after school, I would walk past their yard and watch as it ran clumsily up and
down their fenceline, hoping that I might stop and play. It was a bright and sunny day on
Monday, perfect for frolicking, yet when I came home, I found no dog to greet me. A dull
ache began to grow in my stomach; I knew something was wrong.” The explanatory
connections need not be explicit; the reader here needs infer the causality between the
dog’s usual behavior, the events of this particular day, and the narrator’s uneasy feeling,
yet all the same, the passage asserts a trajectory of events—the details combine to posit
not merely a unified referent (as in description), but a unified sequence of events.
The result is not merely the metonymic logic of description, but what Chatman
terms “chrono-logic.”

174
[W]hat makes narrative unique among the text-types is its “chrono-logic,” its
doubly temporal logic. Narrative entails movement through time not only
“externally” (the duration of the presentation…) but also internally (the duration of
the sequence of events that constitute the plot). The first operates in the
dimension of narrative called Discourse (or récit or syuzhet), the second in that
called Story (histoire or fabula).
162

By this account, narrative content is defined not only by temporal movement, but by a
double temporal movement—the first of the telling or showing, and the second of the
signified events themselves existing entirely independent of the first (note that since our
point at hand is to distinguish narration from other discursive types, I will more explicitly
reference “narration” as the first dimension rather than “discourse,” to avoid ambiguity).
In my account of the brown dog with black paws, we can distinguish the temporal frame
of the telling—the ten seconds or so it takes to read those four sentences—from that of
the story, vaguely spanning several months or years (the ambiguous length of time in
which the narrator would regularly see the dog on his way home from school and leading
to the climactic day in which the dog was missing).
Yet despite Chatman’s emphasis, these separate frames are not only
distinguished by temporal movement. Time is often the clearest example of this division
(as most stories involve a span of time greater than the telling), yet hardly the only one.
In a theatrical presentation of Hamlet, the space of the stage (the space of narration) is
just as clearly distinct from the space of Denmark (that of the story), and even in our
story of the brown dog, we envision a space that is clearly separate from that in which
we read those sentences (though the only explicit mention of spatial components are
“yard” and “fenceline,” we still imagine a street, houses, and other necessary details that
are implied by the narrated events). Yet like many narratologists, Chatman privileges
the temporal dimension over other dimensions of the story, and this is a flaw we find with

162
Chatman, Coming to Terms, 9. Citing Phillipe Hamon, “Qu’est –ce qu-une description?”
Poetique 12 (1972), 475.

175
much of traditional narrative theory—especially those dealing with literary narration, as
literary spaces are often quite ambiguous to begin with.
163164
Yet in his explicit address
of cinematic narration, Edward Branigan, for one, discusses this double framing of story
and narration not only in regards to time, but space and logic as well.
Narrative is a way of comprehending space, time, and causality. Since in film
there are at least two important frames of reference for understanding space,
time, and causality, narrative in film is the principle by which data is converted
from the frame of screen into a diegesis—a world—that frames a particular story,
or sequence of actions, in that world; equally, it is the principle by which data is
converted from story onto screen.
165

Branigan distinguishes the site of narration—the screen—from that of story via the three
principles of space, time, and causal logic, and further stresses not only the story’s
sequence of actions, but the world in which it occurs as well. Narration, he tells us, is
the principle by which information is converted between these two frames, allowing the
audience to understand that data in both contexts simultaneously. Rather than a “doubly
temporal logic,” then, we would better think of narration under a “two world logic,”
distinguishing the time, space, and logic of the discourse from the time, space, and logic
of the content. To amend Chatman: narration’s uniqueness among other discursive
types is in its ability to project an imaginary spacetime independent of one’s immediate
context.
166
This returns us to my original claim of narration: that narration is the
discursive act of projecting a world separate form the here-and-now.

163
Gerard Genette’s esteemed Narrative Discourse, for example—its title claiming the whole of
narration as its subject—spends three of its five chapters on dimensions of time, yet includes nary
a mention of narrative space.
164
This privileging of time may be equally influenced by those literary genres most concerned
with elaborately detailed worlds and spaces—science fiction and horror, in particular—not being
given the same critical esteem or attention as those genres that foreground plot and character
above setting.
165
Branigan, 36.
166
Branigan further enforces these two worlds in his distinction of the diegetic from the
nondiegetic: “The spectator’s organization of information into diegetic and nondiegetic story
worlds is a critical step in the comprehension of a narrative and in understanding the relationship
of story events to our everyday world.” Branigan, 35.

176
When Branigan says that narrative is something that we encounter in our
everyday lives, even outside of “storytelling” proper, this is what he’s addressing—this
two world logic, our ability to hypothesize a time and space separate from our immediate
context. This is “narrative” not as a communication, but as an act of supposition, of
speculating on hypothetical worlds and hypothetical events. Humans are unique in our
ability to enact such speculation, to conceive of a time and place separate from that in
which we exist. Narrative, in this sense, is our ability to mentally project a
spatiotemporal context different from our own.
Narration, on the other hand, is the act of projecting such a spatiotemporal
context to someone else. This is narrative as a discursive practice, and this is our
concern. We're dealing not just with narrative as an ability to hypothesize another
spacetime, but narrative as the act of defining such a spacetime for an audience. In this
regard, when we engage a book or movie—or a videogame—someone else (author,
filmmaker, designer) is defining such a spatiotemporal context for us. They say to us,
via the text: “imagine Victorian England, inhabited by a crotchety old miser who hates
Christmas…” or "imagine a two-dimensional space, with colored blocks at the top, a
paddle at the bottom, and a ball bouncing between them..." The latter may not entail the
elaborately pre-determined event-chain of the former, yet it still involves the projection of
imaginary events in an imaginary space.
Story, then, is not simply narration’s trajectory of events, but the entirety of
narrated content (at least, so it is used in the French structuralist tradition—as in the
work of Gerard Genette, and continued by English writers such as Seymour Chatman—
and so too we will use it). Story is all that is “inside” the narrative, the world projected by
the discourse and everything contained therein—events (plot), actants (characters), and
spacetime (setting) included. In other words: the whole of the diegesis.

177
In truth, story and diegesis mean pretty much the same thing. Clarifying his
terms, Seymour Chatman distinguishes story from narration as such:
[E]ach narrative has two parts: a story (histoire), the content or chain of events
(actions, happenings), plus what may be called existents (characters, items of
setting); and a discourse (discours), that is, the expression, the means by which
the content is communicated. In simple terms, the story is the what in a narrative
that is depicted, discourse is the how.
167

And as Stam, Burgoyne, and Flitterman-Lewis describe, diegesis—as imported into film
narratology via Etienne Souriau and Christian Metz—makes a similar distinction.
Diegesis (more or less synonymous with Genette’s “story” (histoire)) refers to the
posited events and characters of a narrative, i.e. the signified of narrative
content, the characters and actions taken as it were “in themselves” without
reference to their discursive mediation.
168

Diegetic space, diegetic time, diegetic logic, diegetic events, diegetic actions, diegetic
sounds, diegetic voices, diegetic light, or diegetic anything all refer to that which is inside
the story (as opposed to the nondiegetic, which exists at the level of discourse). Both
terms point to “the signified of narrative content,” the information told, separate from its
telling. In use, however, the terms tend to carry different implications.
Most frequently, we use “story” to simply describe “what happens.” If I were to
say “I’m going to tell you a story,” would you expect me to describe a place and time, the
people in that place, and the sorts of things that happen there? Or would you expect me
to provide an organized, causal chain of events? Most likely, the latter—you’d expect to
hear “what happens.” “What happens,” however, is not story—“what happens” is plot,
the specific sequence of events with which the narrative is concerned. We use “plot” to
distinguish an organized, causal chain of events from the broader category of story,
which entails all the additional details of diegetic content. Yet although story entails
much more than a sequence of events, this has become its everyday connotation. Our

167
Seymour Chatman, Story and Discourse: Narrative Structure in Fiction and Film (Ithaca:
Cornell University Press, 1978), 19.
168
Stam, Burgoyne, and Flitterman-Lewis, 38.

178
foremost concern in a traditional narrative is in “what happens,” and as a result, when we
think of story, we think first of the chain of events, and only secondarily of “everything
else.” The term “diegesis” was repurposed, in part, to refer to that “everything else,” thus
when we use diegesis, we think of the world inside the narrative (and only secondarily
what happens therein). So although the two terms technically mean the same thing—
the whole of signified content—they foreground different aspects of the content. And
this allows us a useful distinction.
We could reasonably say that most approaches to narration are story-based
approaches: approaches that privilege “what happens.” The fundamental concerns of
narratology are typically temporal arrangement (order, duration, and frequency of
narrative events), perspective (the point of view by which events are presented), and plot
(what the events are). Even character has been a vastly under-examined topic,
relatively speaking. And as for setting and space? Forget about it.
Videogames, however, entail a different brand of narration. Certainly, they
involve events (something happens in every game, digital or not). They involve causal
logic (we can connect events and make sense of why things happen). They involve
perspective (every player has their own, dictating their reception of the game). And they
even involve characters—or at the very least, actants (the players themselves, the
people to whom the events happen). But while all of these issues remain relevant, the
central concern of videogame narration is in the projection of spacetime.
169
To discuss
videogames, we need a diegetic approach to narration, a rendition of narratology that
privileges the diegetic world over the usual focus on the trajectory of events.
Narratology can indeed speak to the way in which videogames project—essentially,

169
It’s quite telling that when we discuss “perspective” in traditional narration, we focus on how
we view events; when we discuss perspective in videogames, we focus on how we apprehend
space.

179
narrate—their worlds, yet the traditional privileging of linear, plot-and-character-based
storytelling does not serve us well. A diegetic approach would not privilege plot first,
character second, and setting last, but would instead foreground the diegetic world and
the discursive act of projecting that world (and rather than speaking of plot, characters,
and setting—terms carrying all the baggage of traditional narrative—would speak of
events, actants, and worlds, neutral terms that don’t presuppose content that explicitly
represents human experience).

A Diegetic Approach
Michel de Certeau makes one of the strongest cases for such an approach in his
assertion that “every story is a spatial story—a travel practice.”
170
To make this claim, he
relies on a conceptual distinction between places and spaces, describing place as “an
instantaneous configuration of positions” while space entails “vectors of direction,
velocities, and time variables.”
171
If I reference a three-bedroom house on Forest Drive
in Pittsburgh, it is merely a place, a location, a “where” or a “there.” If I show you a
photograph of a bedroom, or describe the layout of the house, it still remains a place.
We could describe it as “a thing in a location”: the house, on Forest Drive. It is an
independent entity unto itself, an abstracted idea in the service of nothing. But if I tell
you of my growing up in that house, of a young boy doing things within its walls (eating
dinner, leaving for school, getting in trouble), it becomes a space. We conceptualize it
differently, as it is now a realm of action, a space in which things occur. The house in
and of itself is no longer just a location—it is now the location of happenings. It becomes
a domain in which actants can act.

170
Michel de Certeau, The Practice of Everyday Life, trans. Steven Rendall (Berkeley, CA:
University of California Press, 1984), 115.
171
Ibid, 117.

180
Alternately, think of a landmark: Mt. Rushmore. Are you thinking of it? Picture it
in your head. Now think of the climax to North by Northwest, as Roger O. Thornhill
evades the villains and saves Eve Kendall atop Mt. Rushmore. Notice a difference?
The first was merely a place, a thing practically. It exists in an instant and entails no
“happening” that would suggest a temporal dimension. The second, however, is a
space, an arena in which things occur, in which actants are able to “do.” Whereas the
former could reasonably be conceived as an object rather than a location at all, the latter
is clearly a space where things happen.
By projecting actions, we activate space, changing what had previously been an
abstracted idea of spatial dimensions into a living, breathing world. In short, space is
more fully-dimensioned than place—a place exists in three-dimensions, while space, by
adding a temporal component, exists in four. A place is spatial; a space is
spatiotemporal. A place is a “thing”; a space is a domain. As de Certeau puts it, “space
is a practiced place,” and conversely, we could say that place is an abstracted space.
172

To examine this distinction, de Certeau goes on to describe two discursive
practices that he couches under the metaphors of maps and tours. Maps are accounts
that assert spatial relationships, as in his example, “the girls’ room is next to the kitchen.”
Tours, on the other hand, imply a movement through space: “you turn right and come
into the living room.” We must guard against falling victim to the surface of these
metaphors—this distinction is not concerned with overhead drawings vs. guided walks,
nor whether we see the space in question at close range or from afar. Rather, de
Certeau is distinguishing whether we account of space as a snapshot capturing one
instant frozen in time, or whether we assert a trajectory of events, activating a temporal
component. He’s making a distinction between described places and narrated spaces.

172
Ibid, 117.

181
De Certeau is dividing description from narration—maps describe space, while tours
narrate it.
173
Both discursively convey information, and both assert details of a location,
yet only the latter projects a temporal component, suggesting changes in state. Only a
tour—and equally, a “space”—allows for a trajectory of events.
174

Via this distinction—“maps” vs. “tours”—de Certeau gives us the roots of a
diegetic approach to narration. By claiming that “every story is a travel story—a spatial
practice,” de Certeau is noting that every act of narration creates a diegetic space, and
equally, that diegetic space does not exist for itself (as “a thing in a location”), but rather
as a domain for the narrative events. As he says, stories “are treatments of space”—
they utilize space and provide their temporal dimension, as when North by Northwest
turns the place of Mt. Rushmore into a diegetic space of action.
175
They do not merely
describe places, but rather narrate spaces. Further, narration implicitly activates space
even without explicit reference, as every story must entail some location in which events
occur. In the familiar tale “boy meets girl, boy gets girl, boy loses girl, boy gets girl
back,” we see no explicit reference to any locations or any temporal span, yet these
characters must exist somewhere at some time. There must be some space in which
boy met girl, some space in which he got her back. So even without deliberate attention,
the diegetic world is always implicitly evoked by the more explicit trajectory of events.
176

173
It may be tempting to distinguish “map-like” interfaces, such as those in SimCity and
Civilization, from “tour-like” interfaces, as in Half-Life or Deus Ex, yet while SimCity and
Civilization may appear map-like, we must remember that they still foreground a landscape
changing over time, and thus still demonstrate “tour” discourse.
174
And tellingly, tour discourse frequently occurs as second-person narration: “you turn right and
enter the living room.” We’ll revisit this idea in chapter four.
175
de Certeau, 122.
176
It is for this reason that Dominique Château refers to diegesis as both a mental referent and a
“world-producing activity,” using off-screen space as example. When a character leaves the
screen, we still imagine that character to exist, regardless of whether they are explicitly
represented at any given moment. The narration may even make passive reference to
characters, objects, or locations which are never seen by the audience, yet we believe them to
exist nonetheless. In this way, “the diegesis forms an implicit system, often becoming more

182
A good way to think of videogames is to switch this emphasis. Videogames must always
assert a diegesis—a space in which actants can act and events can occur—yet the
events therein may remain by-products of a sort, nothing more than “what you do there.”
Seymour Chatman observes that events, at the narrative level, are simply
“changes in state,” either brought about by an agent (“actions”) or simply occurring as
part of the world’s fabric (“happenings”).
177
In this regard, an event is a transformation of
the diegetic world, some manner by which diegetic space has moved through time.
Similarly, Tzvetan Todorov declares “two basic principles of narrative”—transformation
and causality—in that narrative entails the diegetic world undergoing changes that can
be traced to some logic of cause and effect.
178
By either’s model, we can invert the
traditional principles of narrative to focus not on events (and where they happen) but
rather space (and what happens there). Using Todorov’s two principles, Branigan
breaks down the process of narration as such:
[C]hanges in state are not random but are produced according to principles of
cause and effect (e.g. principles which describe possibility, probability,
impossibility, and necessity among the actions that occur). This suggests that
there are two fundamental kinds of predication in narrative: existents, which
assert the existence of something (in the mode of the verb “to be”), and
processes, which stipulate a change or process under causal formula (in the
mode of such verbs as “to go, to do, to happen”). Typical existents are
characters and settings while typical processes are actions of persons and forces
of nature.
179

In describing maps, de Certeau is describing a manner of spatial discourse that merely
asserts existents; without processes, their referents simply “are.” Tours, however, assert
processes onto those existents—assert changes in state. This is the work of narration.

implicit as the story progresses.” Further, we expect the world to continue beyond the specific
details we can discern from the information provided—there’s always more out there. Stam,
Burgoyne, and Flitterman-Lewis, 38-9, citing Communications 38 (1983) Special Issue on
“Enouciation et Cinéma.”
177
Chatman, Story and Discourse, 44.
178
Tzvetan Todorov, Genres in Discourse, trans. Catherine Porter, (Cambridge: Cambridge
University Press, 1990), 27-38.
179
Branigan, 4-5.

183
As de Certeau concludes, stories “carry out a labor that constantly transforms places
into spaces or spaces into places”—which is to say: they both narrate and describe
space.
180

Rather than address description as lacking the temporality of narration, Todorov
instead thinks of these two modes as entailing different manners of temporality:
Description and narrative both presuppose temporality, but the temporality differs
in kind. …[D]escription [is] situated in time, to be sure, but in an ongoing,
continuous time frame, whereas the changes that characterize narrative slice
time up into discontinuous units.
181

A described state—as in a map, or the details of a narrative prologue—occurs at some
point, for some ambiguous length of time; rather than think of such description as a
snapshot, lacking time (“this is how the land appeared at one particular moment, which
we’ll use as representative of its general state”), Todorov addresses its temporal
component more as summary (“this is what the land is like most of the time”). Narration,
on the other hand, focuses on transformation rather than constancy. In assembling a
plot, narration skips periods of invariability (periods in which “nothing happens”) and
compiles the narrative’s relevant changes in state. So while we traditionally consider
plot as “what happens,” we can equally think of it as “what changes.”
Videogames also create spaces rather than places—realms in which the player
can function as an actant—and using the Japan level of Tomb Raider: Legend, we can
readily discern space from place. In playing the level, the player navigates two adjacent
buildings, suggested to exist within the city of Tokyo. Tokyo, however, remains a mere
place, an entity in itself without a temporal dimension. Standing atop the Takamoto
building, I can look out into the city, yet nothing occurs out there; it is only a notion of
Tokyo, a notion of place. “Tokyo” never changes in the game, and offers no trajectories

180
de Certeau, 118.
181
Todorov, 28.

184
of movement or time. Yet the two buildings comprising the level, crafted in three-
dimensions, are domains allowing for movement and activity (both by myself and by
other characters). This is a space. It changes over time and exists not as “an
instantaneous configuration of positions” but instead “actuated by the ensemble of
movements within it.”
182
When Lara Croft shoots a chandelier, allowing it to come
crashing down on some enemy gunmen, a permanent change of state has occurred—
the chandelier will not return to its original state, the gunmen will not reappear to attack.
The space has moved through time, evincing a transformation—a situation that does not
occur with place.
Henry Jenkins work on videogames has sought “a more space-centered theory
of the medium” by foregrounding the worlds they offer.
183
Describing game design as
“narrative architecture,” Henry Jenkins positions videogames within a much longer
tradition of “spatial storytelling” that includes Japanese scroll paintings, amusement park
attractions, paper-and-pencil role-playing games, travel narratives, science fiction and
fantasy writing (typically preoccupied with world-building), and the odysseys and quest
tales characteristic of J.R.R. Tolkien, Jules Verne, Homer, L. Frank Baum, and Jack
London.
184
Important to the point is that none of these evince a singular manner or
methodology for spatial storytelling, yet they all privilege the diegetic space in one
manner or another, well beyond the classical model where space is so frequently
relegated to the background details of “where the plot happens.” Elsewhere, Jenkins
has also claimed that “video games constitute virtual playing spaces” which he does not

182
de Certeau, 117.
183
As he himself describes his work. Henry Jenkins, foreword to a reprinting of “Game Design as
Narrative Architecture,” The Game Design Reader, eds. Katie Salen and Eric Zimmerman
(Cambridge: The MIT Press, 2006), 670.
184
Henry Jenkins, “Game Design as Narrative Architecture,” in First-Person: New Media as Story,
Performance, and Game, eds. Noah Wardrip-Fruin and Pat Harrigan (Cambridge: The MIT Press,
2004), 121-3.

185
see as being all that different from the play-ground—barring, of course, the virtual
part.
185
And to address that virtual part, he cites de Certeau as offering a means for
discussing spatial narration outside the usual concern for plot and character. Using de
Certeau’s terms, Jenkins says that “place may be understood here in terms of the
potential contained as bytes in the Nintendo game cartridge,” and that potential is turned
into space by the playing of the game. “As I play a Nintendo game and master it level by
level, I realize the potentials encoded in the software design and turn it into the
landscape of my own saga.”
186
In this regard, we might think of the videogame-as-text
as offering a place, while the videogame-as-game (an activity performed when the
player engages that text) activates the narrative process, turning it into a space.
In other words: we should think of videogames not as narratives, but rather as
narrators.
187

Game-Texts as Narrators
Videogames are world simulators, spatial narrators, diegesis machines. They
provide spaces ripe with narrative potential, and through the player’s interaction,
generate events therein. When dealing with fictional texts, we often think of the text
itself as the narrative—as “the story told”—but a more accurate conception would be of
the text as a narrator. A narrative text is one which tells a story, projects a spacetime
apart from the ordinary world. Because traditional narrative texts are crafted to tell one

185
Henry Jenkins, “’Complete Freedom of Movement’: Video Games as Gendered Play Spaces,”
in From Barbie to Mortal Kombat: Gender and Computer Games, eds. Justine Cassell and Henry
Jenkins(Cambridge: The MIT Press, 1998), 263.
186
Mary Fuller and Henry Jenkins, “Nintendo and New World Travel Writing: A Dialogue,” in
Cybersociety: Computer-Mediated Communication and Community, ed. Steven G. Jones
(Thousand Oaks: Sage Publications, 1995), 66.
187
Anyone still uncomfortable with “narration” as our discursive model, then, could justifiably
replace “narration” with “spatiotemporal projection,” as my use of the term will be fairly
synonymous.

186
specific story, we’ve come to conflate the text with its narrative, assume that the text is
the narrative (and equally, that every viewing or reading of a narrative text offers the
same story-told). Yet the text is merely the object that tells the story, and just because a
text is able to tell multiple stories, or incredibly abstract stories, does not mean that it
becomes any less narrative. Remember our diagram of videogame discourse from
chapter one:

The game-text is not the discursive instance, but rather the thing that provides it, with the
match (gameplay itself) serving as “the story told.” The text includes the game’s full
system of rules, the entirety of its possibilities, yet the experienced content is what
manifests when the player engages that discursive object.
In truth, this is the case in any narrated instance—the diegetic world is not
“inside” the text, but is instead conjured in the mind of the audience by the text’s
representation, requiring a distinction between the narrated content and the discourse
that produces it. Such is the basis for David Bordwell’s theory of film narration, a
particularly influential model that, by virtue of its constructivist basis as well as its
attention to screen narration, can serve us well. Bordwell sees the audience’s activity as
an act of “construction” that occurs via a cognitive engagement with the text.
Responding to a variety of textual cues (the information provided by the text) and
performing a variety of mental operations (perception, hypothesizing, cognition,
association) in conjunction with a variety of “narrative schemata” (cognitive structures for

187
narrative comprehension), the spectator “constructs” the diegesis from the data
provided.
To make his case, Bordwell relies on a three-part model descending from
Russian Formalism, distinguishing: (a) fabula, the “chronological, cause-and-effect chain
of events occurring within a given duration and spatial field,” (b) syuzhet, “the actual
arrangement and presentation of the fabula in the film,” and (c) style, the text’s
systematic use of medium-specific devices. Syuzhet and style, together, constitute the
narration; the two terms simply distinguish the abstract patterning of narrative
arrangement from those technical properties specific to the medium at hand. The
fabula, on the other hand, is the story, by our use of that term—the events, actants, and
world projected by the discourse, all that is “inside” the narrative—and this, Bordwell
argues, is a commodity that is entirely constructed by the audience, a “set of inferences”
that results from the continuous process of perception and cognition as the audience
engages the narration.
188

For Bordwell, narration cues a set of mental operations, entailing both top-down
and bottom-up processes, such that we cannot easily separate perception from
cognition. As the spectator receives narrative information, they are continually making
assumptions, inferences, and hypotheses based on prior knowledge, cultural
convention, expectations, general experience, and numerous other factors, all of which
they continue to compare against further information as it is received. When the
spectator first views a character, for example, they may make suppositions based on
costuming and appearance, the star playing the role, the genre of the film, and countless
other factors, then as the film continues, the spectator will have to adjust those

188
David Bordwell, Narration in the Fiction Film (Madison, WI: The University of Wisconsin Press,
1985), 48-62.

188
hypotheses based on newer information. In this manner, the reception of a narrative is a
continual process of “filling in,” of building a conceptual model of the diegetic world and
the things that are happening there, speculating on what else may be in the world and
what else may happen, and adjusting that model as further data is received. The formal
system of the text thus “cues” and “constrains” the viewer’s construction of the story, as
narrative data suggests some possibilities while delimiting others, continuously guiding a
diegesis that is shaped by the viewer’s own imagination.
189

Among the primary criticisms of this account is Bordwell’s seeming to take the
text “for granted,” so to speak, giving little credit to the authoring process. It is for this
reason that Seymour Chatman adjusts Bordwell’s argument to say that the viewer
reconstructs the story from the cues contained within the text, giving credit to some
presence (whether the literal author or otherwise) who shapes the narrative data.
I believe that each reading of a narrative fiction reconstructs its intent and
principle of invention—reconstructs, not constructs, because the text’s
construction preexists any individual act of reading. Though reader-response
and other constructivist theories correctly insist on the active quality of the
reader’s share, seeing it as an energetic and creative act, the reader can
constitute only one-half of that actualization. There must already exist a text for
her to activate.
190

Chatman describes this as “the agency within the narrative fiction itself which guides any
reading of it” and “the principles of invention and intent which remain in the text” once
the authoring is complete, finally concluding this agency to be “nothing other than the
text itself in its inventional aspect.”
191
Though Bordwell never explicitly entertains the
notion, his model equally seems to imply the text itself as the locus of narration, as
evinced by his claim of it offering narration yet no narrator, his assertions that syuzhet

189
Ibid, 29-47.
190
Chatman, Coming to Terms, 74-5.
191
Ibid, 74, 75, 86. We will revisit these claims, along with Chatman’s distinctions of an author,
implied author, and narrator, in chapter four.

189
and style—the narration—are the properties of the text, and, if nothing else, by his
numerous references to “the text’s narration.”
It is by this model that we’ll view the videogame text’s actualization of a diegetic
world, an activity that we can credit to the interplay between text and audience. The
interactive nature of the medium, with players not only processing output, but equally
providing functional input, makes that model all the more attractive.
The text may not contain a fixed story—a narrative—but it does contain all that is
necessary to project a causally-linked trajectory of events. Even the most abstract of
videogames still narrate something, asserting the existence of objects and events within
a diegetic spacetime. Super Breakout contains none of the components that we expect
of a “story,” yet it asserts a fictional space, and a trajectory of events therein, which the
player comprehends through perception and cognition. It even provides a character:
you. Videogames narrate to the player, saying “you are in this situation…” and in turn,
the player constructs the playful belief, “I am in this situation…” So while describing
Super Breakout as “narrative” would be very confusing—outside of my prior explanation,
this statement would seem to be saying something very different from what I mean by
that sentiment—describing it as diegetic seems much less problematic. When I describe
Super Breakout as diegetic, I acknowledge its content as a spatiotemporal context
separate from my own, and can distinguish diegetic components such as the paddle and
ball (what’s inside the gameworld) from nondiegetic components such as the videogame
menu and the controller in my hand (what’s outside the gameworld).
Ultimately, while games and stories are indeed different, videogames need be
both. In game-texts, the game is the story—though we’d rarely refer to it as such. It is
the discursive content, the spatiotemporal domain projected by the text along with the
trajectory of events and the actant(s) around whom they are organized. The actant may

190
only be an inferred presence (as in Super Breakout, where the player is nowhere to be
seen and only acknowledged by the results of their actions), the spatiotemporal domain
may be severely abstracted (as in Super Breakout, where the world is a framed, two-
dimensional field), and the trajectory of events may be incredibly boring to a passive
observer (as in Super Breakout, comprised of the repetitive act of deflecting a ball), yet
these components are narrated nonetheless. No one ever said that every story has to
be a good story, after all.
What role, then, does a “narrator” have in a game? As it happens, this notion of
an organizing intelligence isn’t necessarily alien to traditional gameplay. There are at
least two figures in games with whom we can compare the textual agency of a narrator,
two organizing functions vital to videogames’ model of gameplay, and we will use these
two roles to organize the later half of this chapter. First is the referee, the omnipotent
authority of gameplay, the overseer of the play-ground. As a referee, the videogame
maintains the game’s system of rules, upholding the laws of the game. Second is the
dungeon master, the narrating figure that projects the diegetic world of the game. As a
dungeon master, the videogame operates as an interactive narrator, a storyteller
affording a second-person narrative in which players participate.

Formal Transparency and Maintaining the Play-Ground: Videogames as Referees
Try answering these three questions:
(1) What are the rules of basketball?
Looking to the 2007-2008 Official Rulebook of the National Basketball
Association, I find a lengthy 62-page document answering that very question—in
tremendous detail. Organized by topic and indexed for quick reference, the rulebook
diagrams and explains the features of the court, defines terminology and hand signals,

191
clarifies the roles of players, coaches, and officials, designates the procedures of play,
elaborates on all violations and penalties, and even describes such trivial details as
proper uniform standards. A sample regulation:
Rule No. 10… Section II—Dribble
a. A player shall not run with the ball without dribbling it.
b. A player in control of a dribble who steps on or outside a boundary line, even though
not touching the ball while on or outside that boundary line, shall not be allowed to return
inbounds and continue his dribble. He may not even be the first player to touch the ball
after he has re-established a position inbounds.
c. A player may not dribble a second time after he has voluntarily ended his first dribble.
d. A player who is dribbling may not put any part of his hand under the ball and
(1) carry it from one point to another or (2) bring it to a pause and then continue
to dribble again.
e. A player may dribble a second time if he lost control of the ball because of:
(1) A field goal attempt at his basket, provided the ball touches the backboard or
basket ring
(2) An opponent touching the ball
(3) A pass or fumble which touches his backboard, basket ring or is touched by
another player.
PENALTY: Loss of ball. Ball is awarded to the opposing team at the sideline nearest the
spot of the violation but no nearer the baseline than the foul line extended.
192

Of course, if we wished a more informal version, we could make a much simpler
description of the rules of basketball, but the rulebook provides a definitive and
unambiguous answer to our question.
(2) What are the rules of Scrabble?
Again, I can find a document clearly delineating the answer, as Hasbro’s board
game is accompanied by a pamphlet entitled “game rules.” Not nearly as complex as
the NBA’s rulebook, this brief document manages—in a mere three pages—to describe
all the procedures for turns and scoring, and even provide a few variations on the rules
for quicker gameplay. A sample:
1. The first player combines two or more of his or her letters to form a word, and
places the word on the board to read either across or down with one letter on
the center square. Diagonal words are not allowed.
2. Complete your turn by counting and announcing the score for that turn. Then
draw as many new letters as you played…

192
National Basketball Association Basketball Operations Dept., Official Rules of the National
Basketball Association 2007-2008, National Basketball Association,
http://www.nba.com/media/rule_book_2007-08.pdf (accessed December 12, 2007), 34.

192
3. Play passes to the left. The second player, and then each in turn, adds one
or more letters to those already played to form new words. All letters played
on a turn must be placed in one row across or down the board to form at
least one complete word. If, at the same time, they touch other letters in
adjacent rows, those must form complete words, crossword fashion, with all
such letters. The player gets full credit for all words formed or modified on
his or her turn.
193

Rather than a long series of regulations, the rules of Scrabble offer a step-by-step
procedural description, yet much like basketball’s handbook, this document provides
exactly the sort of exposition we expect of a game’s rules: “how to play.” Both examples
essentially provide us with instructions—a full accounting of “how to play” in its entirety.
Now, one more:
(3) What are the rules of Super Metroid?
That one’s not so easy, is it? This game also comes with a manual, and to some
degree, it does seem to resemble the NBA rulebook and Scrabble’s instructions. The
manual delineates the functions of each button on the NES controller, describes all the
components of the interface and game screen, explains procedures for saving,
chronicles the various items and creatures to be found within the game, and provides a
map of the planet Zebes. Yet this booklet hardly seems to encompass the entirety of
Super Metroid in the way that the NBA rulebook or Scrabble’s instructions do for their
respective games, nor does it resemble the sort of exposition that we usually have in
mind when we discuss the rules of a game (it seems more like the directions for
operating machinery or a travel brochure than “how to play”). Yet Super Metroid is
indeed a game, and one of our fundamental characteristics of a game is that it is
comprised by rules. What, then, are the rules of Super Metroid?
Perhaps we ought to back up for a moment by first asking: what are rules?

193
Scrabble Crossword Game Game Rules (Pawtucket, RI: Hasbro, 2003).

193
The Transparency of Videogame Rules
We’ve addressed the rules of a game in numerous regards: as the constraints
which structure and bind gameplay, distinguishing it from more spontaneous and
freeform play (ludus from paidia, in Caillois’ terms); as the formal system that creates
and governs the game’s goal-oriented conflict, the form to the content of gameplay; and
as that which allows for the emergence of play, the rigid system within which individual
players move, creating the bottom-up behavior of an individual match. Yet we never
really stopped to clarify what exactly rules are in a technical sense. Of course, we all
have a generalized sense of game rules—they’re what we are allowed and not allowed
to do during play, strictures on behavior and guidelines for the activity. When we
discuss the rules of the game, we’re usually discussing “how to play.”
Yet Katie Salen and Eric Zimmerman reveal that “how to play”—the kind of
information that we expect in an instruction manual or rulebook, and what we generically
think of as the rules of a game—are really only half of a game’s formal system. “How to
play” is what they term the “operational rules” of a game. Underneath these, however,
are the structural foundation of a game—its constitutive rules:
Operational rules are the “rules of play” of a game. They are what we normally
think of as the rules: the guidelines players require in order to play. …The
constituative (sic) rules of a game are the underlying formal structures that exist
“below the surface” of the rules presented to players. …The “true and unique
identity” of the formal system of [a game] emerges from the interaction between
these two sets of rules. …The constituative and operational rules of a game work
in concert to generate the formal “meaning” of a game.
194

While we often think only of a game’s annunciated rules, Salen and Zimmerman draw
our attention to the underlying and unspoken constraints that equally shape the formal
activity of gameplay. The operational rules of Scrabble designate the procedures
players follow for taking their turn and constructing words—this is the information that

194
Salen and Zimmerman, Rules of Play, 130, 134.

194
must be made known to the players before the game can take place, the constraints
players must agree to follow in order to have a game. A constitutive rule, on the other
hand, might be the frequency with which any given letter appears in the English
language. Qs and Zs are given immense value based on their relative rarity, thus the
statistical difficulty of using those letters are an inherent part of the game, yet players
need not be told such details in order to play. When we address the system of rules,
then, we are discussing more than simply the instructions explicitly delineated to the
player; we’re also discussing the underlying logic, physics, and statistical probabilities
that govern gameplay.
As players, we need not concern ourselves with constitutive properties because
they are “automatically” enforced by the constraints of the game and the logic of the
physical world.
195
While the operational rules of basketball need to explicitly annunciate
that players are not allowed to move out-of-bounds, must dribble the ball while moving it
down the court, and must stand behind a particular line when performing a free-throw,
they need not mention that the ball must obey the rules of gravity in order for a free-
throw to succeed, or the particulars of force, elasticity, and inertia that allow dribbling to
occur. The game could not operate without such laws of physics, of course (and the
sport’s regulators certainly have to consider such issues in determining court size, rim
height, and the material from which the ball is made), yet because they are inherent to
the physical properties of the court, the ball, and the world in general, players need not
worry over constitutive rules. The operational rules, then, are merely the rules which are

195
It is for this reason that Huizinga and Caillois neglect constitutive logic while Salen and
Zimmerman pay it attention—as players, the explicit directions are all we need worry about, while
designers need concern themselves with the structural underpinnings of the game as well. In
writing a textbook on game design, then, only Salen and Zimmerman are concerned with the
mechanics behind the game’s form, for while players need not worry over such issues, designers
certainly do.

195
not otherwise constrained by the logic of the world. In videogames, however, we see
this distinction collapse, as operational rules become part of constitutive logic.
Traditionally, the boundaries of which Huizinga and Caillois speak—the walls of
the “temporary world circumscribed by play” that isolate it from “the laws of ordinary
life”—must be maintained by players and/or designated officials. Huizinga describes the
rules as a fragile membrane because it is only by obeying operational rules that the
world of the game can continue; if a player chooses to transgress the rules, the game
would fall apart. Yet with a computational system maintaining the very world in which
play occurs, “how to play” can no longer be separated from the underlying logic of the
game, and that temporary world becomes much less fragile. Thanks to a videogame’s
ability to implement its own rules, all rules become constitutive, as anything not
permitted is simply made impossible, thus we can no longer divorce the operational from
the constitutive. In fact, any variation between these is nearly always considered to be a
design flaw—in videogames, if you’re not supposed to do it, you shouldn’t be able to do
it. If a behavior is not allowed, the formal system enforces this limitation within the
physical properties of its world, and as a result, the rules no longer need to be made
explicit to the player (precisely why so many players neglect the manual altogether).
Spatial boundaries offer us a clear example. While basketball must annunciate
out-of-bounds markers so that players are aware of where they are and aren’t allowed to
move, Super Metroid simply places physical boundaries to prevent players from going
anywhere they are not permitted—walls, mountains, locked doors, and other physical
obstacles delimit the gamespace, making “out-of-bounds” a moot point. Equally, while
Super Metroid’s instruction manual may indeed delineate operational actions, the actual
coding of those actions makes them inherently constitutive—players need not be
informed that they are allowed to jump but not to throw objects, as they will quickly

196
discover that pressing “A” will make Samus Aran jump, while throwing has not been
coded into her range of actions.
It’s not surprising, then, that while we recognize the NBA rulebook and
Scrabble’s accompanying pamphlet as the rules themselves, Super Metroid’s manual
seems more like mechanical instructions combined with a travel brochure, providing
instructions for how to operate the system along with exposition on what you can do in
the diegetic space. All three offer an exposition on “how to play,” to be sure. Yet in
basketball and Scrabble, “how to play” prescribes the very activity of gameplay—the
game manifests by players enacting those rules, an act of self-limitation in which players
agree to some activities, while swearing off others. In Super Metroid, however, the
player is no longer charged with implementing the rules, as this responsibility is now
governed by the game itself (via the computer). The rules are no longer put in the
player’s hands—and need not be annunciated at all for that matter—and resultantly, the
information in the game’s manual simply becomes exposition to help get the player
started, tips for operating the interface and negotiating the gamespace.
This move away from self-limitation creates a drastic shift. After all, the rules of
basketball tell us what is and isn’t allowed in pursuing our goal; they provide constraints.
The only reason we cannot do certain things on the court is because the rules tell us we
can’t. The rules of Super Metroid, on the other hand, provide a wide variety of things
that we can do in the gamespace in order to achieve our goal, and does not ask the
player to limit their behavior in any way. Super Metroid lets us do anything we want
within its world. Now, in truth, we need qualify that last statement: Super Metroid lets us
do anything we want that is possible within the game. Rather than imposing constraints
upon the player, any action that is not allowed simply cannot occur. As a result, we do

197
not need to consciously and willingly limit ourselves based on arbitrary rules, making
Super Metroid seem to be a game of allowance rather than limitation.
Bernard Suits addresses limitation as one of the fundamental attributes of a
game’s rules. Rather than taking Huizinga and Caillois’ view of rules as delimiting a
space (conceptual or otherwise), Suits instead addresses game rules as a matter of
voluntarily inefficiency. He sees rules as that which “prohibit[s] more efficient in favour of
less efficient means” for achieving a pre-defined goal, and argues that such inefficient
means are accepted solely for the activity that they make possible—the game itself.
196

Among his numerous examples is a footrace around an ovular track: If we take crossing
the finish line as the goal of the race, the most efficient means of achieving that goal
would be to run across the infield and head directly for the finish line. Yet this would not
be to play the game. Instead, the racers agree to take an inefficient route—the curved
path around the track—solely so that the game may take place.
By this perspective, the rules of a game are arbitrary behaviors that the player
agrees to simply because they allow the game to occur. A baseball player may run the
bases despite striking out, a chess player may grab the opponent’s king and declare it
captured on their first move, and a poker player may rifle through the deck to see what
cards will be dealt next, yet such actions would violate the rules, and thus destroy the
game. After all, the goal of the footrace is not so much to cross the finish line, but to
“win the game,” and this cannot be achieved without obeying the rules, no matter how
arbitrary or inefficient they may be. So while the rules do indeed provides limitations,
they also offer allowances—they afford activities which would not otherwise be possible.
Without rules, players may hit a ball over a fence, knock over a king, assemble certain
playing cards, or cross a finish line, yet these would be quite different from hitting a

196
Suits, 34.

198
home run, performing a checkmate, laying a straight flush, or winning a race, as those
latter actions may only happen within the rules of their respective games. We can thus
consider rules as a system of both limitations and allowances.
To be fair, videogames afford limitations and allowances just as readily as any
other game. The rules of Super Metroid could allow for countless actions which might
make the game easier (for example, moving in three dimensions), yet the game instead
limits the possibilities of player activity, just as the rules of basketball could allow players
to move the ball without dribbling, or to step out-of-bounds, but instead prohibits such
behaviors. Alternately, basketball not only limits what the player can do, it also allows
actions which would not be otherwise possible—namely, scoring points. Without the
rules of basketball, a successful jump-shot would be meaningless, as that action only
equates to “two points” when the rules shape the activity, just as Super Metroid provides
a wide range of allowances in that every activity in the game—shooting, jumping, rolling
into the “morph ball”—is only afforded by the game’s rules. The difference between
digital and non-digital games is not so much between limitations and allowances, then,
but rather in what the game provides that wasn’t there before.
In the real world, I can do whatever I want with a basketball; the formal system of
basketball thus delimits my range of possibilities. When I sit down at a blank computer
or TV screen, on the other hand, I really can’t do anything yet; it’s not until I boot up a
game that actions become available—or even a space in which to perform actions, for
that matter—thus the formal system of Super Metroid significantly expands my range of
possibilities. Without the formal structure of the game, there’d be no space in which we
can act, much less actions to perform there. Conceptually, then, the formal system of a
videogame seems to allow activity rather than delimit it, and resultantly, videogames
become characterized by freedom rather than limitation. Whereas basketball says “you

199
can do this, but not this,” Super Metroid says “do anything you want.” Super Metroid
may, in fact, prevent just as many behaviors as basketball, but it never tells us that we
can’t do them. Resultantly, videogames typically create an impression that “there are no
rules.” Because we need not police our own behavior, given free reign to do anything
possible within the world of the game, the very notion of “rules” seems to dissipate. After
all, we rarely think of videogames in terms of what we are and aren’t allowed to do,
making videogames seem like spaces where anything goes. We can address this
“anything goes” mentality as a certain transparency of the rules.
The operational rules of a game are essentially “the rules that must be stated,”
the rules that a player could violate if they wished, thus must be made aware of in order
to maintain the integrity of the game. Yet because videogames are able to maintain
such rules on their own, incorporating them into the logic of a videogame’s world, they
become transparent in that they no longer need be pushed into players’ conscious minds
as limitations; instead, the rules exist merely as the structural fabric of the world at hand.
The rules of a videogame lie largely hidden beneath the screen as invisible code, formal
properties of which the player need never be made aware. While the rules of basketball,
Scrabble, chess, poker, Monopoly, or charades must be made clear so that players do
not violate them, the rules of Super Metroid, Civilization, Metal Gear Solid, or Grand
Theft Auto remain largely unspoken.
Film scholars speak of a similar tendency in the classical Hollywood cinema,
where formal properties such as editing and cinematography are made “transparent” in
that they do not draw attention to themselves. Per Richard Maltby:
A Hollywood movie appears transparent. Its spatial and temporal conventions
work to efface themselves through their very familiarity, producing an apparently
unimpeded access to the events of the plot and their meaning in the story.
Devices such as "invisible" editing draw no attention to their contribution to the
movie's narration. At the same time, the movie's obligation to the continuous
present enhances its sense of presence and immediacy, so that it becomes

200
almost impossible to "see" the movie as an assembly of rhetorical and
descriptive strategies, producing an effect of transparency.
197

Videogames practice a similar technique, though their rhetorical strategies are not only
aesthetic conventions, but equally their underlying rule-based logic. Videogames
“naturalize” their system of rules via diegetic representation so that their limitations (as
well as their allowances) seem logical within a narrative context, what Brenda Laurel
terms “implicit constraints.”
198
Natural boundaries justify spatial limitations (walls,
mountains, locked doors, or bodies of water), plot and exposition explain objectives
(rescue the princess, uncover the treasure, save the world), and sci-fi and fantasy
contexts justify inordinate capabilities (flying, teleportation, superhuman strength).
Further, disallowed actions are generally elided by avoiding situations which might
require them—if the avatar cannot jump, designers would typically not create a situation
where leaping a two foot gap would solve a problem. In this way, players do not
consciously think of their avatar’s limitations, as those limitations never become an
issue. Most effective, however, is intuitive gameplay—control schemes which feel as a
“second nature” so that players quickly forget their conventional nature, allowing even
abstract games to practice some manner of transparency. So long as gameplay seems
intuitive and compelling so that players need not stop to question the rules, the formal
design is able to fade away, allowing the player to “forget” its conventional nature.
199

The Consequences of Invisible Rules
We can follow this sense that “there are no rules” in two directions. On the one
hand, this transparency allows a trait that has traditionally been quite alien to game

197
Richard Maltby, Hollywood Cinema, 2nd ed. (Cambridge: Blackwell Publishing: 2003), 465-6.
198
Laurel, 102-3.
199
Note that a primary exception is in those videogames which are meant to replicate non-digital
games. Sports games, casino games, board game adaptations, and other such videogames deal
in explicit rules precisely because they aim to recreate the experience of those games.

201
mechanics: ambiguity. Discussing their fundamental characteristics of game rules,
Salen and Zimmerman insist that rules must be “explicit and unambiguous”:
Rules are complete and lack any ambiguity. For example, if you were going to
play a board game and it wasn’t clear what to do when you landed on a particular
space, that ambiguity would have to be cleared up in order to play. Similarly,
rules have to be totally explicit in what they convey. If you were playing baseball
in an abandoned lot and a tree was being used as second base, ambiguities
regarding what [parts of the tree] counted as second base could lead to a
collapse of the game.
200

Yet the unspoken nature of videogame rules often leads to just such possibilities. In
Super Mario Bros., it is never made explicitly clear that the player may climb atop the
gamescreen in the underground levels and run along the bricks that frame the screen,
bypassing the obstacles that make up the level altogether. This is unquestionably
allowed by the rules—by doing so, players may find the secret warp zones, among other
things—yet this rule is never made explicit to the player. Instead, it is a rule that
designers expect players to assume to not be the case (since gameplay normally occurs
within the level, not on top of it), leaving this a silent and ambiguous possibility. In non-
digital games, boundaries need be explicitly stated, yet this instance relies on such rules
remaining unclear.
As a result, players can often be surprised by what is and is not allowed within a
game’s formal boundaries.
201
Secret areas, hidden abilities, unexpected uses for items,
and other unforeseen possibilities are rampant throughout videogames; not only is such
ambiguity possible, but it’s become a common convention in the medium, as the
discovery of ambiguous and unforeseen possibilities is often one of the great attractions

200
Salen and Zimmerman, Rules of Play, 122.
201
Salen and Zimmerman claim that this explicitness holds true in videogame rules because the
code is hard and fast, therefore, in their terms, unambiguous. However this would be to consider
the code and rules from a designer’s point of view. When we ask whether the rules are explicit,
we do not mean: are they written down somewhere? Instead, we mean: are they explicitly
presented to the player? In videogames, the answer is often no. (That rules are explicitly written
down somewhere is actually one of Salen and Zimmerman’s other fundamental characteristics—
that “rules are fixed”—and this does indeed apply to videogames just as readily as any other
game.)

202
of videogames. Yet how many instances can we find in non-digital games where a rule
is deliberately kept hidden from players as a “secret?” If we were to see a similar
instance to the Super Mario example in a non-digital context, one could easily imagine a
heated dispute arising over whether running above the screen is a viable tactic, much
like arguing over what parts of the tree count as second base. Yet in a videogame, there
is no question—the action has been permitted by the text, therefore it must be a viable
option.
202

Or must it?
There are, in fact, instances in which an action that is technically permitted by a
videogame is not meant as a viable option, when the operational rules intended by the
designers do not match the actual constitutive possibilities of a videogame’s world. We
call them exploits—programming oversights and loopholes in the system that allow an
advantage unintended by the game’s designers. Exploits offer possibilities which are
technically allowed by constitutive logic, but seem a blatant violation of operational
intent, as demonstrated by the notorious proximity mine ladder of Deus Ex. Like many
linear games, Deus Ex has numerous areas which the player is not meant to access
right away, often blocked by seemingly impassable walls. Clever players, however,
have realized that the proximity mine—an explosive device that affixes to the surface of
walls—offered an inadvertent technique for hurdling such obstacles. The mines create a
small outcropping on an otherwise featureless wall, just big enough for the avatar to
stand on, thus by placing a mine, stepping atop it, placing another, stepping atop that
one, and repeating the process as many times as necessary, players can climb any wall

202
Non-digital games can indeed offer surprising possibilities, particularly in terms of strategy—
players often discover new and novel ways of using the rules at their disposal. Yet this is more a
matter of discovering the underlying fabric of the game—the constitutive rules, rules which are
meant to remain unspoken. There are few cases where a player might discover a new
operational rule, or that the operational rules are, in fact, different from what they expected.

203
and resultantly bypass large sections of the game. This strategy was obviously not
intended by designers; it is merely an accidental result of simplified physics (the
proximity mines have no weight threshold, thus they do not dislodge, and avatars have
no center of gravity, allowing them to balance on a precariously small outcropping). So
while the proximity mine ladder is technically allowed by the game’s rules, it seems a
blatant violation of the “spirit” of the game, something we typically hold in even greater
esteem than the rules themselves.
While the formal system of a game is comprised of constitutive and operational
rules, and these are the only components we need address within the formal game itself,
Salen and Zimmerman draw our attention to a third type of rule that is not necessarily
inherent to the game, yet remains inherent to gameplay: implicit rules.
Implicit rules are the “unwritten” rules of a game. These rules concern etiquette,
good sportsmanship, and other implied rules of proper game behavior.
…[I]mplicit rules can change from game to game and from context to context.
For example, you might let a young child “take back” a foolish move in a game of
Chess, but you probably wouldn’t let your opponent do the same in a hotly
contested grudge match.
203

Salen and Zimmerman clarify that implicit rules are not part of the unique formal identity
of a game, as these unwritten rules of etiquette and “common sense” are at once similar
across many different games, yet may equally vary within the same game. Instead,
implicit rules are at the heart of what we would consider good sportsmanship and the
lusory attitude, the playful willingness to pretend that the rules matter and to voluntarily
take part in the game’s make-believe. While the Deus Ex exploit may not violate any of
the formal rules of the game, it does transgress this voluntary make-believe—after all,
the proximity mine ladder could hardly pass as plausible logic.
Here, we see the flip side of the transparency of videogame rules. As
operational rules become transparent, providing a sense that “there are no rules,”

203
Salen and Zimmerman, Rules of Play, 130.

204
players can easily become less inclined toward implicit rules as well—because “there
are no rules.” And exploits are hardly the only such instance in videogame culture.
Even more drastic are hacks—instances in which players alter the very code of
the game. While exploits take advantage of discrepancies between operational intent
and constitutive logic, hacks occur when players simply change the constitutive
properties themselves, as occurs with a popular hack for Doom 3 that allows players to
use their weapon and flashlight simultaneously. With much of Doom 3 taking place in
darkened corridors, a key feature of its gameplay is that the player may not hold a
flashlight and their weapon at the same time, creating an often terrifying catch-22—
defend yourself blindly, or provide light to see while remaining defenseless. Frustrated,
some players decided to create their own solution to the problem: a programming hack
which allows players to attach the flashlight to their weapon. Rather than finding an
unintended advantage in the game’s constitutive properties, these players simply create
their own, often just as severe a contravention to the spirit of the game—while the
flashlight mod does make diegetic sense, it violates one of the fundamental mechanics
of Doom 3’s gameplay.
Cheat codes similarly change constitutive properties—they just happen to come
pre-coded into the game. SimCity 4 offers codes to unlock rewards, add cash to the
city’s budget, or turn off power and water requirements, while Grand Theft Auto III offers
codes to manifest health, weapons, ammunition, or vehicles—both easy ways to gain
unintended advantages—yet among the most famous in videogame lore is the Konami
code, used in a variety of titles but best known for its use in 1987’s Contra. Players
began Contra with only three lives, and once these lives were spent (barring an extra life
power-up), the player was forced to restart from the beginning. With the game’s fast-
paced action and considerable difficulty in later stages, completing all eight levels of the

205
game resultantly became a fairly challenging task, thus in order to finish the game, many
players relied on the not-so-secret cheat code: up-up-down-down-left-right-left-right-B-A-
start, which, if entered during the opening title scroll, would provide the player with an
abundant thirty lives rather than the usual three—and blatantly subverted the high level
of difficulty for which the game is known. Significantly, such codes are often not
intended for player use, included instead for designers’ own testing purposes in order to
allow quick access to game features during development. Yet much like exploits,
players who discover such codes often take advantage of them, and their popularity has
led many designers to now include them for deliberate player use.
Perhaps most blatant, however, are walkthroughs and game guides—written
accounts which provide a step-by-step solution and/or reveal all the secrets and
strategies of a given title. While Super Metroid and Deus Ex provide a gauntlet of
obstacles to conquer and puzzles to be solved, walkthroughs simply offer a step-by-step
solution for expediently conquering the game. These treat the game as a problem to be
solved—and provide the entirety of the solution—rather than a game to be played, and
once again, would seem contrary to the very spirit of the game. And yet walkthroughs
and FAQs proliferate gaming websites, while published game guides have become a
profitable industry in themselves.
All of these implicit-rule transgressions, in one way of another, seem to violate
the very spirit of their games by negating either plausible logic, narrative involvement,
fundamental game mechanics, or intended challenges. Yet these are all primarily seen
in single-player experiences, those which—while perhaps detrimental to the game’s
intent—affect no one but the player who chooses to employ them. There are, however,
implicit-rule violations which occur in multi-player settings. At their mildest, these may
result from disagreements as to “fair play,” as with the practice of camping—when a

206
participant holes themself up in a corner of the gamespace and simply waits for
unsuspecting victims to pass by. While some feel this goes against the spirit of the
game, others insist that, as a tactic clearly allowed by the rules, it is a completely valid
strategy of play. At its most severe, however, we call it “griefing”—deliberately spoiling
the game for others. Whether shooters like Halo and Counter-Strike or MMOs like
Everquest and World of Warcraft, nearly every online game has its fair share of
experienced or high-level individuals who deliberately prey on inexperienced new-
comers (noobs), simply to spoil their game. Salen and Zimmerman noted that implicit
rules would typically suggest an experienced adult to “go easy” when playing a young
child; in this regard, griefing is the videogame equivalent of a professional athlete playing
a pick-up game against eight-year-olds, not pulling any punches, then taunting the losers
afterwards.
Regardless of what one thinks of these individual phenomena, what's significant
is that there's such an ambiguity over their validity as strategies for play, and that such a
vast number of gamers see them as wholly acceptable. If a poker player is caught with
extra cards up their sleeve, we would undoubtedly call this cheating, threatening to
undermine the game. If a basketball player outwardly refuses to dribble and ignores the
court’s boundaries, we would call this poor sportsmanship, threatening to spoil the
game. Yet if a Deus Ex player bypasses an entire portion of the game by using the
proximity mine exploit…well, what? Does this undermine the game? Does it spoil the
game? While we customarily agree that subverting the rules is cheating, and that
subverting the spirit of the game is poor sportsmanship, we're much less certain as to
whether exploits, hacks, cheat codes, walkthroughs, griefing, or any other videogame
transgressions are, in fact, violations.

207
The Refereed Game
We can begin to find an explanation for this attitude by looking to an exception
provided by Bernard Suits, who parenthetically notes a special type of rule different from
typical operational rules.
It is the kind of rule whose violation results in a fixed penalty, so that violating the
rules is neither to fail to play the game nor (necessarily) to play the game well,
since it is sometimes tactically correct to incur such a penalty…for the advantage
gained. But these rules and the lusory consequences of their violation are
established by the [operational] rules and are simply extensions of them.
204

The type of instance to which Suits refers are those we often find in professional sports.
If a chess player illegally moves a piece while the other player is not looking, we would
not hesitate to call this cheating. Yet if a professional football player performs an illegal
hold, we would not think of it in the same way. Instead, a penalty is assessed, keeping
the violation within the operational rules, and the game is able to continue.
205
Because
the formal system of the game has been designed to account for such violations, it is
able to weather these transgressions and thus continue unabated, making it much less
susceptible to cheating and poor sportsmanship than other games (note that football
also provides penalties for “unsportsmanlike conduct,” and nearly every sport provides
conditions under which a player may be ejected for severely detrimental behavior).
Perhaps even more telling, however, is that if the player is not caught performing the
illegal hold, and thus not appropriately penalized, the game is still able to continue, with

204
Suits, 38. Please note: Suits uses the term “constitutive rules” in a manner that is moderately
equivalent to Salen and Zimmerman’s discussion of “operational rules.” Since I’ve already
established Salen and Zimmerman’s terms, and do not wish to cause undue confusion based on
the superficial semantic differences between the two, I’ve taken the liberty of replacing Suits
mention of “constitutive” rules with the bracketed term “operational” in order to keep our terms
straight.
205
Note that nearly every professional sport has common instances where a rule-violation is
deliberately chosen as a strategic option, with players and coaches fully expecting to be caught
and penalized for the infraction. A losing basketball team may deliberately foul an opposing
player in the waning seconds of a game in order to get the ball back, a hockey player may ice the
puck to prevent a possible goal, and a football player may take a deliberate pass interference call
to prevent an assured touchdown. Yet none of these damage the game in the way that cheating
would.

208
the outcome still considered valid. In this instance, the violation has not been accounted
for, and yet no one labels the player a cheat—and few would claim the player as a spoil-
sport, either. Yet this is neither the result of transparency nor an “anything goes”
mentality, since football openly foregrounds all of its formal structures. Instead, it is the
result of the sport’s refereed nature.
The fundamental difference in these special cases to which Suits refers is that
players are no longer held responsible for maintaining the system of the game. Non-
refereed games are maintained by the players themselves, placing the responsibility of
proper behavior—whether that of annunciated, operational rules or unspoken, implicit
ones—on the players. Players must police what they can and can’t do as well as what
they should and shouldn’t do, as they are held responsible for both cheating and good
sportsmanship (in a friendly pick-up game of football, an illegal hold would probably
seem much more egregious). Yet when an outside agent becomes responsible for
upholding the operational rules of a game, our attitude toward the rules—including
implicit ones—changes.
206

We see a similar dynamic in videogames. With the computer able to maintain
the world of the game and enforce all formal properties of the game, videogames
become refereed affairs. Any behavior that is allowed is seen as viable, and the
distinction between can and should all but disappears (that is: if you can do it, why
shouldn’t you do it?). Much like an illegal hold that is not called, videogame exploits and
griefing are accepted because the officiating entity—the game itself—has not

206
Also notable is that while there are still plenty of examples in pro sports which we unanimously
acknowledge as cheating—using performance-enhancing drugs, unlawful equipment (corked
bats, etc.), or illicitly-obtained information on the opposing team—nearly all of these must be
couched under the explicit, operational rules of the game. Numerous performance-enhancing
substances, for example, were considered legal until officiating bodies began to explicitly ban
them. Yet there are few actions that are not explicitly forbidden that would still be considered a
violation of the game. Which is to say: in a refereed environment, it is still possible to cheat, yet
very difficult to be a spoil-sport.

209
disapproved it. Hacks and cheat codes, on the other hand, become the digital
equivalent of “house rules”—a tweak to the formal structure which basically offers a
variation on the official game, yet is nonetheless approved by some sanctioning body.
Salen and Zimmerman note referees as “extensions of the formal system of a
game,”
207
and in this regard, they become the maintainers of the play-ground. Referees
are the figures who police the game’s margins and uphold its system of rules, the
supreme authority of the game’s internal reality. In such a situation, rules can quickly
become less about regulating one’s own behavior and more about testing the
boundaries of the referee, seeing what the referee will allow. We’ve already observed
that play is fundamentally about testing boundaries, experimenting within the borders of
the rules to see what possibilities lie therein; no longer bound by the responsibility of
upholding the rules themselves, players are simply given more license to aggressively
test those boundaries in a refereed environment.
As refereed spaces, videogame worlds become an internal reality that we need
not question nor consciously attend to. Affording a transparency of rules, they naturalize
the game’s logic, allowing the player to “forget” the game’s conventional nature—an
unlikely occurrence in traditional gameplay—and focus instead on the fictional universe
created by those rules. And in this fictional universe (as opposed to the ludic universe of
the play-ground), the videogame operates as a very different sort of authority.

Participatory Storytelling and Narrating the Diegesis: Videogames as Dungeon
Masters
Gathered around a table, [players] would listen as the Dungeon Master cracked
open the D&D rule book—which contained descriptions of monsters, magic, and
characters—and fabricated a scene: down by a river, perhaps, a castle shrouded
in mist, the distant growl of a beast. Which way shall you go? If the players

207
Salen and Zimmerman, Rules of Play, 276.

210
chose to pursue the screams, the Dungeon Master would select just what ogre or
chimera they would face. His role of the die determined how they fared; no
matter how wild the imaginings, a random burst of data ruled one’s fate. It was
not surprising that computer programmers liked the game or that one of the first
games they created, Colossal Cave Adventure, was inspired by D&D.
208

While the game-text’s role as referee concerns the maintenance of the gameworld, we’re
here concerned with its role as a storyteller in a more traditional sense—a narrating
presence that projects a trajectory of events within a diegetic context. If the iterative of
gameplay is indeed an act of narration, then the game-text is the narrator who projects
that fiction, and in this regard, the videogame functions not merely as a referee, but as a
dungeon master.
Often credited as immediate forerunners to videogames, table-top role-playing
games such as Dungeons & Dragons afford their players the chance to explore fantastic
environments and encounter all manners of mystical creature—much as I traipsed
across the landscape of Oblivion—all under a rigorously structured rule-set of statistics,
probabilities, and dice throws. Players donning the roles of “player characters” (avatars)
serve as the heroes of these adventures, gathering around tables covered in maps,
graph paper, die-cast miniatures, and an assortment of multi-sided dice, and enacting
personas rooted in elaborately determined character stats. They slay monsters, cast
spells, and hoard treasure, traversing worlds rooted in both imagination and elaborate
rules. Yet it is the Dungeon Master who serves the most pivotal role of the campaign.
The stars of [role-playing games] were the Dungeon Masters, the storytellers
who devised the adventures. The best game leaders could transport a room full
of players sitting in a living room in Houston back in time to an ancient place
where anything was possible. The only limitation was the imagination of the
players.
209

208
David Kushner, Masters of Doom: How Two Guys Created an Empire and Transformed Pop
Culture (New York: Random House, Inc., 2004), 7.
209
Brad King and John Borland, Dungeons and Dreamers: The Rise of Computer Game Culture
(New York: McGraw-Hill/Osborne, 2003) 20.

211
The Dungeon Master is a referee of sorts; much like a sporting umpire, the DM
maintains the rules and holds all ludic authority (some RPGs even use that very term to
describe this central moderator). Yet “referee” alone does a disservice to the Dungeon
Master’s trade. Certainly, it is the DM’s responsibility to know the game’s rules inside
and out, and it is for this reason that this role is often assumed by the most experienced
of players. Rolling dice, tabulating statistics, and serving as the final authority over the
interpretation of the rule book, the DM is the arbiter of gameplay. Yet as King and
Borland’s above quote indicates, the DM is foremost a storyteller, and it is this latter role
with which we are here interested.
210

A specialized, rule-based manner of interactive narration, we could reasonably
call dungeon mastering the most refined manner of participatory storytelling to come out
of the twentieth century. Though himself using the term “referee,” Gary Alan Fine notes
that the much more common metaphors for the DM’s role are terms like “storyteller,”
“playwright,” or in many cases, simply “God.”
The “world” is, after all, his creation—he is “God” in that he creates the world in
which his players must survive; he maintains ultimate interpretive authority. …He
chooses how the game will be constructed, both in terms of the setting and the
scenario. In theory he is the dreamer; he is in control.
211

In this sense, the dungeon master is a narrating deity-figure, the omnipotent arbiter of
the fantasy world. It does not seem much of a stretch to apply that same metaphor to
the videogame itself, operating in the DM’s tradition of projecting a participatory universe
in which players imagine themselves to be active. Our goal in this section is to consider

210
In truth, “dungeon master” is actually a brand-specific term, originating in the Dungeons &
Dragons franchise and still commonly associated with that series (or at its broadest, with general
swords and sorcery RPGs). The more generic variant is “game master,” and other franchises will
at times refer to this role as referee, judge, storyteller, or some similar term. Yet for the sake of
both the evocative nature of the term— focusing specifically on the space that is conjured—and
the canonical role of the Dungeons & Dragons brand, I’ll stick with “dungeon master” as our
metaphor of choice. To indicate my generic usage, please note that from here on out I’ll refrain
from capitalizing the term.
211
Gary Alan Fine, Shared Fantasy: Role-Playing Games as Social Worlds (Chicago: The
University of Chicago Press, 1983), 72.

212
the game-text in this regard, as the dungeon master of the game—not just the referee
who maintains the rules, but the storytelling presence who projects the diegetic world.
As dungeon master, the game-text projects a fictional world that is separate from
the player’s here-and-now, invoking the two world logic of narration. As Bordwell and
Branigan both suggest, we can identify three fundamental dimensions necessary for the
projection of a diegetic world—space, time, and logic—and when engaging a videogame
world, we always encounter these via two distinct though simultaneous frames of
reference—the systemic context of gameplay and the diegetic context of a story world.
And much of the structure of videogame narration comes from the arrangement of and
relationship between these two contexts. So to consider the dungeon mastering role,
we’ll address each of these three issues: (1) the manners by which we construct a sense
of diegetic space in relation to the images we see on screen; (2) the manners by which
we construct a sense of diegetic time in relation to the actual time of gameplay; (3) and
the manners by which we construct a sense of diegetic logic in relation to the actual
rules of the game. And since we can hardly address a diegetic world without
acknowledging the trajectory of events that occur therein, we’ll also consider (4) the
manners by which narrative events result from the gameworld.
212

212
One might question my choice to deploy and adapt concepts of traditional narrative,
particularly cinematic narrative, to discuss the manner by which videogames project their worlds.
A fair argument could be made that videogame narration is so fundamentally different that
traditional models distract us from their uniqueness (and as I’ve repeatedly made clear, I wish to
avoid this very pitfall). However, we cannot avoid the undisputable reality that for the vast
majority of us, the experience of cinema and television will always precede our engagement with
videogame space, and that we cannot help but be informed by their basic similarity of delivery—a
three-dimensional world represented on a two-dimensional screen. The heavy use of cinematic
and televisual techniques in countless videogames evince that we cannot divide videogame
narration entirely from our most culturally familiar strategies of screen narration. We have a long
history of identifying with characters on screen and experiencing time and space in the disjointed
and fragmented assemblage of cinematic narrative, and as such, traditional screen narration
becomes one of the many cognitive schemas that inform our understanding of videogame
discourse. Proffering an entirely new mode of diegetic projection would be to pretend that
videogames are experienced in a medial vacuum. And they certainly are not.

213
Space
We’ll begin with space, the most primary component of the diegetic world. We
construct the diegetic space of a videogame via a variety of cues—visual perspectives,
the arrangement of disjointed images, auditory information, and various modes of
exposition. Katie Salen and Eric Zimmerman note:
The structure of a digital game space always grows directly from the formal
system that defines the game. However, the space that a player experiences is
also a function of representation (how the space is displayed to the player) and
interaction (how the player navigates through the space). These three
elements—formal structure, structure of display, and interactive structure—
together constitute the experience of a digital game space.
213

In its deployment of space, a videogame creates spatial relations between existents so
that the player can conceive of a coherent and consistent space. When a player
interacts with the door of a building, for example, they will typically perceive the ensuing
screen as existing “inside” that building. In this respect, players construct a mental
impression of the game’s space through their perception of and interaction with the
game’s representations.
The most familiar example of constructing diegetic space onscreen is the
cinematic practice of continuity editing: cutting between multiple camera perspectives,
using numerous cognitive cues (such as sightlines, theatrical use of the 180-degree
plane, objects that appear in both frames, and diegetic context), in such a manner that
the spectator is able to assemble these separate images into one coherent sense of
space. Videogames also assemble space, though frequently in very different manners,
and below, we’ll consider a few of the ways in which videogames do so.
In addressing videogame space, we are similarly dealing with the “camera” of the
game—not exactly the correct term, but an appropriate metaphor, due to our familiarity
with the device. Yet unlike cinema, our understanding of space is often further

213
Salen and Zimmerman, Rules of Play, 394.

214
complicated by our relationship to the avatar (an issue we will examine further in chapter
three), as the player must create a coherent understanding of space while equally
functioning within it. As a result, one of our first concerns in constructing a videogame
space is in where we are in that space, requiring attention to the locus of our visual
perspective (the “camera”) as well as the location of our body (the avatar)—while a
subjective point-of-view from behind the eyes of the avatar, as in Half-Life or Halo,
simplifies matters by making the player and avatar’s positions simultaneous, this is far
from the only manner by which we apprehend the gamespace. As such, spatial cues tell
us not only how space is arranged, but also how we relate to it as a player and where we
are in relation to our avatar, whether inside them (internal POV), behind them (trailing
perspective), beside them (side-scrolling perspective), over top of them (overhead
perspective), jumping in space around them (cinematic perspective), or by some other
relationship.

Continuous space:
When considering how we arrange spatial details, simplest and clearest are
those instances which simply offer one continuous perspective, so that any two locations
are connected by a continuous expanse of space between them, whether experienced
through an internal point of view (Half-Life, Halo) or any manner of disembodied camera
(Tomb Raider, Contra, The Sims). In any of these instances, space is created by a
continuous movement through it—or more precisely, by that space continuously moving
past the player. As the landscape shifts on screen, the player is given an impression of
expanding space, revealing more of the diegetic world and providing seamless continuity
between locations.

215
Adjacent space:
Adjacent “rooms” are among the most common manners by which we relate two
discontinuous perspectives, as when an avatar exits screen right, then enters screen left
in a different area, creating the impression that the two spaces are continuous (as in
Adventure, King’s Quest, Metroid, or Metal Gear). These need not always connect literal
rooms; adventure titles like King’s Quest and The Secret of Monkey Island often build
large open spaces (forests, towns, etc.) from a lengthy series of adjacent screens, yet
the room analogy provides an appropriate description of these individual units—each
screen operates as a bounded and self-contained space, with some manner of passage
(typically at the borders of the screen) connecting them, much like doorways. Adjacent
space is also frequently used between levels, as in F.E.A.R., where each level is
suggested to continue in the precise spot where the last left off—conveniently facilitated
by some physical barrier preventing the player from returning back from whence they
came. Though the separate rooms or separate levels are technically entirely separate
arenas (as far as the programming is concerned), the player retains the impression of
continuous movement through one consistent domain.

Nearby space:
Somewhat of an amendment to adjacent spaces, nearby spaces use similar
techniques, so that players retain a sense of proximity and direction, though the
transition between the two spaces is clearly not instantaneous. We see this in adventure
titles like The Longest Journey, when leaving screen left from one area instantaneously
leads to appearing screen right in another, yet those two spaces clearly could not be
continuous. If April (the player’s avatar) begins on the screen which depicts the front of
her boarding house, she can exit to appear in a second space containing a series of

216
bridges over the canals of Venice. Based on the similar environment—urban setting,
brick buildings, etc.—we could plausibly believe these two to be adjacent spaces, though
one would suspect some distance to exist between them. Yet once April exits this
space, she appears in a lush park. Filled with trees, grass, and open space, we’d have
a hard time believing that the park is a continuation of the urban bridge screen; to make
logical sense of this transition, we’d have to assume some lapse in space (and time) to
occur in the transition between these areas. If April continues through the park, she
could exit that screen to appear at the Venice Academy of Visual Arts, and though the
environment is indeed similar to that of the park, it again seems unlikely that these two
locations are immediately contiguous.
Rather than adjacent spaces, these are nearby spaces: the player maintains a
sense of direction and proximity so that these areas still comprise one consistent
diegetic space—to get to the academy from the boarding house, one would head to the
bridges, then through the park—yet the player assumes some expanse of extraneous,
unrevealed space to connect these areas. It’s particularly noteworthy that to help
facilitate these transitions, The Longest Journey often uses distance or passageways to
veil the “cut” between such spaces, eliding the suggestion of continuity editing via a brief
distancing from the character—when April exits the bridges, she disappears behind a
wall before the transition, and when leaving the park, she walks off into the distance. As
such, the game suggests “April walked off into the distance, and eventually arrived at the
academy” rather than use a more direct cut that would imply an instantaneous transition.

Embedded space:
Another common technique is the embedding of space, as when moving from an
exterior to an interior. We see this commonly in “open world” games, where the player

217
navigates one large, open and continuous space, yet the movement indoors occurs
through a transition to a new space. While exploring the land of Cyrodil in Oblivion, I
might arrive at the north gate to the city of Anvil. Anvil is surrounded by an immense
wall (as is each city) so that I cannot directly enter the city, yet upon interacting with the
gate, I’m given a brief load-screen, then find my avatar just inside the city’s gate.
Similarly, I could then approach the door to the Fighter’s Guild, and after another load-
screen, I would myself inside that building. Within the structural fabric of the game,
Cyrdoil, Anvil, and the Fighter’s Guild are all autonomous spaces—evincing this fact, the
interior of each location is significantly larger than their exterior would actually allow.
Moving between these locations is essentially an act of teleportation, disappearing from
one space and re-appearing in another. Yet via a sense of embedded space—that the
guild is inside Anvil, which is in turn inside Cyrodil—we understand a spatial relationship
between these areas, allowing us to suspend our disbelief and conceive of one unified
land of Cyrodil.
Doors are a common trope for connecting both adjacent and embedded spaces,
an additional cue to help the player make sense of two seemingly related, though in fact
entirely disjointed, areas. Some games, such as Resident Evil, will even include the
sound of a door opening and closing to help strengthen our connecting of the two.

“Cinematic” space:
Of course, we cannot overlook that some games use the common tropes of
cinema and continuity editing—not only in cut-scenes, but equally in gameplay. Whether
assembling a series of fixed angles, as in Resident Evil, or employing a floating camera
that only intermittently changes perspective, as in God of War, numerous games use the
conventions of screen direction and continuity that cinema has perfected over the past

218
hundred years. These are cases in which multiple perspectives are used not to join
together two separate spaces, but rather to represent the same space (or at least parts
of the same space) from multiple angles, requiring the player to resituate themselves in
relation to the space at hand.

Player-controlled perspective:
A markedly different manner by which we construct space is through our own
willing examination of space when provided a player-controlled camera or the option to
change perspectives at will. These instances allow the player to request more spatial
information, to assemble a sense of space by their own desire. In Ratchet and Clank,
the player’s perspective trails the avatar of Ratchet as he runs about the gamespace, yet
the game offers two manners by which the player can examine space. First, the player
may pivot the camera around Ratchet, independent of where he is facing; while the
camera’s focus remains centered on the avatar, the player may move around him in 360
degrees, allowing a better sense of the area surrounding him. Additionally, the player
may also switch to Ratchet’s point-of-view by pressing the L1 button on their controller,
allowing a new perspective of space. In this instance, the player assembles two
discontinuous perspectives by their own volition, already knowing the spatial relationship
between the two points of view and using that knowledge to unify the spatial data.

Mapped space:
Maps frequently help our sense of space by indicating the spatial relations
between locations, whether discontinuous or not. The world map of Super Mario World
and the galactic map of Mass Effect both allow the player to choose a travel destination,
and serve as the interface by which the player moves from one segment of the game to

219
the next, yet they also provide spatial exposition, indicating the proximic relationship
between locations and connecting each destination within the diegetic world. The
overhead map that is accessible from outside the diegetic space of many games
(typically in the pause menu or some similar non-diegetic component of the interface)
has similarly become a standard trope as gamespaces have become more unwieldy,
allowing the player to acclimate themselves within a sprawling diegetic world and avoid
the very real possibility of getting lost—trying to navigate a vast open world like Grand
Theft Auto: San Andreas or Oblivion without any map to orient yourself would quickly
become a frustrating and confusing exercise. The “mini-map”—a small representation of
the player’s immediate vicinity accessible during gameplay, typically at the edge of the
frame—has become equally common, further assisting the player’s spatial orientation.
In some cases, these maps may have a diegetic bearing—the protagonist of Dead
Rising is given a map by another character, while Mass Effect’s hero accesses the
galactic map from the command deck of their ship—however maps frequently exist
solely for the player’s benefit, justified simply for allowing a better understanding of the
diegetic world.
Maps provide a different cognition of space than our usual experience. This was
one of the crucial distinctions that de Certeau made between maps and tours, for while
we generally experience space in a tour-like fashion—moving temporally through it,
experiencing the space from a singular point of view—maps instead abstract
configurations of space, providing a generalized conception of proximities and spatial
relations outside of a subject’s point of view; in many cases, the map may not even
provide accurate proximities, instead only providing directional relationships. Tours
provide what psychologists term “egocentric space,” space oriented around one’s own
spatial position, providing a “personal positional reference scheme” (if you stop, right

220
now, and think of where the nearest restroom is, you’re probably utilizing egocentric
space, defining its position in relation to where you are situated).
214
Cognitive mapping,
on the other hand, provides a different manner of organizing spatial information,
affording an awareness of the environment far beyond our field of perception (rather
than thinking specifically of the nearest restroom, imagine the general layout of the
building as a whole).
215
We form cognitive maps even without literal maps—our ability to
devise a shortcut, for example, demonstrates an ability to abstract spatial relations
outside of personal experience. Yet literal maps help fill in this data, isolating larger-
scale spatial patterns outside of a subject’s perspective and providing a tangible
foundation for cognitive mapping. As such, maps become a way to represent “off-screen
space,” calling to mind areas of the diegetic world away from the player’s immediate
position.
We should note that mapped space differs from the worlds of SimCity or
Civilization, games whose very presentation are “map-like.” While these do entail many
of the same functions as true maps—abstracting space, presenting it in an expansive
view that is detached from first-hand experience—the map-like layout does not exist to
orient the player within a separate perspective, but rather is the very world of the game.
While these games may call to mind other representations, the abstracted terrain is still
the world in which the player is active, the very space to be mastered (these games do,
however, feature their own sub-maps to represent the entirety of the space in an even
further abstracted fashion).

214
Philip Winn, Dictionary of Biological Psychology (London: Taylor & Francis: 2001), 30.
215
Ibid, 163.

221
Abstracted space:
Much like Adventure’s avatar—the non-descript square representing a knight that
we used in chapter one—Adventure’s environment offers an equally abstracted set of
signifiers that we “read” as indicative of a much more complex referent. While the player
navigates Adventure’s minimalist maze of pixilated rooms, they comprehend this space
via an understanding of the context it represents: castles, labyrinths, dragons, swords,
and all the trappings of medieval adventure. Equally, the heavily-pixelated worlds of
Frogger and Battlezone, or the animated, surreal worlds of The Legend of Zelda: Wind
Waker and Katamari Damacy all exist simultaneously as an abstract play-ground but
equally a diegetic world that we comprehend via their abstracted representations. In any
of these cases, the diegetic world is conceptually much more than the reductive or
cartoonish representations that depict them, existing in the mind’s eye just as much as
they do on the screen.
We construct our sense of diegetic space not only in the spatial relationship
between separate areas, but equally in the representational relationship between the
screen’s signifying space and the signified space of the fictional world. We covered
semiotic terms in chapter one, so we need not rehash these issues, yet it’s worthwhile to
draw our attention to a few of the vital issues involved in how we conceptualize the
“reality” of what we see on the screen.
Of course, nearly every gamespace is abstracted to some degree, based solely
on the nature of the screen—a two-dimensional space. Resultantly, depth cues are
often vital—vanishing points, lines of perspective, movement, changing dimensions of
size, shadows and lighting, and other indicators that allow us to infer spatial relationships
based on our experience perceiving the world. These may occur via three-dimensional
rendering, functioning in a similar manner to a photographed image (at least, as far as

222
the player is concerned), though this is by no means the only manner for affording depth.
Two-dimensional side-scrollers like Contra and Metroid still animate their worlds to
suggest depth, even if the actual action occurs in a flat, 2-D plane.
Space is also frequently abstracted in terms of size—Oblivion’s continent of
Cyrodil takes up the real-world equivalent of roughly 16 square miles—a large chunk of
land, to be sure, yet hardly the size we’d expect of a continent. Heck, most cities are
larger than 16 square miles, yet this relatively small patch of land represents a continent
filled with mountains, rivers, roads, fields, a sizeable bay and nine full cities. We see a
similar abstraction of size in the game’s use of embedded space, with interiors markedly
larger than the exteriors that are meant to contain them. In other cases, games may
choose a relatively small arena of action to indicate a much larger world, as in Mass
Effect, where moderately small patches of land are used to convey the general
landscape of entire planets. As videogames attempt to represent larger and larger
spaces, the practicalities of exploring those enormous spaces begin to require some
degree of abstraction, as that desire for enormity clashes with the constraints of a
containable game.
But perhaps most indicative of spatial abstraction are “unreal” spaces, those
which, recognizably, could not exist in the world as we know it, demonstrated in non-
Euclidean spaces that defy the conventional logic of continuous space. The
“wraparound” space of Ms. Pac-Man or Asteroids—where an object exiting one edge of
the screen immediately re-enters on the opposite side—as well as the nonsensical
geography of Adventure’s black labyrinth and The Legend of Zelda: Ocarina of Time’s
forest maze (both of which defy any sense of spatial relations) all demonstrate spaces
that could not occur in the world as we know it, requiring a non-traditional conception of
space. We still encounter these with our standard sense of spatial logic, yet their design

223
requires a break from some of that logic’s principles in order to make sense of our
navigation through them.

Verbalized space:
Some games, of course, are built on an entirely different manner of abstraction—
language. We might call this “verbalized space.” We shouldn’t assume only pictorially-
rendered spaces to comprise videogame worlds. A vast range of hypertext games have
provided lavishly-detailed spaces, from the fantasy worlds of Zork and Colossal Cave
Adventure to the sci-fi landscapes of Planetfall and The Hitchhiker’s Guide to the
Galaxy. Yet wholly text-based games are not the only games to use verbalized space.
Role-playing games like Wasteland, Fallout, Baldur’s Gate, and the earlier
entries in the Ultima and Final Fantasy franchises all rely on textual description and
exposition to compliment the relatively simple images seen on screen. Wasteland, for
example, really only depicts an overhead map of the landscape to indicate a generalized
spatial arrangement—the vast majority of the game’s action, characterization, and
spatial details come from textual descriptions of the avatars’ surroundings and the
diegetic goings-on. All the details of the story are explained via brief, in-game passages
of text, along with much lengthier passages that must be read from an accompanying
booklet; even the game’s turn-based combat occurs entirely through textual commands
and descriptions. And such verbal accounts of space are hardly limited to a bygone era
of rudimentary graphics and limited data storage. 2007’s Mass Effect, while offering a
sprawlingly large universe rendered through state-of-the-art graphics, relies on copious
amounts of textual description in order to forgo exploring every planet of the galaxy, as
many planets exist only in text-based descriptions and brief narrative passages (e.g.,
“while exploring the mountainous planet, you discover a substantial mineral deposit…”).

224
Spatial exposition:
Related to verbalized space is spatial exposition, instances in which
discontinuous spaces are not necessary linked by a specific spatial representation, yet
the exposition surrounding gameplay provides a diegetic link between the two. We see
this in each installment of the Halo franchise, as the exposition between levels justify
why the ensuing level will occur in a new location and how the Master Chief arrived
there, thus establishing diegetic links between the different locations and their position
within the same diegetic world, yet their specific spatial relationship—along with the
generalized configuration of the Halo universe—remains ambiguous.
In many cases, we may be able to rely on our familiarity with specified locations
to create a cognitive map—as when Midnight Club III names its three cities San Diego,
Detroit, and Atlanta, allowing us to situate those spaces within our own familiarity with
the continental United States—yet what’s most important is the diegetic justification for
being transported to an entirely new domain, thus linking each area of the game within a
consistent diegetic world. Sly Cooper and the Thievious Raccoonous is built on a series
of locations around the glode, with each episode centering on different villains and their
own story-arc, such that one could conceivably look at each episode as a separate
game in itself, yet the game’s exposition links the separate locales in a unified and
sequential diegetic frame, providing a sense of one continuous world.
As videogame spaces have become larger and larger, numerous games have
chosen to offer the player expedient means for traversing the landscape, a process that
Obilivion calls “fast traveling,” a term we might use for the broader practice of offering
expository justification for, essentially, teleporting between locations. In Oblivion, fast
traveling is simply a way of saying “…and then your avatar spent twelve hours walking to
Anvil,” allowing the player to quickly arrive in a new location while equally justifying the

225
movement within the diegetic context. In a similar manner, the Grand Theft Auto series
allows player to board the subway to quickly move between locations. Yet this device
need not only be applied to open world games. Countless games use some device of
transportation to move the player between individual levels, as when Bioshock’s player
enters the blathysphere (an underwater subway of sorts) to indicate the movement from
one location to another. Jak and Daxter and Super Mario Galaxy cartoonishly
acknowledge the device for what it is—a magical teleporter—as the former lets Jak jump
through mystical portals while the latter sends Mario implausibly hurtling through the
galaxy at the speed of light to land on a new planet. Regardless of the context, these
are all instances of spatial exposition, providing a plausible context to justify the player’s
instantaneous movement between discontinuous locations.

Unexplorable space:
In a similar vein, we should acknowledge the common practice of showing
extended space that cannot actually be “touched” in any way. When playing the two-
dimensional Super Mario Bros., players see mountains, plains, and clouds in the
distance, adding depth to the environment, and in Super Mario World, these background
objects even exist on independent planes so that the avatar’s movement cues changes
in perspective, providing an even greater sense of three-dimensional space. Yet the
spaces of these animated backgrounds can never be explored or experienced—they’re
merely window dressing, much like a painted backdrop on a theatrical set (Nintendo’s
Paper Mario series self-consciously engages this very metaphor by depicting its
backgrounds, literally, as two-dimensional sets). In such 2-D environments, these
backgrounds afford a sense of three-dimensions, fleshing out the diegetic world to make
them appear “deeper”—in both the literal and metaphorical sense—and thus more

226
immersive, yet these background spaces do not actually “exist” within the play-ground of
action. We saw a similar practice in Oblivion as I looked down the hillside past the
invisible wall at the border of the game space. This space may be rendered differently,
appearing in three-dimensions and seeming just out of reach, yet just as in the Mario
games, it still exists as unexplorable space—space that doesn’t exist within the play-
ground’s space of possibilities, but rather only in the diegetic world in which we imagine
the space to continue further into the distance, beyond the borders of gameplay.

Other manners of perception: sound and “touch”
Our chief concern here has been in our visual apprehension of space, yet we
shouldn’t overlook our other manners of sensing our environment. Sound is frequently a
vital cue to constructing space; when changing perspectives, a continuing soundtrack of
ambient noise will immediately create an impression that the player still occupies the
same locale, indicating that the change in camera is merely providing a new angle on
the same space (and equally, a sudden change in ambient noise will indicate an entirely
new location).
Yet we also need consider how the space “feels”—that is, how we’re able to exist
within it, as our impression of the gamespace is equally informed by our interaction with
it. The vast difference between unexplorable space and the rest of the game is that we
can never feel the former, don’t get a sense for its boundaries by our testing of them.
Our interactions are nearly always facilitated through an avatar, yet the manner by which
we interact may vary, from the deeply immersive sense of complete control over a game
body to the much more detached sense of interaction that occurs through typed
commands or point-and-click interaction. It is through the avatar that we’re able to touch
the diegetic world, get a sense for its spatial dimensions and make sense of the world

227
beyond merely what we see. Certainly, Civilization—where the player provides
commands to individual units but never directly interacts with the environment—projects
a different manner of space than those games in which the player navigates the
environment through an individual avatar. Equally, Dead Rising’s highly interactive
environment, chock full of usable items, feels much different from Resident Evil, where
the majority of objects are simply painted onto the surroundings.

Time
Mark Wolf observes that “cinema rendered time more malleable than it had been
on the live theater stage, but the video game presents even more possibilities for
temporal structuring.”
216
Indeed, while one might expect videogame time to occur in a
fairly straightforward manner, the medium has demonstrated some intriguing
manipulations of diegetic time. We’ll examine a few of them.
Narratologist Gerard Genette offers one of the more definitive accounts on the
relation between story and discourse time, identifying three ways in which the two relate
to one another: order, duration, and frequency. We will address each of these, yet also
add two more: abstraction and tense.

Order:
While chronology is often among the most interesting formal structures by which
a traditional narrative arranges its content, it’s actually among the least interesting for
videogame discourse, as games tend to occur in their natural order. Jesper Juul
explains:

216
Mark J.P. Wolf, The Medium of the Video Game (Austin, TX: University of Texas Press, 2001),
77.

228
Notwithstanding inspirations from cinema, time in games is almost always
chronological, and there are several reasons for this. Flash-forwards are highly
problematic, since to describe events-to-come would mean that the player’s
actions did not really matter. Using cut-scenes or in-game artifacts, it is possible
to describe events that led to the current fictional time, but an interactive
flashback leads to the time machine problem: The player’s actions in the past
may suddenly render the present impossible.
217

All the same, there are quite a few videogames that do indeed rearrange temporal
order—Call of Duty 2 opens by putting the player in the midst of the Normandy Beach
invasion before jumping backwards in time to the bombing of Pearl Harbor, while
Resident Evil 2 has the player move chronologically through the events of one evening,
but then jumps back to the beginning in order to provide a parallel series of events via a
different avatar, thus repeating the same diegetic span of time from a different point of
view. Such examples, however, rely on a strictly linear series of events in order to
guarantee that they will properly link together in chronological order. As such, Juul’s
point remains valid—when disrupting chronology, designers need guard against the
dangers of interactivity.
Much more frequently, achronology occurs via an unearthing of the past, as the
player moves forward in chronological time but discovers exposition on past events,
allowing a previous chain of events to be discovered. J.C. Herz sees such game
structures as descendents of the murder mystery, describing them as “a narrative Easter
egg hunt using strategically placed characters or puzzles as a way of revealing a
predetermined back story,” such that the game ultimately offers two superimposed
stories:
One is the sequence of events that happened in the past, which you can’t
change but is a very good story. The other is the sequence of events that
happens in the present (e.g., you are wandering around trying to solve puzzles),
which is a lousy story but is highly interactive. As you hurdle obstacles in the
present, you can see a larger picture of the past.
218

217
Juul, Half-Real, 147-8.
218
J.C. Herz, Joystick Nation (Boston: Little, Brown and Company: 1997), 150.

229

In such an instance, the diegetic time of the player’s actions still occurs in chronological
order, yet previous diegetic events are revealed via other expository means. Herz offers
one of the more notorious examples of such a game with Myst (most likely the reason for
his claim of a “lousy story,” as Myst’s player occupies a world devoid of any character
but themself), though we see this strategy at work in countless games—Mass Effect and
Bioshock are among the more recent titles to heavily use the strategy. Arriving in
Bioshock’s dystopian underwater city of Rapture, the player must overthrow the
villainous forces that have led to its decay in order to survive, yet while going about that
agenda, the player uncovers the events which led the once-idyllic city to its present
state. As such, the player’s actions still occur in strictly linear chronology, while relevant
diegetic events are revealed in a more complex order.

Duration:
Discrepancies between diegetic time and the time of playing are much more
frequent in terms of duration, and these take a variety of forms.
Narrative ellipses are probably the most recognizable technique descending from
previous narrative forms, as irrelevant passages of diegetic time are “skipped over” in
much the same way that literary or cinematic narratives assemble only those episodes
relevant to the plot. We see such lapses most commonly between levels, and in this
manner, level structure frequently mimics the scene structure common to theatrical and
cinematic narrative—individual levels proceed in continuous “real time,” while the time
between these episodes are readily lapsed. It’s not insignificant that such lapses are
frequently cued by cut-scenes, putting the player in a cinematic mindset that readily
justifies such temporal gaps.

230
More unique is videogames’ frequent employment of the pause—stoppages of
diegetic time in which the fictional world comes to a complete standstill. Pauses may
occur under various circumstances—some are functional necessities, such as the load
screens which delay diegetic time while moving to a new level, while others are
stoppages of convenience, as when pausing the game to use the restroom or saving
one’s game in order to continue at a later time. Yet stoppages of diegetic time are not
only employed for simply putting the game “on hold,” as many games allow player
activity—allow the game to continue—while the diegetic world is stopped. Oblivion
features a copious number of extra-diegetic menus through which the player may review
information (character stats, map, mission objectives, etc.), manage their inventory,
equip and unequip items or spells, and so on, and this activity occurs while the diegetic
world is paused. At times, this may even create an ambiguous relationship between the
player’s actions and the diegetic world, as when the player of Resident Evil mixes items
or reloads their weapon—seemingly diegetic actions—while diegetic time remains
stopped. And in games like The Sims and Freedom Force, pausing time becomes a
primary part of gameplay, as players can pause the diegetic world to assign character
actions and perform other tasks before allowing time to resume and seeing the results of
those actions.
Finally, games may also effect duration by manipulating speed, allowing diegetic
time to move slower or faster than play time. While games like Max Payne and True
Crime: Streets of L.A. allow the player to temporarily slow time, replicating the slow-
motion effect of action cinema in order to afford a greater accuracy of movement and
gunfire, games like The Sims and SimCity conversely allow a speeding up of time,
allowing the diegetic world to pass by in rapid order. In more abstract games, we see
manipulations of speed not necessarily as discrepancies between diegesis and

231
discourse, but rather as part of the challenge of the game. As a player proceeds through
a game of Tetris, the falling blocks continue to move faster and faster, affording much of
the game’s challenge, making manipulations of speed integral to the game’s sense of
progression.
Though not necessarily a manipulation of duration per se, I think it worthwhile to
mention one further occurrence, one that we’ll call an interlude. I take the term from
Mark Wolf, who discusses interludes as “moments in which the game’s interactive
potential is briefly suspended,” a situation that merits its own address.
219
These include
such instances as cut-scenes, load screens, and other such interstitials, instances that
may have varying relationships to diegetic time (cut-scenes continue in “real time,” while
load-screens are typically pauses), yet all equally afford a respite from activity, a
“breather” so to speak from the action at hand. This is not so much a discrepancy
between discursive and diegetic time, but rather between discursive and play time—the
discourse continues in one manner or another, while gameplay is put on hold. These
include instances of extra-diegetic narration, moments of exposition that occur
independent of diegetic time. When Mortal Kombat provides the backstories of its
kombatants, or the player of Mass Effect reads through the game’s menu-based
encyclopedia, these occur outside of gameplay proper, serving as a respite from the
rigors of action and occurring separate from diegetic time altogether.

Frequency:
Genette describes narrative frequency as the relation between singulatives and
iteratives—that is, singular events and repetitions. While our usual assumption of
narration is to show an event once that happens once, narration over the years has

219
Wolf, The Medium of the Video Game, 83.

232
found numerous ways to manipulate this relation (flashbacks, alternating perspectives,
etc.). However, in videogames, one manipulation easily stands out from the rest:
replaying.
When we replay—whether an entire game or a particular moment—we “go back”
and revisit diegetic events a second time (or third, or fourth…). When the avatar dies
and we return to a previously saved state, we frequently think of this as a negation of the
last playing—best evinced in Prince of Persia: Sands of Time when the narrating prince
tells the player “no, no, no—that’s not how it happened” before returning to the last
saved checkpoint. Yet we cannot overlook the narrating repetition involved here, as
such an occurrence is a clear re-narration of events. Rather than simply revisiting the
same trajectory of events (like rewinding a movie), the game offers a revisiting of a
particular diegetic state—the diegesis returns to a previous state, allowing the player to
choose a different course of events through the same time and space. Sands of Time
plays off this dynamic by incorporating it into the logic of the game itself—using the
“Sands of Time,” the player is able to turn back brief spans of time to undo a mistake,
making replay and the manipulation of diegetic time a part of the game’s internal logic.
Yet in truth, this is exactly what occurs anytime that we restart a game or load a
previously saved game-state.
In match-based games—sports games, racing games, fighting games, etc.—we
can simply conceive of a second play-through as another diegetic instance, a separate
match from the last playing. Yet in deliberately narrative games like Halo and Tomb
Raider, we clearly revisit the same story. Much like the replay after dying, this a return
to a diegetic state—the initial moment of the story—allowing a new progression through
the same span of diegetic time and space. Events may occur differently the second time
through—precisely because we remember what happened the first time, making the

233
repetition directly influential on play—yet the discourse is still returning to the same
diegetic time.
This manner of diegetic repetition is fundamental to gameplay—it is why I
referred to gameplay as an “iterative” experience in chapter one. Each playing is a
repeating of the “general game,” the generic trajectory of events that we abstractly
conceive as the game itself (which, as we noted, doesn’t actually exist anywhere). In
this sense, every playing is a repetition, a return to an initial spatiotemporal instance that
always promises the same potential, yet never unfolds in quite the same manner. It is
by their very repetition that we expect to learn a game mechanic, to improve our skill and
strategy, thus repetition is a vital structural component for videogames. Abstract and
puzzle-based games often employ repeated activities as an inherent part of the game’s
structure—Tetris, Breakout, Pac-Man, Space Invaders and countless other titles loop the
same activity over and over again, often with minor variations to increase difficulty
(increased speed, more complex configurations, etc.). Yet even when such repetition is
not built into the game’s structure, their design may still rely on the assumption of
repetition. Many games are near-impossible to win the first time through without
stumbling along the way, and their design assumes that players will need to replay
particular segments in order to finally master the game—many of the Super Mario
games require the player to memorize patterns of movement to defeat various levels and
bosses, all-but-necessitating the player to replay several times before succeeding. And
while many of the lengthy, narratively-driven games of today are hearty enough to make
a single play-through worth the price of the game, the earliest of videogames were
frequently designed with an assumption that players would play the game over and over
again to achieve mastery (especially true in arcade titles, whose success relied on
players returning to pump more quarters into their slots). Mark Wolf refers to such

234
repetitions as “looped time” in that players return to the same spatiotemporal context
multiple times over.
Repetitions, cycled images, consistent and repeating behaviors, revisited
narrative branches, and the replayability of many of the games themselves
create a sense of expectation, anticipation, and familiarity for the player. They
encourage the player to find underlying patterns which allow him or her to take
control of the situations encountered, and this assured orderliness may well be
an important factor in the allure that videogames have for many people.
220

Whether through replaying or the repetitive structure of the game itself, looped time is a
vital component to the trial and error process common to play, and equally requisite for
developing skill and learning the logic of the gameworld.

Abstraction:
Manipulations of order, duration, and frequency are common to any narrative
medium, and in their overriding concern for plot, traditional narratives focus much of their
energy on how to arrange the telling of events. Much more unique is the videogame
medium’s ability to abstract time, a growing commonality in games of all sorts.
Temporal abstraction occurs when diegetic time behaves in manners that we
cannot reconcile with a consistent and unified temporality, evincing some contradictory
sense of time. Frequently, these instances require diegetic time to be understood within
multiple, divergent timestreams, defying any reasonable logic of temporality.
We see this in Oblivion in the game’s day/night cycle, which occurs at roughly
1/60 of normal time, such that one minute of play time sees an hour pass in diegetic
time. We cannot, however, simply call this a manipulation of duration (a “speeding up”
of diegetic time), as diegetic events still occur in real time. Neither dialogue, nor combat,
nor walking speed are sped up—the time it takes a character to speak a sentence or
swing a sword is neither shorter nor longer than we’d expect it to take—yet days and

220
Ibid, 82.

235
nights pass in markedly faster time than real time would allow. Recalling our semiotic
terms from chapter one, we’d best describe this as an iconic representation of time—
various attributes of temporal behavior are isolated and depicted in ways that
conceptually evoke their real-world referents (actions, days, months), yet the
representation of time itself is significantly abstracted from the real thing. This mirrors
the manner by which Ghosts N’ Goblins’ iconic protagonist conceptually evokes a knight,
though his image is clearly a reductive, cartoon version of an actual knight. As such, we
might call Oblivion’s day/night cycle a cartooning of time—it clearly conveys a sense of
time, and has a non-arbitrary resemblance to temporality as we understand it, though no
one would mistake the representation for temporal reality.
This is not the only manner by which videogames abstract time, mind you. In
SimCity 4, the player can speed up, slow down, and even stop the month and year
calendar in the game’s heads-up display, allowing the player to stop time while they
manipulate the terrain, construct new buildings, or re-zone districts, then watch years
pass by in minutes to see the results of their actions. Most striking, however, is that
regardless of whether time is zooming by in fast-forward or at a complete standstill, the
game’s day/night cycle remains consistent. Regardless of the pace of the calendar, the
player can watch their city’s citizens going about their daily lives, see the sun rise and
fall, completely independent of the calendar. In other words: the calendar and the clock
exist in entirely different timestreams, such that pausing one has no effect on the other.
Much like Oblivion, we can understand individual temporal concepts that the game
represents, yet the representations are so abstracted that we’d never confuse them for
the real thing.
Evincing both of these trends, Metal Gear Solid suggests a continuous stream of
real-time, yet numerous discrepancies suggest otherwise. Pivotal events are repeatedly

236
put on-hold until Solid Snake’s arrival, and regardless of whether the player completes
the game in five hours or fifty, they’ll arrive at the climactic ending “just in time,” such that
while individual actions occur in obvious real time, the larger temporal frame of the
game’s overall story remains much more ambiguous. Further, Snake holds lengthy
conversations over the Codec radio while the world around him seems to be put on hold.
Snake frequently enters radio conversations in the midst of a battle, yet his enemies
never attack while he’s speaking and are always waiting in the same spot once the
Codec screen is closed—despite some of those conversations lasting ten or twenty
minutes. So while time passes on the radio, time stands still in the rest of the diegesis.
Civilization and similar turn-based strategy games abstract time in yet another
manner. The Civilization player is given as much time as they wish to move their units
before finally choosing the “end turn” option to advance the calendar by one year, and as
such, the calendar is paused similar to SimCity’s. Yet more unusual is that in
Civilization, no two actions ever occur simultaneously. While the calendar remains
paused, the player is able to move each of their units, watching the results of each
action in sequence—a rifleman attacking a city, a scout moving out into uncharted
wilderness—all while every other civilization waits patiently. Once the player finishes
their turn, each ensuing civilization performs their own maneuvers in sequence, and
once all turns have been taken, one year passes on the calendar before gameplay
returns to the player. Time is essentially structured so that: one action occurs, one
action occurs, one action occurs, one action occurs, and then time passes by. Once
again, we cannot make logical sense of diegetic time’s behavior, though we can make
sense of temporal representation in a more abstracted sense—each action makes sense
on its own, the passing of a year makes sense on its own, the game’s movement

237
through historical eras makes sense on its own, yet combining all of these becomes
highly illogical.

Tense:
Seymour Chatman observes:
[T]here are perforce two NOWs [in narration], that of the discourse, the moment
occupied by the narrator in the present tense (“I’m going to tell you the following
story”), and that of the story, the moment that the action began to transpire,
usually in the preterit.
221

It is specifically because narrative content occurs in the past-tense that its discourse is
able to manipulate its temporal arrangement, and here is where we hit a theoretical
conundrum: despite the vast fluctuations that we’ve seen between the time of play and
diegetic time, in a videogame, both of these time frames are suggested to be occurring
“now.” Much like traditional narration’s time of discourse, a videogame’s time of play is
obviously in the present-tense—the player is playing “right now”; yet because of the
interactive nature of the narrated content, diegetic time must also be occurring “right
now.”
Though games may take place in past or future eras (medieval England, the
WWII Pacific theater, or a distant planet in the distant future), diegetic time in
videogames is always narrated in the present-tense, a drastic shift from traditional
narration—and equally, a temporal paradox. Manipulations of time in traditional
narration are readily explained by the past-tense of the content; since we assume all the
narrated events to have already occurred, the narrator can recount them in any manner
they choose. We can always assume a logical diegetic chronology that can be
assembled retrospectively, as the discourse and story occur in different tenses—the
telling in the present, but that which is told is in the past. Among the great curiosities of

221
Chatman, Story & Discourse, 63.

238
videogame time is that the game narrates a present-tense scenario, despite a wide
variety of temporal manipulations. Curiously, even when we lapse time (when diegetic
time is longer than discourse time), we still remain in the present.
Chapter four will help explain this situation as we discuss narrative framing and
the multiple contexts involved in the narrative act; for now, we can observe that while
both discourse and diegesis occur “now,” they entail entirely different renditions of “now.”
The story occurs “now” and gameplay occurs “now,” but they’re not the same “now.”
Cut-scenes, on the other hand—traditional cinematic narration that generally
occurs in the third-person—carry with them a suggestion of past-tense, just as the
cinematic predecessors from which they take their structure. As such, movement
between gameplay and cut-scenes provides an alternation of tense. If we were to
verbalize the opening moments of Gears of War, as the player (in the role of Marcus
Fenix) exits a prison compound and boards a helicopter, triggering an ensuing cut-
scene, we might narrate the situation as such:
You exit the prison compound. Locust soldiers attack and you shoot down the
last of their ranks. You charge across the prison courtyard to an awaiting
helicopter. You climb aboard.

[cut-scene] Fenix climbed aboard the helicopter, and just as the chopper lifted off
the courtyard, a massive Locust beast burst through the crumbling pavement. As
the helicopter lifted away from the prison, its young gunner looked to Fenix. “Are
you the Marcus Fenix?” he asked.

Just as cinematic narration carries the implicit suggestion of a previous happening—that
the events documented on-screen are not happening “now,” but rather happened at
some point in the past—so too cut-scenes suggest such a past-tense. So while the
modulation between gameplay and cut-scenes shuffles the player between participant
and observer, it equally shuffles between simulation and documentation, one suggesting
events happening right now, the other suggesting events that have happened in the
past.

239
Logic
Logic also has a dual deployment, though it proceeds somewhat differently than
space and time. In any instance that evinces some manner of logic in a videogame text,
we can interpret that logic in at least two manners—as the product of a system of rules,
or as the internal reality of a diegetic world. And as videogame players, we are always
negotiating these two logics.
When we play Tomb Raider: Legend or Prince of Persia: Sands of Time, the
diegetic world is comprised of countless details; if we had to focus on all these details in
order to figure out how to solve each puzzle, we’d never finish. Yet by focusing on their
“game-ness”—considering the world as a rule-based arena—we can narrow our options.
Resultantly, the world becomes a series of climbable ledges and a scant few usable
objects, and we can better determine our options. It’s not insignificant that in both
games, climbable ledges, moveable blocks, grapple-points, and other usable objects
tend to stand out from the background, allowing the player to distinguish those spatial
elements which are part of the game arena from the diegetic backdrop—whether
intentional or not, these distinctions aid the player’s ability to navigate the game, to fall
back on rule-based logic over the vast possibilities of the diegetic fiction. This becomes
especially relevant as various obstacles and affordances are repeated throughout the
game, allowing the player to remember their previous encounter with said objects
(demonstrating the important of the repetition we addressed earlier). Anytime we see a
long red banner descending a wall in Prince of Persia, we immediately recognize those
banners from our previous encounters, knowing that the prince can use them to slide
down great distances. Such games are adept at enforcing rule-based constraints while
simultaneously eliding them under the guise of a diegesis—while we can discern the
game’s rule-based nature, its logic is equally understandable as a diegetic reality.

240
We see this dual-logic in countless examples, even those much more abstracted
than Prince of Persia or Tomb Raider. The Civilization example which opened chapter
one evinces this in a way, as we can read Civilization as a game (with its own bizarre,
rule-based logic) or as abstractly representing a real world context (with a more rational
and understandable logic), and in this case, those two realities are strikingly divergent.
As we see blocks fall in Tetris, we are informed by our understanding of gravity and
spatial physics, so that even while we process the game by its rules, the logic of its
representation still effects our engagement with that space. Similarly, when we play
Pong, Breakout, or any other ball-and-paddle game, our familiarity with deflection
informs our understanding of where the ball will travel—when it hits a surface at an
angle, we expect to bounce back at a similar angle in the opposite direction; if the ball is
traveling through empty space, we expect it to continue in that direction until it hits a new
surface. There’s no reason that the ball couldn’t suddenly change direction of it own
volition (as a virtual object, it can behave in any manner that designers wish), yet based
on its representation, we expect certain behaviors—a certain logic—until shown
otherwise. Further, our understanding of ball-and-paddle games are always preceded
by our knowledge of tennis, ping-pong, and similar real-world games, providing an
immediate frame of reference as to what will occur.
Although the game’s outward representation helps rationalize its logic, then, this
does not preclude our comprehension of the game as a rule-based system. Though we
grasp the diegetic logic behind the Prince of Persia sliding down curtains, we also
recognize said curtains as a rule-based component of the gameworld that allows for one
specific action—sliding down them. One of the vital balancing acts that designers must
often perform in such instances is to present objects in a manner that seems believably

241
integrated into the diegetic world, while equally recognizable as the rule-based object.
As such, we are always encountering those objects via both logics at once.

Causality
David Bordwell observes that classical narration—and make no mistake,
videogame narration is generally quite classical—centers on one principle above all else:
causality.
In classical fabula construction causality is the prime unifying principle.
Analogies between characters, settings, and situations are certainly present, but
at the denotative level any parallelism is subordinated to the movement of cause
and effect.
222

Logic is largely a product of causality—the game-text arranges information in such a way
that the player can make discernable sense of events and existents, particularly in
regard to causal motivation. This may come about by straightforward exposition (directly
explaining how to do something, why something occurs), though it frequently happens in
more implicit manners, such as a proximity of events (I performed action A, and event B
immediately occurred, so I deduce that A caused B). Our concern, however, is not only
for what has happened and why, but equally what could happen and how one would
achieve that end (knowing that A causes B, I can assume that performing action A in the
future will again result in event B). As Bordwell describes, audiences continually make
and test hypotheses, trying to make logical sense of whatever they perceive in the
discourse.
223
In a videogame, however, this is not only a process for understanding
what has happened, but also for deducing how to accomplish tasks—strategizing and
puzzle-solving.

222
Bordwell, 157.
223
Ibid, 37.

242
Salen and Zimmerman identify two characteristics necessary to the internal logic
of a game: that it is discernable and integrated.
224
Discernability entails the player’s
ability to perceive a causal logic between action and outcome. When Mario consumes a
mega-mushroom and suddenly doubles in size, we discern the former to have caused
the latter. But perhaps a better example is in indicators of damage during boss battles.
When battling Super Metroid’s Kraid, the player will have to barrage the monsterous
creature with a long series of attacks to defeat him. Further complicating the battle,
however, is that Kraid is only vulnerable in particular parts of his body. Yet how is the
player to know this? No one shot will cripple the beast, so how can the player discern
whether their attacks are having any effect? To resolve this, designers added an
aesthetic cue to facilitate this causal logic—the creature flickers red anytime that it takes
damage. As such, the player is able to discern which actions are damaging the beast,
and which have no effect at all. This, however, is merely localized causality, providing
an immediate significance to individual events. Equally important is that such meaning
is integrated into the larger context of the game, allowing an understanding of how such
localized actions play into the game as a whole. In Super Metroid, this is partly achieved
by the game’s structural patterns: learning a boss’s weaknesses and attack patterns will
allow the player to defeat it; defeating the boss will provide access to a new item; that
new item will provide access to new areas of the game; and so on.
Piecing together the logic of the gameworld is one of the fundamental processes
of gameplay—as players, we speculate on possibilities and probabilities, certainties and
unachievables. And we can identify numerous factors for constructing causal logic in a
videogame: representational logic, the logic of our own world and its representational
relationship to the world of the game; conventional logic, our familiarity with conventions

224
Salen and Zimmerman, Rules of Play, 34.

243
from previous games, previous stories, and other cultural instances; tested logic, that
which we surmise from trial and error, repeating actions to discern their effects; temporal
relationship, the proximity of instances within the game’s trajectory of events; and
exposition—when all else fails, simply explaining the system to the player. But to be
reductive, we can broadly say that we build our sense of game logic through the
combination of rules and representation—surmising the systemic logic of the game and
correlating it with the diegetic world that it conveys.

Verisimilitude
Seymour Chatman observes that fiction uses a different rhetoric than that of
history, as the latter “suades us to believe that its stories are factually true,” and thus
demands a different degree of logic. Fiction makes no such claim. Rather than fact or
truth, fiction trades in verisimilitude—a manner of plausible logic that Chatman describes
as an “appeal to the probable, rather than the actual.”
225
Etymologically, verisimilitude
descends from the Latin verus and similis, conveying not “truth,” but rather the quality of
being “like truth.” When dealing with fiction, that which we accept as “valid” need not
correspond to factuality in any sense, but merely to a logic that we find acceptable given
the world in which the fiction takes place (whether a world that replicates historical
eighteenth-century London or a mystical fantasy realm lacking realistic demands on
science and rationality).
An unconscious process of “filling in” details to create coherent sense of a
fictional reality, verisimilitude is greatly reliant on the formal transparency I addressed
earlier, whereby the game’s rules and representations become naturalized and
resultantly efface their very existence, eliding the text’s conventional nature.

225
Chatman, Coming to Terms, 151.

244
Audiences come to recognize and interpret conventions by “naturalizing” them…
To naturalize a narrative convention means not only to understand it, but to
“forget” its conventional character, to absorb it into the reading-out process, to
incorporate it into one’s interpretive net, giving it no more thought than to the
manifestational medium, say the English language or the frame of the
proscenium arch.
226

This concept has traditionally been applied to representational realities—those depicting
a world that we compare to our own—yet we see the process occurring in abstract
puzzle games just as readily as immersively mimetic ones. When playing Bookworm,
Bejeweled, Hexic, or Puzzle Quest—all games that feature geometric shapes packed
into the field of play—players never question why game tokens continue to descend from
above as others are eliminated. Taking the cues provided—that each token seems to
have dimension, that further tokens descend downward when the space below them
become unoccupied—players can “fill in” causal logic based on fundamentals of spatial
physics and gravity, making the representation seem entirely sensible (this is evinced by
the frequency with which descending tokens are described as “falling” from above). This
occurs from the same process of naturalization to which Chatman refers, even if in a
somewhat different context.
Yet we do not only naturalize logic based on personal experience of the world.
Chatman further notes that what constitutes “reality” or appears “natural” is largely a
cultural phenomenon, and such norms change from culture to culture and era to era.
227

Branigan similarly notes that logic “is often based on what seems familiar and natural,”
and while this may be based on personal experience, it is quite often instead naturalized

226
Chatman, Story & Discourse, 49.
227
As example, consider the fantastic manipulations of physics common to Hong Kong martial
arts films, conventions that were readily accepted in Asia yet seemed ridiculous by American
standards—until American audiences had seen enough to absorb those conventions into their
own frame of reference

245
by cultural convention.
228
Our experience of other texts, other games, societal norms,
and learned beliefs all influence our cognitive reception of a videogame’s reality.
[T]he norm for verisimilitude is established by previous texts—not only actual
discourse, but the “texts” of appropriate behavior in the society at large.
Verisimilitude is an “effect of corpus” or of “intertextuality” (hence
intersubjectivity). It is a form of explication, pointing from effect to cause, and
even reducible to a maxim. Further, because maxims are public, that is, “tend to
go without saying,” they may be implicit or backgrounded.
229

As a result, overt explanation only becomes necessary for acts that would seem to
violate the norm by contemporary standards of reality or convention. More common,
Chatman argues, is “unmotivated” verisimilitude, where overt details of causal logic are
unnecessary as the audience is able to deduce some reasonable logic to connect events
(at least, reasonable within the given context).
Verisimilitude is greatly aided by what some will term the “reality” or “truth effects”
of fiction: the arrangement and deployment of seemingly unimportant details as an
assurance of authenticity. In a sense, this is what we frequently think of as a fiction’s
“atmosphere,” the nonessential background details that help further the illusion of
verisimilitude by “setting the stage.” We do not need the animated backdrop to Super
Mario World, nor the river and hillside beyond the border of Oblivion’s gamespace, as
those unexplorable spaces serve no purpose to the functionality of gameplay, yet such
details facilitate our willing surrender to the game’s internal reality, and as such, still
serve a vital purpose.

Events
Despite our diegetic focus on the world of the game, we do need pay some
attention to the events of a videogame. Narration, after all, entails not only the projection

228
Branigan, 28.
229
Chatman, Story & Discourse, 50-1.

246
of a spatiotemporal environment, but a trajectory of events therein. But how does such
action occur within a videogame?
Traditionally, critics and designers have answered this with a binary approach,
distinguishing two manners of game—one using a linear structure reminiscent of
traditional narration, and the other using a generative design more in line with traditional
gameplay. Jesper Juul distinguishes these as games of progression and games of
emergence:
Progression is the historically newer structure that entered the computer game
through the adventure genre…[and entails the player performing] a predefined
sequence of events. If the player does not perform the right actions, the game is
over. It is characteristic of progression games that there are more ways to fail
than to succeed. The progression structure yields strong control to the game
designer: Since the designer controls the sequence of events, this is also where
we find games with cinematic or storytelling ambitions.

Emergence is the primordial game structure where a game is specified as a
small number of rules that combine and yield a large game tree, that is, a large
number of game variations that the players deal with by designing strategies.
Emergence is found in card and board games, most action, and all strategy
games. Almost all multiplayer games are games of emergence.
230

These categories merely offer two extremes, of course. Using single-player action
games and massively multiplayer online games as examples, Juul acknowledges that
many videogames actually combine components of both. Yet these categories make a
useful design distinction, and help identify the paths by which conventional narration and
gameplay have entered the medium.
Not surprisingly, given Juul’s account, the former category is frequently
addressed using the standard terms of narratology, while the latter is instead discussed
as rule-based systems (and equally, to discuss videogame narratives, we typically use
examples from the first category, while to discuss gameplay, games from the second).
To some degree, then, these categories divide those games that are most narrative-like
and most game-like, providing a design-centric distinction in much the same vein as

230
Juul, 72-4.

247
Salen and Zimmerman’s approach, stressing the underlying mechanics of the
videogame’s structure.
To address the nature of a game’s trajectory of events in a more player-centric
manner, however, I find Henry Jenkins’ terms to be more useful. Moving away from the
conventional discursive model of “storytelling” to foreground the spatiality of game
design, Jenkins’ suggests a new model of “narrative architecture” or “environmental
storytelling,” entailing an extensively different manner by which diegetic events
materialize and are understood. Rather than a binary approach, Jenkins addresses four
different manners by which narrative meaning might manifest from a videogame.
Environmental storytelling creates the preconditions for an immersive narrative
experience in at least one of four ways: spatial stories can evoke pre-existing
narrative associations; they can provide a staging ground where narrative events
are enacted; they may embed narrative information within their mise-en-scene; or
they provide resources for emergent narratives.
231

Jenkins acknowledges that this list is by no means exhaustive, yet it does cover a
significant range of videogame narration, and as such, it suits our purpose for this brief
discussion. Further, we can see each of these potentials at work in this chapter’s
opening example of Oblivion.

Evocative Spaces
First on our list are those spaces that evoke pre-existing narrative associations,
whether specific cultural texts (The Empire Strikes Back, American McGee’s Alice) or
general myths and genres (Space Invaders, Grand Theft Auto III). These allow players
to “enter physically into spaces they have visited many times before in their fantasies,”
and “do not so much tell self-contained stories as draw upon our previous narrative

231
Jenkins, “Narrative Architecture,” 123.

248
competancies.”
232
Relying on our cognitive ability to build upon previous knowledge and
construct new trajectories, such spaces expect the player to do much of the narrative
legwork. Evocative spaces offer a narrative context—the broad outline of narration—
and allow the player to fill in the details.
Traipsing through Cyrodil calls to mind the Tolkienesque brand of medieval
fantasy common to a wide range of texts spanning literature, cinema, table-top gaming,
and other videogames. Filled with goblins and trolls, swords and sorcery, adventure and
exploring, Oblivion evokes a massive amount of narrative data well outside of what
actually happens in the game. And against this backdrop, anything that happens in the
game—simply by virtue of its contextualization—will take on an untold collection of
narrative meanings.
To a degree, evocative spaces are all products of genre—even when their
associations arise from one specific text or franchise. Discussing the genre film, Rick
Altman makes a seminal distinction between the semantic and syntactic elements of
genre, dividing the immediately recognizable semantic building blocks of genre (the
iconography, characters, locations, musical motifs, and other readily identifiable tropes
that are associated with a particular corpus of texts) from the deeper structures of
meaning that emerge from bringing those elements together (plot, theme, and so on).
233

Evocative spaces are those built on semantic elements, allowing players to form their
own syntactic connections. Such worlds are built on recognizable characters, settings,
and other iconography that carry a vast range of pre-associations, whether textually
specific icons (Luke Skywalker, Superman, Middle Earth) or literally generic ones
(robots, super-powers, goblins), and through both the activity of gameplay and the

232
Ibid.
233
Rick Altman, “A Semantic/Syntactic Approach to Genre,” in Film Genre Reader, ed. Barry
Keith Grant (Austin, TX: University of Texas Press, 1986), 26-40.

249
cognitive processes of association and verisimilitude, players assemble those elements
into a narratively meaningful trajectory of events.
As such, everything I do in Cyrodil is informed by the associations carried in its
iconic world. Whether engaging a deliberately designed sequence of events or simply
stumbling into various encounters as I wander across the countryside, much of the
meaning I draw from Oblivion’s “story” is built in my own mind, evoked by the pre-
associations fixed into its evocative landscape.

Enacted events
Rather than a space evoking pre-associations, we can also think of the
gameworld as a stage upon which narrative events can be enacted. In this sense,
everything that the player does feeds into the “story” of the game, as the player operates
as the actor in their own personal storytelling environment. In one regard, enacted
events seem the most obvious locus for narrative meaning in videogames—as a
medium of interactivity, the player’s very interaction seems our most obvious concern.
Yet enacted events are equally the most opposed to our traditional conception of
narration—these events are performative rather than expository, resulting largely from
“what is done” by the audience rather than “what is told” by the text.
Much of Oblivion emerges from my own choices as a player, the narrative
decisions that I make within the game’s pre-built fantasy context. Of course, we cannot
overlook that last phrase—we’ll deal more with the game’s pre-determined components
in a moment—yet at every turn of the game, my own choices as a player help shape the
unfolding of events. I may choose to join the Fighter’s Guild and become a hearty
warrior, or perhaps instead the Mages’ Guild, practicing magic and sorcery. I may play a
noble hero who is fair and honest, helping those whose paths I cross, or I may turn to

250
murder and thievery, choosing only those actions that suit my needs. While Oblivion
does rely on predetermined trajectories of events, it is equally reliant on the choices I
make through my avatar—on the path that I choose through its narrative possibilities.
This is the root of the participatory storytelling of which Gary Alan Fine speaks.
“Fantasy gamers…construct a world imbued with meaning,” Fine notes, as theirs is a
world “that is not inherently meaningful but is made meaningful by the significance given
to it by its participants.”
234
It is for this reason that Fine concludes that while the
dungeon master is, in theory, the author of the game, “he can only create a meaningful
fantasy world with the support of the players,” as the unfolding of his scenario results
from the players’ activity.
235
Players enact the game both through deliberate,
consciously chosen paths of action as well as spur-of-the-moment choices and
reactions, so while the videogame may provide the narrative framework, the
improvisational enacting of the game relies on players’ participation in the storytelling
process. Observing that few fantasy role-players actually “role-play,” Fine even argues
that storytelling offers a more appropriate analogy than performance. “Rather than
conceiving of gaming as improvisational acting, a better metaphor might be storytelling—
with each storyteller having authority over one character—producing a collective
fantasy.”
236
In a similar context, the videogame and players enter into a dialogue of
interactive storytelling, with the players donning the role of one or more characters and
the game performing the roles of every non-player character (NPC), along with the very
world itself.
Jenkins addresses this as episodic storytelling of a sort, wherein each episode
may become compelling in and of itself without greatly contributing to the overall “plot.”

234
Fine, 231.
235
Ibid, 80.
236
Ibid, 213-4.

251
In the classical mold, such episodic storytelling has often been dismissed, as the
interchangeability of episodes—their capacity to be rearranged without consequence to
the plot’s effect—evinces a looseness of plot that can be readily seen as sloppy or
unstructured. Yet Jenkins assures us:
Spatial stories are not badly constructed stories; rather, they are stories that
respond to alternative aesthetic principles, privileging spatial exploration over plot
development. Spatial stories are held together by broadly defined goals and
conflicts and pushed forward by the character’s movement across the map.
237

This account is especially appropriate to address Oblivion, comprised of countless—and
I truly mean countless—narrative episodes spread across the landscape. The iterative
experience, then—the specific trajectory of gameplay—results from the player’s
enactment of those episodes, assembling their own unique path through the game’s
near-infinite possibilities.

Embedded events
Embedded events, on the other hand, account for the pre-determination of those
episodes. Fixed dialogue, pre-developed characters, triggered events, and other
instances of designer-controlled happenings all fall under the heading of embedded
events. This is where videogame design borrows its most deliberate use of conventional
storytelling, as fixed occurrences are planted in the gameworld, manifesting by virtue of
some pre-determined trigger.
Jenkins describes two manners of such events. In the first, the game is a
“memory palace,” where past events are inscribed into the landscape, waiting to be
discovered by the player. This is the manner of storytelling seen in Myst and Bioshock,
where a past-tense story can be assembled as the player uncovers more and more of
the environment’s secrets. In these cases, since the pre-determined chain of events has

237
Jenkins, “Narrative Architecture,” 124.

252
already occurred, the player may uncover those details in any order while still
maintaining a coherent story, and as such, the player’s navigation through such a space
may remain relatively unstructured. Yet Jenkins also points to a separate mode of
embedded narration, one in which pre-authored, present-tense events are triggered
within the landscape. This is the mode we frequently associate with cut-scenes, when
arriving at a particular location—as when we enter a room or complete a level—triggers
a fixed, cinematic passage of narrative exposition. Games like Halo and Metal Gear
Solid, built on a long series of such pre-rendered sequences, are hugely reliant on such
exposition. Yet such heavy-handed tropes of conventional narration are by no means
the only such instances, and Jenkins suggests a particularly strong example of
integrated narration in Half-Life. Here, the player retains autonomous control of their
avatar throughout gameplay, always maintaining a sense of first-hand enactment,
however pre-authored occurrences remain embedded in the landscape nonetheless,
triggered by the player’s arrival at pre-determined locations. When the player first
emerges from the Black Mesa underground research facility, a squadron of marines will
automatically deploy from an overhead helicopter, and similarly, encountering Barney
the security guard will trigger that character to relate a fixed monologue of exposition.
This manner of embedded narrative requires a much more structured sequence of
events, of course—as a rearranging of episodes would likely result in a nonsensical
story—and as such, Half-Life, Halo, and Metal Gear Solid all heavily constrain the
player’s route through their worlds. Resultantly, the gameworld becomes one long
gauntlet with exposition embedded along the way. This isn’t the only means of
narrational constraint, mind you—Grand Theft Auto III offers a very different landscape
in its open world design, constraining the player instead via mission trees, wherein event
B cannot be accessed until event A has been completed, similarly guaranteeing that

253
certain events occur in their deliberate order. In either case, though, designers dictate
the course of events not through the temporal arrangement of conventional narration,
but rather a spatial embedding of events into the landscape.
We see both of these modes at work in Oblivion. Traversing Cyrodil, a plentiful
amount of back-story is to be found. In following the game’s primary mission tree—
foiling the plot of Mankar Camoran—players uncover an extensive history involving the
history of the kingdom, the emperor’s bloodline, and the rise of the Mythic Dawn cult. If
players join the Fighter’s Guild, they will gradually learn about a rival band of
mercenaries, The Blackwood Company, with plentiful information (both significant and
trivial) to be unearthed through both deliberate missions and more casual interactions.
Most of the game’s autonomous side-missions even entail some back-story to be
discovered. Equally, however, the game entails both manners of present-tense
narration. Like Grand Theft Auto, Oblivion employs a mission tree structure that
guarantees certain events occur before others. I cannot infiltrate the Mythic Dawn
shrine, for example, until after I escort the king’s heir to safety in the mountains of
Bruma. Equally, the quest trees for the Fighter’s Guild, Mage’s Guild, and Thieves’
Guild follow a deliberate path, as each successive quest cannot be accessed until the
previous one is completed. Yet many of these individual missions also employ spatial
constraints to route the player through embedded events, much like Half-Life. We saw
this in the game’s opening, a linear gauntlet through the tunnels beneath the prison that
allows a deliberate trajectory of pre-determined events. Both of these latter methods
entail a certain degree of “on-rails” narration, corralling the player through a specific
route and/or a specific trajectory of events, yet they equally afford an organic sense of
stumbling from event to event by using natural boundaries, a logical progression, and
other verisimilar justifications for constraining the player.

254

Emergent events
Lastly, gameworlds may provide the means for a spontaneously generated
trajectory of events via dynamic components that interact in complex and surprising
ways. “Emergent narratives are not prestructured or preprogrammed, taking shape
through the game play, yet they are not as unstructured, chaotic, and frustrating as life
itself.”
238
This is the mode of discourse that we often cite as a “sandbox” design,
affording a variety of “toys” to play with (objects, vehicles, characters) that not only
interact with the player, but with one another, in novel and unpredictable ways.
The Sims has become the standard-bearer for such design, wherein each
character’s behavior is algorithmically generated via numerous desires, needs,
personality traits, skills, and “memories,” compounded by cues from their environment
and their interactions with objects and other characters. In and of themselves, this wide
range of influences can already dictate novel behaviors within each Sim, yet that
behavior is further complicated by the interaction of numerous such characters—along
with the commands and interactions of the player. Further, the game is filled with
evocative narrative components—character appearance and wardrobe, objects and
furniture themed by an assortment of designs, and countless staples of romance, work,
play, domesticity, and interpersonal relationships. Not only do the characters behave in
emergent ways based on the interaction of autonomous rule-sets, but narrative meaning
emerges from the collision of these narrative components, as all the pre-associations
evoked by those signifiers color the game’s emergent events. As such, the game’s
resulting chain of events and discursive meanings become a spontaneous and
unstructured story that cannot be credited to any one authoring intelligence.

238
Jenkins, “Narrative Architecture,” 128.

255
Though not nearly as complicated, many of Oblivion’s characters are also driven
by algorithmically-generated behaviors, and the very appearance of many of the game’s
creatures are spontaneously generated (that is, exact creatures aren’t “planted” in
specific locations, but rather chosen based on a variety of factors). A trip through a
particular cave may entail a few wolves in one session, or a slew of undead in another.
And as the player navigates the gamespace, cueing events by a wide assortment of
means, the overall configuration of the game becomes an unpredictable series of
episodes.
Combined, Jenkins’ four modes of spatial narration provide us four sources from
which the details of story might manifest in videogames: evoked from outside, enacted
by the player, embedded by designers, or emerging from the computer. And with
Oblivion, we see a clear example in which all four work in conjunction, creating an
elaborately complex narrative experience that is neither predetermined nor chaotic, but
instead the interactive product of all of these influences.

In Experiencing Narrative Worlds: On the Psychological Activities of Reading,
Richard Gerrig identifies two metaphors that he finds most frequently used to
characterize a narrative experience: transportation and performance, metaphors that are
especially apt for addressing videogame worlds.
239
Gerrig cites Victor Nell’s assessment
of being “lost in a book” as example of the first metaphor, evincing the transportive effect
that audiences feel when engaging a narrative world separate from the here-and-now.
There are, of course, other metaphors that we might alternately use for this quality; when
addressing videogames, “immersion” is the more popular choice, though this gets at the

239
Richard Gerrig, Experiencing Narrative Worlds: On the Psychological Activities of Reading
(New Haven, CT: Yale University Press, 1993), 2.

256
same point (Gerrig even implies a similar dynamic by correlating this effect to the “depth”
of the narrative experience).
240
To address performance, Gerrig instead turns to
cognitive psychology, pointing to the same general ideas on which Bordwell’s
constructivist approach is rooted—and in this metaphor, Gerrig is indeed addressing a
similar phenomenon: the audience’s necessary activity of imagining (“constructing”) the
fictional world for themselves. As example, Gerrig similarly cites Frederic Bartlett’s
research on schema theory, which demonstrated that “readers’ memories for texts
represent a merging of textual information with pre-existing knowledge,” a process that
Bartlett termed “rationalization” but that we’ve instead discussed as verisimilitude—the
filling in of unstated details in order to make logical sense of the goings-on.
241
The
audience’s “performance,” then, is the imaginative construction of a diegesis based on
both narrative information and one’s own cognitive interpretation.
Both of these metaphors, however, become especially apt in videogame worlds.
The transportive effect is not only a psychological engagement, but a participation in the
fictional world, a simulated sense of presence in the diegetic context. The performative
quality is not only a cognitive act of imagination and interpretation, but a more literal
enacting of content, a taking-part in the storytelling itself. And both of these processes
occur through the avatar—the topic of our next chapter.

240
Ibid, 8.
241
Ibid, 18.

257

Chapter Three: AVATAR

The nuclear weapons disposal facility on Shadow Moses Island, in Alaska’s Fox
Archipelago, was attacked and captured by Next-Generation Special Forces
being led by members of FOX-HOUND. They’re demanding that the government
turn over the remains of Big Boss, and they say that if they’re demands are not
met within 24 hours, they’ll launch a nuclear weapon.

You’ll have two mission objectives. First, you’re to rescue DARPA Chief Donald
Anderson and the president of ArmsTech, Kenneth Baker. Both are being held
as hostages. Secondly, you’re to investigate whether or not the terrorists have
the ability to make a nuclear strike, and stop them if they do.

As these introductory words are spoken in voiceover, shadowy images of a
nuclear submarine’s command bridge fill the screen, along with close-ups of an
enigmatic figure preparing to be launched in a one-man submersible. The face of this
mysterious individual is cloaked by scuba gear, yet a superimposed title reveals his code
name: Solid Snake. As Snake’s torpedo-like vessel is released from the sub, the
voiceover continues, recounting an earlier exchange between Snake and his supervising
officer, Colonel Campbell. Campbell continues explaining the details of Snake’s
infiltration mission—the limited support he’ll receive, the obstacles he may face, the
enemies he may encounter. Snake exits the submersible, and as he swims toward the
light above the water’s surface, the image fades to white before revealing a title screen:
“Metal Gear Solid. Press start button.”

258
I press “start” on my PlayStation’s DualShock controller and select “new game”
from the scroll menu of options, then watch as the game’s introductory sequence
continues.
Solid Snake, covert Special Forces operative and hero of the previous three
Metal Gear titles, swims through the murky waters underneath the Shadow Moses
facility. A stylishly cinematic sequence reminiscent of numerous spy thrillers, the movie-
like atmosphere of this opening is reinforced by the voice-acting and production credits
superimposed over the image. Part James Bond thriller, part Tom Clancy intrigue, part
Japanese anime, and part militaristic action spectacular, everything about the sequence
drips of Hollywood realism and sheen. Snake emerges from the water on the outskirts
of the facility’s underground docks and begins a conversation with Campbell via his
Codec radio.
“This is Snake. Colonel, can you hear me?”
“Loud and clear. What’s the situation?”
As Snake speaks with the Colonel, the game takes considerable effort in
justifying the protagonist’s ability to converse by radio at any given moment without
nearby guards overhearing the verbal exchange. The Colonel explains: “The Codec’s
receiver directly stimulates the small bones of your ear. No one but you will be able to
hear it.” This conversation system is necessary for providing the player with relevant
exposition throughout the game (as here, where Campbell explains that Snake must
covertly reach the elevator at the far side of the docks), yet within the narrative context,
the game must rationalize why the sound of the conversation does not alert nearby
guards to Snake’s presence. Equally, many of the awkward plot devices necessitated
by the basic mechanics of gameplay are rationalized by such explanations. The “save
game” option—another functional necessity—is later elided when Campbell introduces

259
Snake to an “operator in charge of communication data processing” who is available to
record a log of Snake’s “mission data” whenever he contacts her by radio.
Considering these nods toward plausible realism and the suspension of disbelief,
the Colonel’s explanation to Snake of how to use the radio seems somewhat unusual: “If
you need to, contact me by Codec. The frequency is 180.85. If you want to use the
Codec, push the select button.”
Push the select button? Hmmph. That’s an odd thing to say to Snake in this
context. After all, the select button is on my Playstation controller, not on Snake’s
radio…
I disregard the oddity of this statement, and as the game continues, I finally take
control of Solid Snake. Snake is my avatar—the digital body that I inhabit in the game,
my representative in this narrative world—and with intuitive flicks of my controller’s
thumbstick and some well-timed button-pressing, I fluidly navigate the hero through the
underground docks of the Shadow Moses compound. Avoiding the guards on patrol in
this area, I dart between a few rows of large storage crates and sneak into a freight
elevator to reach the surface, whereabouts—after some more expository radio dialogue
from Campbell—I begin my efforts to infiltrate the stronghold. I stealthfully maneuver
toward the building, avoiding the beams of overhead searchlights and the automated
gaze of security cameras, yet as I dodge around the large helipad in the center of the
area, I spot a roving patrolman approaching, and with nowhere better to hide, I duck
inside the back of a cargo truck while the soldier passes by. To my pleasant surprise, I
find a SOCOM pistol inside the vehicle—quite a relief, having not been equipped with a
weapon. An audio beep indicates that Colonel Campbell is beckoning, so I press the
select button to converse via radio once again:

260
Good, you’ve got yourself a weapon. To use it, first hold down the R2 Button to
enter “Weapon Mode.” Then select the weapon you want with the Directional
Button. After you’ve selected the weapon you want, let go of the R2 button to
exit “Weapon Mode.” The weapon you selected should appear in your hand. To
use the weapon follow the directions displayed in the window.

Okay, hold down the R2 button, then… wait a minute! The R2 Button? Weapon Mode?
The directions displayed in the window? Snake would have no idea what any of these
things mean!
While this dialogue’s context suggests that Campbell is speaking to Solid Snake,
these instructions are addressed to me, the player, and as such, remain utterly
nonsensical in the diegetic situation. While the instructions are certainly helpful—I doubt
that I would have been able to figure out how to equip and use the weapon on my own—
they seem completely contrary to the illusion of reality and the suspension of disbelief
that the game has been trying so hard to provide. Yet such tutorial moments are hardly
unique in videogames. I could probably list a hundred games that employ similar
conventions, and resultantly, gamers tend to take such moments for granted, hardly
giving them a second thought.
242

Yet consider how an audience would react if such a moment occurred in a
different medium, such as the Hollywood movies to which Metal Gear Solid pays
homage. Imagine a similar exchange in the James Bond franchise: After some curious
turn of events, the film’s musical underscore suddenly swells, as Q contacts Bond by
radio. “Do you hear that music, 007? That’s Goldfinger’s theme. It means that
whatever’s happening right now has some connection to the villain. Oh, by the way, if
you’re feeling a bit hungry, there’s popcorn in the lobby.” These comments might be
informative for the audience, yet they would seem completely illogical to the character of

242
Some games do, however, acknowledge the absurdity of such dialogue in context—characters
in The Legend of Zelda: Link’s Awakening similarly explain gameplay elements like the save
function to the avatar of Link, but then conclude the exposition with such declarations as, “I have
no idea what any of that means, but, hey, I’m just a kid.”

261
Bond, as they reference components of the film’s soundtrack and the environment of the
movie theater, things that do not exist to Bond inside the world of the film. Colonel
Campbell similarly references buttons on the Playstation’s controller, actions available
through the game interface, and extra-diegetic components seen only by the player, yet
Snake does not seem to find anything odd in his comments. And indeed, such
moments, even in a serious-minded, cinematically realist game like Metal Gear Solid,
occur rather seamlessly. Such a moment in any other medium would be a stark
example of reflexivity, drawing attention to the artificiality of the text and reminding the
audience that the story exists outside of reality, yet in videogames, these references to
the mechanics of the game are relatively inconspicuous.
243
To the average gamer, this
is merely a standard device for teaching the player how to use the game’s controls.
Despite this device’s longstanding familiarity, this exchange between Campbell
and Snake indicates a slippage between the avatar of Snake and the player holding the
Playstation controller in hand—a supporting character’s address of the protagonist is
being confounded with the game’s address of the player. Further, this slippage is hardly
limited to such in-game moments. Numerous textual devices demonstrate this
modulation between player and character—many of which will be discussed over the
course of this chapter—however we need look no further than the discourse around
videogames to see evidence of this phenomenon. Just consider how we go about
referring to our avatars.
We often see such confusion in descriptions of gameplay or when providing
instructions, as commonly demonstrated in instruction manuals. The manual for Metal
Gear Solid sees just such a slip when describing elements of the on-screen display:

243
Metal Gear Solid actually does have its own deliberate moments of reflexivity later in the
game, which will be discussed later in this chapter, yet self-consciousness is hardly evoked
during these tutorial exchanges.

262
Life Gauge: Solid Snake’s remaining life. It is displayed when Snake receives
damage. When the Life Gauge reaches zero, the game is over. When a Boss
enemy is defeated, your maximum life is increased according to the difficulty
level you choose.
244

The first sentence refers to the gauge as Snake’s remaining life, while the fourth
sentence calls it your life. This occurs throughout the manual, sometimes addressing
the avatar in the second-person, as “you,” and sometimes in the third-person, as “he” (or
“Snake”).
Conversational discussions of gameplay typically see a similar slippage in
pronouns, as indicated in this excerpt from a review of Jak II:
The controls are spot-on -- Jak moves with a grace and smoothness that's
unmatchable. He switches from attacking, spins and lunges, to handling a rifle,
shotgun and mini-gun with gorgeous ease. … You can shoot enemies from afar,
run up and then smack them using the traditional spin attack, or you can hammer
them with the butt of your gun.
245

Again, we see a confusion between avatar and player, at first referring to the things “he”
can do, then the things “you” can do. For a comparable example, just look at my own
description of Metal Gear Solid’s opening moments that began this chapter—did you
take any notice of the shift between “I navigate the hero through the underground
docks…” to simply “I dart between a few rows of large storage crates…” in the following
sentence?
A comparable pronoun shift also occurs in the discourse amongst players within
and while playing games. This excerpt of a conversation in the massively multiplayer
online game Star Wars Galaxies: An Empire Divided aptly demonstrates, occurring as
two players take a breather after narrowly surviving a particularly harrowing battle:
Player 1: Damn, that was CRAZY!
Player 2: Nice bit of XP, though [experience points].
P1: My mind is almost at zero [mind points]—I thought I was done for!
P2: lol. Wanna head back to Mos Eisley and meet up with the others?

244
Metal Gear Solid Instruction Manual (Redwood City, CA: Konami of America, Inc., 1998), 15.
245
Douglass C. Perry, review of Jak II, IGN, posted on October 9, 2003,
http://ps2.ign.com/articles/454/454132p1.html.

263
P1: Sure, but I’m gonna have to log off soon. It’s almost 3am here, and I’m falling
asleep.
246

Note that the “I” reference wavers—“I thought I was done for” refers to the player’s
avatar, while “I’m falling asleep” refers to the player himself. Such exchanges are
common in online gaming, with players confounding references to themselves and to
their avatars (as well as other players and their respective avatars).
These examples abound throughout the medium. And certainly, most gamers
fail to notice them, having come to take such discourse just as much for granted as the
tutorial example in Metal Gear Solid, yet such examples certainly suggest a rather
curious relationship between players and their avatars, whether Solid Snake, Lara Croft,
Super Mario, Pac-Man, or any other. We’ve settled into a comfortable relationship with
the pixilated protagonists that we inhabit while trudging through the fantastic landscapes
of videogame worlds, enough so that we rarely give them a second thought. Yet as this
discourse indicates, it’s a considerably schizophrenic relationship. We exist both as a
participant in the gameworld inhabiting the body of the avatar (“I infiltrate the weapons
facility”) and as an observer outside the game watching the avatar (“He infiltrates the
weapons facility”) simultaneously, putting ourselves in two places at once.
Equally, then, the avatar also exists as two figures—the embodiment of the
player, yet also a discrete character in its own right. After all, I watched Solid Snake
enter the Shadow Moses complex of his own volition, listened to him converse with
Colonel Campbell, and developed a sense for his personality—a personality quite
different from my own—all before being offered any opportunity to provide input into the
game, allowing me to easily recognize Snake as a distinct figure outside of myself. And
yet I also recognize him as “me.” The first words on the back of the game’s packaging
clearly emphasize this point: “You are Snake, a government agent on a mission to

246
Conversation condensed for the sake of clarity and expediency.

264
regain control of a secret nuclear weapons base from terrorist hands…”
247
So I am
Snake, but I’m not Snake. Snake is me, but he’s also someone else.
Although a unique dynamic in today’s day and age, this sort of relationship is not
without precedent. In fact, it was spoken of thousands of years before videogames—or
anything remotely close to a computer—ever appeared.

Boomerang Etymology: The Evolution of a Word
Long ago, according to Hindu mythology, the demon Ravana performed a great
penance to the god Brahma, creator of the universe, which lasted for several thousands
of years. Recognizing this feat, Brahma finally appeared to Ravana with the offer of a
boon—an unbreakable promise of the gods. Ravana first asked for immortality, yet
Brahma denied this request, so Ravana instead asked for invincibility against gods,
demons, spirits, serpents, and wild beasts, a wish that Brahma begrudgingly granted.
With this newfound invulnerability for protection, Ravana assumed leadership of
the demon army of the underworld and headed toward Lanka—Sri Lanka by today’s
map—to usurp the kingship of the island. With the aid of the evil Rakshasas, Ravana
laid waste to the land, ravaging the good and the righteous, wreaking havoc over the
world, and—most disturbing to the gods—disrupting the rituals and sacrifices of Brahmin
priests.
Observing this devastation and sacrilege, the gods met with Brahma to discern
what could be done. Certainly, Ravana needed to be stopped, however the demon king
was now impervious to celestial attack. The gods remained helpless against him. Yet in
his arrogance and contempt for mortals, Ravana had neglected to request invulnerability
against humans, instead asking only for protection against those he thought legitimately

247
The italics are my emphasis; the ellipsis is included in the quote.

265
capable of causing him harm; mortals, he thought, were too insignificant and weak for a
being of his might to worry over. Vishnu was already known to take Earthly form in times
when dharma (justice and virtue) was threatened, and so Brahma implored Vishnu to
take a human embodiment so as to do battle with Ravana.
Meanwhile on Earth, a benevolent king was beginning to worry as to who would
inherit his kingdom, as the aging monarch had not yet been borne a son to succeed him.
At the recommendation of his trusted priests and ministers, the king organized a
sacrifice to ask the gods for progeny, and from the sacrificial fire emerged a divine being
sent by Vishnu, offering a golden urn of holy nectar to be drunk by each of the king’s
three wives. Before long, each wife was with child, and in due course, the queens gave
birth to Bharata, the twins Lakshmana and Shatrughna, and he who would be the king’s
eldest son and heir to the throne, Rama—soon to be recognized as the seventh great
incarnation of Vishnu.
248

And so begins the Ramayana, one of the great Sanskrit epics of Ancient India.
Literally translated as “The Travels of Rama,” the first six books of the Ramayana detail
the life and career of Rama, from his early feats and marriage to Sita, his exile from his
father’s kingdom, and the abduction and subsequent search for his wife, to his arrival in
Lanka, the ensuing war, and Rama’s eventual defeat of the demon king. The epic’s
seventh and final book then relates Rama’s life after the war, as well as his and Sita’s
eventual passing on to the next world.
The events of the Ramayana were not the first time Vishnu had manifested in
mortal form. By the traditional Hindi account, Vishnu had previously descended to Earth
in the forms of a fish, a tortoise, a boar, a man-lion, a dwarf, and the human

248
Story culled from: Swami Venkatesananda, The Concise Ramayana of Valmiki (Albany, NY:
State University of New York Press, 1988); and Arthur Anthony Macdonell, “Ramayana,” in
Encyclopedia of Religion and Ethics Part 20, ed. James Hastings (Whitefish, MT: Kessinger
Publishing, 2003), 574-8.

266
Parashurama, and following his incarnation as Rama, Vishnu also took shape in the
figures of Krishna and Buddha. His tenth descent, as Kalki, is expected to occur at the
end of time. Significant for our purposes is that while all ten of these forms are
recognized as embodiments of the god Vishnu, these beings also exist as earthly
creatures in their own right. Though perhaps technically correct, the terms
“embodiment” or “incarnation” to reference these appearances remain misleading;
rooted in the English “body” and the Latin “carne” (flesh), both terms indicate a mere
corporealizing of the deity, the taking of bodily form. While Vishnu is indeed “made
flesh” in these figures, these terms fail to indicate the complex relationship between the
deity and that body. Instead, ancient Hindus employed another term to reference such
figures—avatars.
From the Sanskrit ava [down] and tarati [to cross], the term avatarah literally
translates as a ”descent” of a deity from the heavens. Yet the concept is more complex
than this strict translation indicates, as an avatar is ultimately two beings in one: the deity
incarnate, yet also an earthly mortal. Note that Rama lives a long and eventful life
outside of his battle with Ravana, the task for which Vishnu took this human form. Even
after Ravana’s defeat, with dharma restored and the seventh incarnation’s purpose
fulfilled, Rama remains in the world to govern his father’s kingdom, living a long life
before crossing to the next plane. Further, Rama remains flawed, despite his divine
origins—in the final book of the epic, Rama, under pressure from his citizens, banishes
his wife from the kingdom, a decision he later regrets upon finding her in the forest years
later, along with the twin sons she bore him, now fully grown. These details suggest
Rama to exist as a mortal man of earthly flesh, outside of his status as an embodiment
of Vishnu, as the deity would be neither needed in the world after Ravana’s defeat, nor
subject to such human imperfections. Yet most indicative are the events of the climactic

267
Battle of Lanka itself, where Rama and his army finally defeat the demon king. During
the struggle, the gods come to the aid of Rama’s army, and Vishnu himself participates
in the battle. Vishnu comes to the aid of his own avatar, thus the two beings exist within
the same time and space simultaneously, suggesting them as two discrete entities. As
an avatar, Rama must be understood as an embodiment of Vishnu, yet also as an
individual unto himself.
While the Dasavatara—the ten great avatars of Vishnu—are by far the most
famous applications of the term, the word “avatar” is prevalent beyond these instances.
Other accounts cite twenty-five
249
avatars of Vishnu rather than the traditional ten, while
acknowledging the existence of innumerable others, and many Hindu figures are
attributed as avatars of other deities—Rama’s wife, Sita, for example, is considered an
avatar of Lakshmi. Further, adherents of dharmic tradition often use the term in order to
explain similar incarnations of God from other religions, as the avatar concept sees
parallels in numerous other faiths. The Christian conception of Jesus Christ is probably
the most obvious example from Western cultures, and it is indeed strikingly similar. By
Christian belief, Christ was the embodiment of God, while having simultaneously existed
as the mortal Son of God, who after death “ascended into heaven and [now] sitteth at
the right hand of God the Father Almighty,” suggesting God the Father and God the Son
to be separate figures.
250

Despite the close similarities between the Hindu use of this term and its
contemporary application in videogames and other digital arenas, the word “avatar” took
a somewhat circuitous route toward its modern usage. Long before finding its way into

249
Or by some accounts, twenty-two or twenty-three.
250
From “The Apostles’ Creed,” as translated in The United Methodist Church, The Church of
England, and others. Other Christian denominations, such as the Roman Catholic Church, use
close variations on this rendition.

268
digital applications, the term expanded beyond its Hindu origins to signify not only the
incarnation of a deity, but other embodiments as well:
av·a·tar (ăv'ə-tär') n.
[Sanskrit avatārah, descent (of a deity from heaven), avatar : ava, down +
tarati, he crosses.]
1. The incarnation of a Hindu deity, especially Vishnu, in human or animal form.
2. An embodiment, as of a quality or concept; an archetype: the very avatar of
cunning.
3. A temporary manifestation or aspect of a continuing entity: occultism in its
present avatar.
251

— The American Heritage Dictionary of
the English Language

Broken away from a strict attachment to individuals or deities, the word essentially
becomes synonymous with “embodiment”—any tangible thing (typically a person or
object) that exemplifies something intangible. That intangible element might be an
expansive concept:
“High shouts of exultation, in every dialect, by every vehicle of speech and
writing, rise from far and near over this last avatar of Democracy...”
252

a characteristic:
“It is the sadness of one who, like the youth in Alastor, has beheld in dream a
veiled maid, the avatar of perfect love and beauty, … who alone gives meaning
and harmony to the fret and turmoil of human life.”
253

or an abstracted paradigm:
“Political conditions had thus made Sainte-Beuve one of the high priests of anti-
clericalism and free-thought. A curious avatar, no doubt, for the agile and many-
sided critic, whose mind seemed to grow more active and broader with
advancing years!”
254

As these examples demonstrate, the term’s migration out of a religious context saw its
meaning collapsed and simplified to a mere synonym for “embodiment,” “manifestation,”

251
“Avatar,” The American Heritage Dictionary of the English Language, 4th ed. (Boston:
Houghton Mifflin, 2000). http://www.bartleby.com/61/36/A0543600.html (accessed June 25,
2006).
252
Thomas Carlyle, Latter-day Pamphlets (n.p.: Chapman and Hall, 1850), 13.
253
Edward Dowden, The Life of Percy Bysshe Shelley (n.p.: Lippincott, 1887), 127.
254
Albert Léon Guérard, French Prophets of Yesterday (n.p.: T.F. Unwin, 1913), 205.

269
“personification,” or “archetype,” in all cases without the unique complexities of the
word’s original use.
The word then found its way into videogame and computer culture via several
routes, and resultantly, the recent history of the word gets muddled, with multiple
sources often credited for coining the term in application to a computer user’s digital
body. I can’t claim to give official etymological credit to any one of these, yet I can, at
the very least, lay out the historical touchstones of the term’s deployment.
William Gibson is often credited with introducing the modern conception of a
digital avatar in his 1984 landmark work of cyberpunk fiction, Neuromancer, with its plot
concerning “cyber-cowboys” (hackers) navigating a digital cyberspace environment
known as the matrix. By Gibson’s account, cyberspace exists as a world with its own
spatiotemporal dimensions, entirely separate from those of our own:
Cyberspace, as the deck presented it, had no particular relationship with the
deck’s physical whereabouts. When Case jacked in, he opened his eyes to the
familiar configuration of the Eastern Seaboard Fission Authority’s Aztec pyramid
of data.
255

Here, data manifests in three-dimensional form in a fully-navigable spatiotemporal
environment with no relation to physical reality. Resultantly, cyber-cowboys must
navigate this space via a digital representation fairly similar to our modern concept of an
avatar, though quite detached from the traditional Hindu account of such a figure.
Case’s digital body hardly exists outside of Case himself, and by occupying this digital
embodiment, Case is described as simply moving from one world into another; the novel
does not give any impression of the two existing separately nor of Case functioning in
both worlds simultaneously. Further, to describe these digital bodies, Gibson chooses
the term “constructs” rather than the more recent “avatars,” a choice he continues to
stick by in his later novels. All the same, Gibson’s name does seem to arise in many

255
William Gibson, Neuromancer (New York: Ace Books, 1984), 105.

270
discussions around the origin of computer avatars, and this association seems fair
enough—we must at the very least credit Gibson with inspiring some piece of our
modern concept of an avatar, as Neuromancer has proven a seminal work ingrained in
most accounts of digital environments and digital bodies since.
256

The first deployment of the word itself in relation to computers—and equally, its
first use in a videogame—arose in a 1985 role-playing game for the Apple II. The
application of the term was quite different from Neuromancer’s constructs, and does
seem predicated on the religious overtones of the term’s original usage, if not its precise
religious meanings. In Ultima IV: Quest of the Avatar, the titular avatar is indeed an
embodiment, yet it is neither of a deity nor of a cyber-cowboy. Instead, the avatar is an
embodiment of virtues. This, of course, is not surprising for this moment in time, with the
word typically used to indicate an embodiment of a quality or concept.
Unlike most role-playing games that preceded it, the central objective of Ultima
IV is not to seek out and conquer some supreme evil, as was the standard premise for
such fare; rather, the hero’s quest is to discover and understand the game’s Eight
Virtues in order to lead a virtuous life and thus become a spiritual leader for the kingdom.
By proving his understanding of these virtues and meditating at each of their eight
shrines, the hero becomes the Avatar—the embodiment of virtue. In subsequent games
of the Ultima series, the protagonist continues to be known by this title, thus the hero of
the franchise comes to be called simply “the Avatar.”
Because of the vagaries of continuity in the series, it is unclear whether every
protagonist in the ensuing Ultima games are the same specific character of the Avatar,
or whether instead the individual protagonists of each title are, generally speaking, the

256
No doubt, Gibson’s credit in this regard is also linked to his coining of a related term that did
catch on and enter the public vocabulary: cyberspace. Next to videogames, the data-based world
of the internet—colloquially referred to as cyberspace—is probably the most frequent user of the
term “avatar.”

271
Avatar of their respective game. Keep in mind that players are afforded the opportunity
to customize their character in numerous ways, including name, gender, profession, and
in later sequels, appearance, thus “the Avatar” is merely a title given to the game’s
protagonist, whoever that character may be, rather than the name of a specific
individual—I may play a bard named Wagimyr while someone else may create a paladin
named Barney, yet both characters become “the Avatar” for that particular playing.
Further, in Ultimas I, II, and III, the protagonist is simply known as “the Stranger,” and
Ultima IV suggests these to be three different characters, each of whom are also
separate from the Avatar. Yet later sequels seem to imply that all the games’
protagonists have been the same individual. Regardless, the Avatar has become
accepted as the franchise protagonist by many accounts, and the flexibility of character
creation in each installment leaves the protagonist vague enough to be read either way.
However, while the Avatar was indeed an avatar—that is, the figure that
embodies the player within the digital gamespace—this was not the intention of the
term’s usage. That the character of the Avatar happens to be the player’s avatar, one
could argue, is but a retrospectively serendipitous choice of name. On the other hand,
one could also argue that the evolution of the word was not so coincidental, that the
Avatar and the Ultima franchise were influential on our more colloquial use of the term,
much like Kleenex, Band-Aid, and Xeroxing have come to indicate more than their
individual brands—after all, it’s not such a far leap from referring to the player-controlled
protagonist of each Ultima installment as the Avatar, regardless of each character’s
individual name and description, to referring to such protagonists in any game as an
avatar, regardless of the individual character. By either account, we can be certain that
when designer Richard Garriott chose to name his hero “the Avatar,” the term was in no
way meant to convey the character’s status as the player’s digital embodiment.

272
Credit for the term’s first deliberate usage in modern parlance instead goes to a
groundbreaking but short-lived online role-playing game from LucasFilm, a game that
also marked the term’s first application in a “cyberspace” environment (though the early
capabilities of the Internet were certainly nothing close to the world that Gibson had
envisioned). The game was entitled Habitat, and although the game did not make its
beta test debut until 1987, the game was first designed by Randy Farmer and Chip
Morningstar in 1985, and went online by 1986, giving Ultima IV some competition for the
claim of first usage (either way, Ultima IV certainly retains its claim to the first public
usage).
257
An August 1986 issue of RUN magazine featured an article on Habitat that
opened with this description:
Habitat: A make-believe world inhabited by small, colorful creatures, called
Avatars. Human beings may visit Habitat and move freely about its regions,
interacting at will with Avatars. Human beings reach Habitat by traveling many
miles through tiny telephone lines and entering through a large gateway, called
QuantumLink [a dial-up service].

Once a human being enters Habitat, he or she takes on the visual form of an
Avatar, and for all intents and purposes becomes one of these new-world beings.
In the world of Habitat, people can play games and go on quests, but mainly they
meet other people and have fun.
258

Here, we see the term deployed in the manner we’ve become accustomed to in more
recent years: in reference to the player’s digital embodiment. The term has not yet
reached the “generic” quality of referencing any digital embodiment, as indicated by its
use of a capital “A”—“Avatars” are the specific title given to the beings in this game, and
not meant to be applicable to the characters of other games or online environments. Yet
Habitat plants the seeds for future usage of the term through its description of those
creatures, as well as the player’s relationship to them. Further, note that this description
introduces the characters independent of their users (“A make-believe world inhabited by

257
Chip Morningstar and F. Randy Farmer, “The Lessons of LucasFilm’s Habitat,” Cyberspace:
First Steps, ed. Michael Benedikt (Cambridge: The MIT Press, 1991).
258
Margaret Morabito, “Enter the On-Line World of Lucasfilm,” RUN, August 1986, 24.

273
small, colorful creatures, called Avatars”) before mentioning the player’s ability to “take
on the visual form of an Avatar” and become one of these beings. Already, the term is
moving back toward an indication of beings that have some semblance of existence
outside of their users.
The term found intermittent usage over the next few years—the 1989 paper-and-
pencil role-playing game Shadowrun, for example, referenced its player-characters as
avatars. Yet the term did not seem to be properly legitimized for widespread deployment
until its appearance in another cyberpunk novel, with Neal Stephenson’s Snow Crash in
1992. Unlike Gibson’s “constructs,” Stephenson’s “avatars” stuck in the geek-
community vernacular.
In telling the story of hacker/samurai/pizza delivery driver Hiro Protagonist,
Stephenson introduces his usage of the term “avatar” with this passage:
As Hiro approaches the street, he sees two couples probably using their parent's
computer for a double date in the Metaverse. He's not seeing real people, of
course. It's all part of a moving illustration created by his computer from
specifications coming down the fiber optic cable. These people are pieces of
software called avatars. They are the audiovisual bodies that people use to
communicate with each other in the Metaverse.
259

The Metaverse is described as a vast virtual reality-based world roughly 1.6 times the
size of the Earth, and could be conceived as an evolution of or successor to the
contemporary Internet. By logging on through a computer terminal, denizens navigate
the Metaverse via these personalized digital bodies, which can be designed and molded
to the user wishes. Resultantly, a user’s avatar becomes something of a status symbol
in Stephenson’s world, with highly-detailed bodies indicating superior technical acumen
as a hacker and programmer, and low-res bodies often betraying the use of a lowly
public access terminal. Also note that unlike Ultima IV or Habitat, Stephenson deploys
the word with a small “a” rather than a capital “A,” indicating a more general use of the

259
Neal Stephenson, Snow Crash (New York: Bantam Books, 1992), 33.

274
word. But much like Gibson’s constructs, these avatars hardly engage the two-beings-
in-one dimensionality given to those of Hindu mythos.
However, Stephenson’s deployment of the term does neatly match its standard
usage in today’s computer vernacular:
avatar n. Syn. 1. Among people working on virtual reality and cyberspace
interfaces, an avatar is an icon or representation of a user in a shared virtual
reality.
260

— The New Hacker’s Dictionary

avatar A digital “actor” or icon that represents who you are and where you are in
the virtual world.
261

— Netlingo: The Internet Dictionary

avatar 1. A graphical representation of a person in an IRC chat room, interactive
game, or other area of cyberspace. An avatar can be a cartoon drawing, picture,
or other item that the user chooses to represent his or her virtual identity.
262

— Dictionary of Computer and Internet Words:
An A to Z Guide to Hardware, Software, and
Cyberspace

avatar [COMPUT SCI] A virtual representation of a person or a person’s
interactions with others in a virtual environment, conveying a sense of someone’s
presence…; examples include the graphical human figure model, the talking
head, and the real-time reproduction of a three-dimensional human image. (‘av-
ə,tär)
263

— Dictionary of Computing & Communications

avatar An electronic image or other embodiment of an individual which is
computer-generated and holds some essence or presence of individuality or
actuality beyond that of photographic image or scan. This concept is prevalent in
the imaginary world of computer gaming (and Internet chat areas), especially in
virtual reality simulators.
264

— The Telecommunications Illustrated
Dictionary

260
Guy L. Steele, The New Hacker’s Dictionary (Cambridge: The MIT Press, 1996), 50.
261
Erin Jansen, Netlingo: The Internet Dictionary (Ojai, CA: NetLingo Inc., 2002), 50.
262
American Heritage Dictionaries, Dictionary of Computer and Internet Words: An A to Z Guide
to Hardware, Software, and Cyberspace (Boston: Houghton Mifflin Co., 2001), 22.
263
MacGraw Hill Book Company, Dictionary of Computing & Communications (New York:
McGraw-Hill Publishing, 2003), 24.
264
Julie K. Petersen, The Telecommunications Illustrated Dictionary, 2
nd
Edition (Boca Raton, FL:
CRC Press, Inc., 2002), 83.

275
The term has come to find a broad variety of referents, from the complex, immersive
bodies of virtual reality to the simplistic names and icons associated with Internet chat
rooms and instant messaging—and of course, our subject at hand, the player-controlled
protagonists of videogames.
265

Ultimately, neither Ultima IV, Habitat, nor Snow Crash can be single-handedly
credited for defining the term’s current usage. A better account would be to credit the
three’s combined usage of the term for reinforcing one another and influencing the
cultures of gaming, computer science, and science-fiction—cultures that traditionally see
a significant amount of participant overlap—to take up the term. Without Habitat or
Snow Crash, the word may have simply remained a character’s title in a role-playing
franchise; without Ultima or Snow Crash, the word may be remembered merely as the
curious name of the beings in an online world; and without Ultima or Habitat, Gibson’s
“constructs” may have won out over Stephenson’s preferred term.
I detail the roundabout evolution of this word to demonstrate the broad inspiration
of its Hindu deployment that led to each of these sources employing the word, as well as
the distance the word has traveled away from the complexities of that original meaning—
neither Ultima IV, Habitat, nor Snow Crash employ the word for the sake of the two-
beings-in-one dynamic that characterized the avatar Rama’s relationship to Vishnu
(though Habitat might mildly hint at this dynamic). It is somewhat fortuitous, then—as I
will argue in the pages that follow—that the word’s initial meaning is actually very
appropriate to its contemporary usage, as it aptly characterizes the dual-positioning
inherent in videogame players’ relationship to their avatars.

265
It should be noted that I am not concerned with these other deployments of the term “avatar,”
and as such, any of my statements made in regards to avatars should not be taken outside of the
context of videogames.

276
A Spectrum Approach to Dual-Positioning
Functionally, the avatar is essential to the videogame situation—this game body
is what allows the player diegetic presence. Provided agency and permitted to take
action within the text, we, as players, are offered participation in the gameworld and
allowed to conceptualize ourselves as being within that space—the earlier discussion on
videogame discourse (“I infiltrate the weapons facility”, “I thought I was done for”, etc.)
demonstrates the point. Yet simultaneously, the diegetic world also remains
“somewhere else,” not present to us (“here”) but rather on the other side of the screen
(“there”), thus an avatar must be offered as our proxy. The game situation cannot exist
without an avatarial presence to mediate our experience, to become our representative
in the game (even if that presence is only assumed, as in a first-hand POV, or
abstracted, as in Bejeweled or Pong), and yet despite the vitality of this device to the
medium—comparable to such fundamental devices as continuity editing in the cinema or
the narrating voice of literature—little work has been done toward examining this figure.
While cinema, theater, literature, and other previous media might similarly
present their audience with a diegetic world, such media rarely suggest the audience
member to be a participant in that world. Certainly, we romanticize the experience of a
good story transporting us to another place, allowing our own reality to fall away as we
are immersed in the fictional world of a good book or film, however we are never given
any agency within such texts, nor do we conceptualize ourselves as existing within their
worlds. For while we might allow the real world to fall away, we need remember that our
selves fall away with it; we become only invisible, disembodied observers with no
presence in the diegetic space.
In videogames, however, we are indeed offered diegetic presence—a presence
only made possible by the avatar’s mediation. With the player existing entirely outside of

277
the game text, and the diegetic world remaining inside the game, the avatar acts as the
bridge connecting these two components divided by the screen.

The avatar operates as both a systemic component of the formal interface—allowing the
player access to the formal operations of the game—and a diegetic presence within the
gameworld, providing the basis for our four-part model of videogame discourse.
In addressing players’ relationship to videogames and their positioning therein, a
common assumption seems to be that players simply identify as the avatar, and relate to
the gameworld as a present participant by inhabiting that character themself. This
seems to be the general notion involved whenever we begin discussing immersion, a
continuing buzzword in game studies that scholars have been asserting in one form or
another since Janet Murray famously invoked it in Hamlet on the Holodeck,
266
and a
concept which game designers, game reviewers, and gamers themselves tend to revere
as the Holy Grail of the medium. Indeed, games are typically described in the most
immersive of terms: they “suck” you in, they “plunge” you into their worlds, they let you
“become” their characters. But as the Metal Gear Solid example demonstrated, diegetic

266
Janet Murray, Hamlet on the Holodeck (New York: The Free Press, 1997).

278
immersion—identifying one’s self as the avatar within the gameworld—is typically not the
only thing going on in this medium.
Consider the most conventional of examples: Super Mario Bros. Here, I’m
supposed to be Mario, a character in the narrative. I control his movements, tell him
when to run, jump, duck, and so on. And yet, I’m watching him on the screen, like a
cinemagoer, as a character outside of myself. He often performs actions outside of my
control—he even waves to me at the end of each level!—and he has a distinct persona
apart from my own. Much like the avatars of Hindu belief, Mario is my embodiment on
an alternate plane of existence while equally a discrete being unto himself, and
resultantly, I’m positioned as a character within the diegetic world and as an observer
watching that character from outside of the diegetic world simultaneously, addressed in
two places at once. Such is the dual-positioning inherent to the videogame medium.
Any account of the player’s experience or their relationship to the avatar would need to
account for both of these positions, distinguishing them as separate and yet
acknowledging their simultaneity.
In the brief history of interactive and gaming studies, others have already
suggested models of perspective and identification, and they have typically
acknowledged some division between two opposing roles. For example, in establishing
a typology for ergodic literature, Espen Aarseth distinguishes between the personal and
impersonal perspective, where personal roughly equates to identifying with an avatar,
being positioned as a “you” within the narrative, while impersonal belongs to the outside-
looking-in perspective of an observer.
267
Marie-Laure Ryan similarly recognizes an
internal/external binary:

267
Espen J. Aarseth, Cybertext: Perspectives on Ergodic Literature (Baltimore: The Johns
Hopkins Press, 1997) 63. It should be noted that perspective is just one of seven elements in

279
In the internal mode, the user projects himself as a member of the fictional world,
either by identifying with an avatar, or by apprehending the virtual world from a
first person perspective. In the external mode, the reader situates himself outside
the virtual world. He either plays the role of a god who controls the fictional world
from above, or he conceptualizes his activity as navigating a database.
268

The problem with both these models, however—along with many other examples that
follow their lead—is that they conceptualize the player’s positioning as an either/or
situation; the player either inhabits an avatar or retains an outside, god-like perspective,
but not both. If forced to categorize, we could probably classify most titles under one
heading over the other, however this either/or proposition doesn’t really account for the
experience of most videogames.
Truth be told, there’s always some degree of both positions at work in any given
videogame—always an external, god-like identification because we are always in a
position of privilege outside the narrative, and because we always have access to the
narrative world via the computer interface; always an internal, character-based
identification because we always have some point of interaction within the game itself,
allowing us to recognize our presence within its fictional world.
269

We might then imagine a gamut running between these two extremes, with the
player positioned somewhere in the middle:

Aarseth’s typology, and he does not pay any particular attention to this over any other of those
elements in establishing that typology. In short, this is part of a larger argument, not the heart of
any particular discussion.
268
Marie-Laure Ryan, “Beyond Myth and Metaphor: The Case of Narrative in Digital Media,”
Game Studies vol. 1, issue 1 (July 2001), http://gamestudies.org/0101/ryan/ (accessed April,
2003). Also read as the keynote speech at the Computer Games & Digital Textualities
conference, Copenhagen, March 2001.
269
Though I dispute the use of an either/or model, I otherwise think Ryan’s terms of internal and
external are quite appropriate, and as such, see no reason to employ a new set of terminology.

280
On one end of the continuum, the player identifies themself within the game’s space by
inhabiting an avatar of some sort—whether a tangible body that appears on-screen or a
presence implied by their activity—while on the other, the player takes a disembodied
position outside of that space. Essentially, these poles represent the ideal of virtual
reality on one end—complete and seamless immersion—and the viewing situation of
cinema on the other—complete spectatorship, positioned as a removed overseer outside
the diegetic space. However, no videogame fully reaches either end; reaching one
would be to attain the fantastic ideal presented in such sci-fi paradigms as Star Trek’s
Holodeck or The Matrix’s titular cyberworld; reaching the other would be to completely
remove the player from the game situation by either rendering the player entirely inactive
in the text’s goings-on or turning the text into an “interactive movie” (which, while
perhaps related to videogames, is generally a different object).
This model is not without precedent, as it echoes a similar approach that film
scholars have often employed when discussing identification in the film spectator. Judith
Mayne notes, “It is tempting to make a short-circuited argument about the nature of
identification in the cinema and to assume that the primary form it takes is identification
with characters, with human figures on screen.”
270
The control which the player exerts
over an avatar makes this assumption all the easier in gaming. Yet Mayne goes on to
note that anterior to any possible identification with characters on screen is an
identification with the viewing situation itself. This notion was perhaps best established
by Christian Metz in making his distinction between primary and secondary identification.
Metz noted that while the spectator in cinema does indeed identify with the protagonist
on screen, that identification is only secondary, as the viewer’s primary identification

270
Judith Mayne, Cinema and Spectatorship (London: Routledge, 1993), 26.

281
remains with the camera or projector itself, or better yet—in his words—with “myself
looking.”
271

Certainly, the player is allied with the computer or console as source and access
to the gameworld in much the same way the cinemagoer must identify with the camera
as source of the filmic image. Regardless of the connection we may feel toward our
avatar, we are always positioned (literally) behind the computer interface, on the other
side of the screen and apart from the projected content, thus the computer’s view of the
world becomes our own. The player’s mastery and control over the diegetic space
furthers this connection. Yet we must rethink Metz’ primary/secondary distinction here.
Keep in mind that when we discuss “identification” with characters in the cinema, it is
largely a metaphorical term—in truth, the audience would be better described as
empathizing with the protagonist rather than identifying themselves as that figure. Noel
Carroll notes:
[W]hether the notion of character-identification is merely a metaphor or is meant
to be a literal description of a mental state is generally not determinable in the
context of daily conversation. …Clearly, if the notion of character-identification is
to make sense, it cannot be based on postulating an audience illusion of being
identical with the protagonist.
272

As such, it is no wonder that identification with the protagonist in cinema must remain
secondary to our literal position as spectator. Yet videogames provide a different
situation, as they do indeed allow players to locate themselves within the text—this is
what the avatar provides. By taking control of the avatar and interacting with the diegetic
world through that body, the player can recognize themself as an active participant in the
fictional content. This is precisely why the videogame player can answer the question
“where are you on the screen?” while the cinemagoer cannot.

271
Christian Metz, The Imaginary Signifier, 52.
272
Noel Carroll, The Philosophy of Horror (New York: Routledge, 1990), 89-90.

282
Think about that for a moment. Imagine yourself sitting in a darkened theater—
or simply in front of your television—watching your favorite movie. Were I to walk into
the room, look at the screen for a few moments, then ask, “so, where are you in this?”
you’d probably think I was crazy (or completely unfamiliar with narrative cinema). You’d
respond with something along the lines of, “what are you talking about? It’s a movie, I’m
not in it.” Yet consider the same situation while playing a videogame. Were I to walk
into the room and ask the same question, you’d most likely respond without hesitation:
“I’m the mustachioed fellow in overalls,” or “the little frog hopping across the freeway,” or
“the Kombatant on the left.” Even without a specific anthropomorphic body to reference,
you’d still have little trouble explaining your presence: “I’m the paddle on the left,” or “I
control all the players of the team in blue,” or “I’m driving the red car,” or “I’m not
onscreen because you’re seeing everything from my perspective.” Even without a
visible character to reference, you probably wouldn’t think twice about answering the
question—despite obviously sitting in the room in front of me, the question still seems
completely sensible and readily answerable. That’s the power of avatars. So while Metz
and others deploy the term “identification” to reference our relationship to filmic
protagonists, suggesting some degree of coexistence between audience and character,
everyday discourse reminds us of the distance that remains between these figures; we’d
probably better respond were the question posed as “who are you following?” or “who
are you rooting for?,” questions that keep “you” and the character as separate figures.
Yet in videogames, common discourse sees no problem in suggesting a simultaneity
between player and avatar. Much like the earlier examples of pronoun shifting—
indiscriminately switching between “I/you” and “he/she” when referencing an avatar—
everyday conversation reveals the intimate connection between audience and character
in this medium, and a significantly more literal application of the word “identification.”

283
Without one position primary to the other, then, videogames are afforded the
opportunity to offer a wide variety of identification options, with some games immersing
the player in character, others positioning the player as a god-like overseer, and still
others deploying complex amalgamations of the two. Indeed, the medium excels at
affording a multiplicity of positioning options, with the vast majority of videogame texts
negotiating these two extremes in complex ways, and such complex amalgamations are
precisely why a simple either/or model of internal vs. external positioning fails us—the
textual cues that influence our reception of any particular game are typically not steering
us toward only one end of the spectrum. Some cues push us one way, some push us
another, and some are ambiguous from the get-go. On first glance, such a text would
seem a plethora of contradictions creating a disjointed, schizophrenic experience, yet
remember that the very premise of an avatar is based on these same contradictions—
being in two places at once. Thus in a well-designed game, these would-be
contradictions are deployed in manners that create a seamlessly rich and fluid play
experience. Such is the case with Metal Gear Solid.

A Multiplicity of Positioning Cues: Textual Influences on Player Positioning
To provide a framework for such textual cues, I have chosen ten areas that
impact the player’s relationship to their avatar. These distinctions may be mildly
arbitrary, as there is an obvious bit of overlap between them, yet the specific divisions
are not my ultimate purpose. The categories offered here are merely tools for
recognizing the wide variety of positioning cues that are at work in any single game,
continually adjusting the player’s relationship to their avatar and to the diegetic world of
the game. Their essence lies in their ability to reveal the complex configurations of
player positioning at work in most videogames. Rather than simply dropping the player

284
at one end of a positioning binary—inside or outside of the videogame’s world—
videogames typically provide a variety of interrelated and often conflicting positioning
cues that create composite, nuanced, and fluid experiences along the broad spectrum
between these poles.
We will look at each of these categories in turn, examining the positioning options
they entail and considering several different examples of the variety that each allows.
Yet as a particularly rich yet equally demonstrative example, we’ll pay particular attention
to Metal Gear Solid as a game providing vastly conflicting cues regarding the player’s
involvement.

Embodiment: character vs. observer
Recall the opening words on the back of Metal Gear Solid’s packaging—“You are
Snake”—as well as the manual’s tendency to reference the player (“you”)
interchangeably with the avatar (“Snake” or “he”). Outwardly, the game seems clear in
positing the player as being one with the avatar of Solid Snake, and much of the game’s
discourse enforces this claim, suggesting the player to be embodied in that character.
Every videogame entails an avatar of some sort—as I’ve already indicated, such a
presence is requisite of the medium. Yet as our first concern, we consider the nature of
that presence and the degree to which the player recognizes themself within that figure.
In discussing the player’s embodiment, I am referring to the manner and the
extent to which the player is “present” in the gameworld. Does the player recognize
themself as an avatarial character in the diegesis? Or is the player a disembodied
observer without a tangible presence in the game? If there is a clear avatar available, to
what extent does the player identify themself as that character? Does the player simply

285
occupy that body, making no distinction between themself and the protagonist, or do
they regard the avatar as a distinct individual outside of themself?
On one end of the continuum, we might consider a game like Bioshock, which
eliminates all but the most inherent elements of a character outside of the player’s own
consciousness. Not only does Bioshock provide a clear and unambiguous embodiment
within the diegetic world—the character of Jack, the sole survivor of a plane crash with
only a faint memory of his past—but the game provides very little distinction between
Jack and the player. With the entirety of the game occurring from the avatar’s point of
view, the player is literally put into Jack’s body, receiving the entire narrative through that
character’s eyes. And while navigating the underwater dystopia of Rapture, the player is
provided very little indication of a character outside of themselves—the tattooed hands
that extend before their eyes indicate a white male, and the unfolding narrative fills in
details of Jack’s forgotten past, yet so long as the player is willing to role-play and accept
this loose persona, they can easily merge their own identity with his. Resultantly, the
player never sees the avatar as a character outside of themself—we never hear Jack’s
voice, see Jack’s face, or witness Jack in any way acting outside of the player. We
essentially experience the game from “inside” that character.
Compare this embodiment to that of SimCity 4, in which the player maintains a
burgeoning metropolis from a god-like vantage far above the landscape. The game
does suggest an embodiment for the player by addressing the player as the city’s mayor,
thus positioning them as a diegetic member of the game’s populace, and various civic
advisors and individual citizens directly address the player as such. Yet this is much
more of a conceptual embodiment than a tangible one, as the player’s deity-like
omnipotence and omnipresence defies any corporeal manifestation of that role. The
player engages the gameworld from a god-like vantage far above the city, able to view

286
the entire metropolis at once or zoom in to an individual street corner, and they are
equally immune to diegetic time, able to slow down, speed up, or even stop the
advancing of the clock. Further, the player wields more power and control over the
microcosm than any one human—much less an elected civil servant—could possibly
attain; not only can they make zoning decisions, change tax rates, build hospitals, or
bulldoze entire city blocks at will, but equally terraform the land, carve rivers, raise
mountains, spawn wildlife, and generate literal acts of god: tornados, meteors,
earthquakes, volcanoes, or even a giant killer robot from space. The problem of the
mayor’s house conveys the game’s confused perspective: early in the game, the player
is allowed to build their own house within the community; they can witness citizens
coming to visit or gathering outside, and it is assumed that the mayor is inside that
house. But if such were the case, then the player is clearly not the mayor, instead
looking down on that figure as a deity-like figure from above. In the end, the mayor
embodiment is a metaphor at best, leaving the player without any palpable embodiment
in which they can locate their presence.
The standard situation, however, is somewhere between the two, the situation
that we encounter in Metal Gear Solid, Tomb Raider, Grand Theft Auto III, Super Mario
Bros., and countless other titles, where the player functions as both character and
observer. The player can certainly locate their presence in Solid Snake or Lara Croft—
this is the body that allows them to act within the gameworld. But on the other hand, the
player still views Snake and Lara as “someone else”—they can see these characters on
the screen, often performing actions outside of their control, and the player does not
receive the gameworld from their perspective. This is the standard hesitancy of
identifying with an avatar: we know that the avatar stands in our stead, yet the game

287
equally provides numerous cues to suggest otherwise, keeping us both embodied and
disembodied at once.
More complicated are those games which feature multiple embodiments. Many
videogames force the player to switch characters in mid-narrative—in Halo 2, the player
begins by controlling the avatar of the Master Chief as he battles the Covenant, an alien
race at war with the forces of Earth. However once the game’s first level ends, the
player is switched into the role of the Arbiter—not only a separate character, but a
member of the very race they had just been battling! The player continues to alternate
between these two figures for the game’s entirety, shifting allegiances and essentially
playing both sides of the war. How can the player be said to inhabit either figure if they
are repeatedly transferred into the body of a character antithetical to their goals? Metal
Gear Solid’s sequel presented a similar situation, as the player begins in the role of
Snake, yet switches to the new avatar of Raiden midway through the story (with Snake
later becoming an ally). Other games, such as the Final Fantasy series, allow the player
to control multiple avatars simultaneously—Final Fantasy VIII begins with a single
avatar, Squall, yet other characters quickly join him, and the player has full control over
those avatars as well. There are even points in the game when the player can leave
Squall out of their party of active characters, thus controlling three avatars, none of
which are the primary hero of the game. All three of these games would seem to
promote rather fragmented game experiences, yet nonetheless, they are able to
maintain narrative consistency and players have little trouble making sense of the action
and engaging the gameworld. The player’s embodiment can only be explained as
negotiating between character and observer, caught somewhere in-between and fulfilling
both roles simultaneously.

288
Avatar autonomy: inertia vs. sentience
Among the more notable features at the time of Metal Gear Solid’s release was
its extensive inclusion of cinematic vignettes and “talking head” radio conversations,
both used to reveal vital plot exposition, provide character development, incorporate
relevant background information, and develop the game’s vivid atmosphere. These
conventions allowed for a lengthy and highly dramatic narrative, offering a more
sophisticated mode of storytelling unavailable to earlier titles—yet it also carried
undeniable problems. While the game’s advocates hailed its complex narrative and
vigorously elaborate atmosphere, its detractors supplied an alternate take: Solid is
essentially a protracted movie with passages of gameplay interspersed throughout,
rather than a videogame using cinematic techniques. Regardless of which side we wish
to take, let us consider this technique’s effect on the player’s relationship to the game
and their avatar.
While actual gameplay provides a state of internal embodiment in the avatar (to
one degree or another), cinematic passages—moments during which the player
provides no input—are essentially the opposite. During cut-scenes, we suddenly lose
control of Snake; he converses, moves about, and performs actions entirely independent
of our control. We are essentially ejected from our game body, thrown into a situation of
watching “ourselves.” The opening sequence of Metal Gear Solid demonstrates this
phenomenon: before the player is allowed any participation, Solid Snake scuba dives
into the Shadow Moses compound, emerges from the icy water in the facility’s
underground docks, and holds a lengthy conversation with Colonel Campbell and other
operatives via radio, all without any input from the player, and such passages continue
to occur throughout the game. If the player is expected to be Solid Snake, how can they
explain witnessing these events outside of their control? In such instances, the game

289
falls back on the conventions of cinema, positioning the player as an invisible,
disembodied observer watching the story unfold on the screen rather than allowing the
player to inhabit the protagonist and experience the narrative firsthand. Exacerbating
this issue, Metal Gear Solid contains such an inordinate number of cut-scenes that the
player can actually spend more time watching Snake than actually playing him.
With avatar autonomy, then, we are concerned with avatar’s existence outside of
the player’s control. Is the avatar an inert figure, reliant on the player’s actions to give it
life, or is it depicted as an autonomous character capable of acting of its own volition? Is
the avatar’s body simply a hollow vessel waiting to be inhabited by the player, or do we
read the character as a sentient personality outside of ourselves, similar to a character in
a movie or a book?
Bioshock amply demonstrates the hollow vessel avatar—Jack barely exists
outside of the player, and he never acts, moves, or speaks outside of the player’s
control. He remains completely static unless acted upon. Yet this is a rare exception in
the medium, as most avatars exhibit some degree of autonomy or sentience,
distinguishing them as individual from the player.
The most blatant examples of this phenomenon follow from conventions lifted
from traditional narrative media, conventions rooted in telling someone else’s story—not
only cut-scenes, but first-person narration, autonomous dialogue of any sort, and
development of the protagonist as a character, all conventions meant to depict
protagonists as sentient and autonomous beings. We see all three of these in Prince of
Persia: The Sands of Time, beginning with an opening cinematic narrated by the prince
himself.
Most people think time is like a river that flows swift in sure in one direction, but I
have seen the face of time, and I can tell you: they are wrong. Time is an ocean
in a storm. You may wonder who I am, and why I say this. Sit down, and I will
tell you a tale like none of which you have ever heard.

290

Know first that I am the son of Shahraman, a mighty king of Persia. On our way
to Azad with a small company of men, we passed through India, where the
promise of honor and glory tempted my father into a grievous error.

As the prince introduces himself and his tale, the accompanying cinematic shows the
prince arriving in India with his father, backed by an army of warriors. As they lay siege
to the town, an old vizier points the prince in the direction of the maharajah’s treasure
vaults, and the prince embarks into the palace as the war-torn city crumbles in ruin
around him.
Do you think I felt regret as I gazed upon the destruction of the city, or at least
humility at the speed with which a world can be transformed from a good world
into a hell? If you think so, you are mistaken. For in that moment I thought of
one thing only: the honor and glory I would bring my father by fighting like a
warrior in my first battle.

The first complication we see, much as in Metal Gear Solid, is in the very nature of
cinematic cut-scenes—expository interludes during which the player provides no input, a
literal equivalent to watching a movie. Yet here, the effect is furthered by the prince’s
voiceover.
While the player controls the title character during gameplay, much of the story is
narrated by the prince himself, as here, where he introduces himself, tells his audience
to sit down and hear his tale, even challenges their assumptions over his motives and
mindset. If embodied in the prince, how can the player explain witnessing these events
outside of their control, much less being told the story that they are playing by their own
avatar? It becomes impossible to imagine the prince as a mere extension of the player,
as we must account for two different versions of the prince: one in the diegetic present
(the avatar) and another narrating after-the-fact. Resultantly, the player must essentially

291
exist both outside the narrative and within the narrative simultaneously, watching and
experiencing events simultaneously.
273

Compounding the effect in both Prince of Persia and Metal Gear Solid, such
narrative passages also develop the avatar as an autonomous personality distinct from
the player. The prince is given an elaborately detailed persona—he begins as a brash
and arrogant young man eager to impress his father and earn glory for both himself and
his kingdom, but evolves over the course of the game to become a more wizened,
responsible, and compassionate individual, developing a romance with the maharajah’s
daughter and struggling to make right his own mistakes. Solid Snake is just as
distinctive—he’s gruff, cynical, and defiant, yet genuine compassion often glints through
his hardened façade; he’s a loner, a survivalist, and a flirt with female operatives; he has
a lengthy backstory and a history with several supporting characters. These qualities
define both characters as very particular individuals—and clearly delineate them as
separate from the player. The personalities which have popularized many such
franchise characters—Mario’s happy-go-lucky demeanor, Sonic the Hedgehog’s bad-
boy attitude, Lara Croft’s aristocratic airs—all distinguish them as discrete characters
rather than merely the player’s embodiment in the gameworld. The moment that a
player thinks, “wow, the Solid Snake is pretty cool,” they are implicitly acknowledging
that Snake is someone other than themself. Further, every player recognizes that these
characters are in no way unique to their own experience of the game—every other
player who plays Metal Gear Solid is equally positioned as Snake, thus the avatar very

273
This is strengthened once we reach the end of the game, discovering that the prince has been
telling his tale to the maharajah’s daughter after he has used the Sands to turn back time to a
point before the game’s fateful events. As such, the entirety of the game has already occurred,
and as audience to the prince’s tale, we have been in the position of the maharajah’s daughter—
a mere listener—all the while. We will return to this issue in chapter four, revisiting a very similar
situation in Max Payne.

292
clearly extends outside the player’s individual experience and cannot be considered a
singular extension of themself.
To counteract these traditional storytelling conventions, many games have
attempted to integrate some manner of player activity into cinematic passages in order
to keep some sense of diegetic participation. Resident Evil 4 and Tomb Raider: Legend
both require the player to provide input during cut-scenes, affecting the scene’s outcome
and thus able to take some responsibility for the avatar’s actions (as Lara Croft dives
through a booby-trapped passageway, the player must hit a particular sequence of
buttons to dictate whether she passes through unhindered or perishes inside). Other
games, such as Deus Ex or Mass Effect, allow the player to choose from a limited range
of dialogue options, providing a clear sense of influence over conversation and
exposition. While offering varying degrees of effect, these evince a clear attempt to
counteract the distancing of conventional narration and exposition, bringing the player
back into the situation at hand.

Capacity for action: mediation vs. unbounded action
Despite Snake’s vast capacity to function independently during cut-scenes, Metal
Gear Solid goes to considerable lengths to disallow the reverse—the player’s ability to
function outside the avatar. Instead, the game justifies most of the player’s actions—
even those that frequently occur out-of-game in other titles—as occurring through
Snake, making very little of the player’s activity unmediated. Even features that are
frequently relegated to extra-diegetic menus—the save game function, hints and
instructions, expository information on weapons and equipment—are all integrated into
the story via the Codec radio, framed as the activity of various operatives aiding Snake
in his mission. Incorporating these actions into the diegetic world, the game allows the

293
avatar to mediate these options, preventing activity that would otherwise need to be
understood outside the avatar’s capacity. As such, the vast majority of the player’s
activity is filtered through Snake.
Capacity for action is the inverse of avatar autonomy; rather than the extent to
which the avatar can function outside of the player, this considers the player’s activity
outside of the avatar (we might think of this as “player autonomy”). What is the extent of
the player’s capacity for action, and how much agency is he or she given outside of
functioning through the avatar? Is the avatar the player’s only means of interacting
within the gameworld, or is the player given the ability to provide input free of the
avatar’s mediation? Further, can the player provide input outside of the diegetic context,
or even outside of the game proper (that which we would common-sensibly think of as
“playing the game”)?
In contrast to Solid, games like Resident Evil and Gran Turismo allow the player
to perform actions in extra-diegetic menus which, while perhaps metaphorically
indicating actions on the part of the avatar, seem to occur independent of that character.
Resident Evil’s player can pause the game at any time to combine items, reload their
weapon, or review exposition, while all of the non-racing operations of Gran Turismo
occur without any clear indication of an avatar’s activity. In both instances, the diegesis
is paused, essentially put on hold, while we manipulate objects or select upgrades in a
spatio-temporal limbo; while these actions do have consequences in the gameworld,
there’s no immediate evidence of their being performed by a character. Similarly, the
mayor embodiment of SimCity fails for precisely this reason—the player’s capacity for
action far exceeds what we could reasonably credit to that character. The vast range of
civic responsibilities is problematic enough; the player’s god-like abilities to terraform the
land and wield literal acts of god only furthers the point. Resultantly, we cannot plausibly

294
justify the player’s activity through the avatar, and must instead look to their literal
activity as a player.
Perhaps most indicative of the player’s capacity outside the avatar, however, are
those activities that occur apart from “gameplay” altogether. Cheat codes have long
been a component of the medium, often offering the most reasonable means of solution
to frustrated gamers and demonstrating player action that occurs apart from gameplay
proper—the sort of activity demonstrated in chapter two with Contra’s Konami Code for
attaining thirty lives. Most notable about this device is that the Konami Code has
become synonymous with most players’ remembrance of Contra—the cheat seems part-
and-parcel to the game experience. Innumerable games now contain such codes, some
deliberately meant for the general audience, others intended merely as tools for
developer testing (but inevitably discovered by the public at large). In any such case,
the player resorts to activity outside of what we would legitimately consider “playing the
game” to produce results inside the game.
The capacity for out-of-game action continues to grow as a component part of
the gaming experience, and now even occurs beyond the videogame altogether.
Players of Civilization III can use the game’s accompanying map editor—a program
separate from the game itself—to terraform landscapes and build a planet to their
precise specifications before ever beginning the game, and The Sims and Spore both
offer similar programs for editing objects, characters, or features of the game. The Sims
and SimCity both allow players to download content online which can be easily imported
into the game; if the SimCity mayor wants to build Stonehenge, The Parthenon, or the
Capitol Records Building as city landmarks, they’re out of luck—the game does not
include these attractions. Yet by going outside of the game, the player can download
those landmarks from the SimCity homepage and quickly import them into the game.

295
Similar is the growing tradition of mods and hacks—changing the very fabric of the game
itself, as demonstrated earlier with the Doom 3 flashlight hack. In any of these cases—
whether part of the design or not—players go outside the game to effect the gameworld,
stepping out of their embodiment as an avatar and utilizing their power over the game as
a player outside the diegetic world.

Sensory perspective: subjectivity vs. objectivity
Metal Gear Solid’s game engine immediately demonstrates its hesitation
between internal and external positioning via the simple matter of visual perpsective.
Solid provides the player with an overhead view of the gameworld—a fairly detached
and objective point of view, particularly when compared with the deep subjectivity of the
avatar’s POV. Rather than seeing the world through Snake’s eyes, we are given an
externalized, “god’s eye” view, placing the player as an observer outside of the game
looking down on Snake and the surrounding world. Still, this is not to say that the
player’s perspective is entirely detached from the avatar. The camera does follow
Snake’s movements so that Snake is always centered on the screen, and as such, the
camera retains some degree of subjectivity by being motivated by the avatar’s
movements and limiting the player’s range of vision to the immediate vicinity surrounding
him. The game also provides Snake’s POV in isolated instances, particularly when
crawling through ventilation ducts or aiming a sniper rifle, further adding a degree of
internal perspective. And even while the player’s visual perspective is usually distanced
from Snake, they still feel much of his experience. Using the Dual Shock controller’s
rumble feature, the game allows the player to feel the vibrations of explosions, the
shocks on an interrogator’s torture, and the pounding of Snake’s heart when an alarm is

296
tripped. The player’s sensory experience is thus neither internal nor external, but
somewhere in between.
Clarifications as to what we mean by “perspective” or “point of view” can quickly
turn into a muddled and convoluted discussion, thanks to the disparate meanings that
both terms often carry—trying to clarify the issue, Gerard Genette makes nuanced
distinctions between such interrelated concepts as “mood”, “perspective”, “distance”,
“voice”, and “focalization”,
274
while Edward Branigan finds no less than eight different
“levels of narration.”
275
For the purposes of this methodology, I’ve distinguished three
primary manners by which the player receives the game in relation to the avatar’s “point
of view”—sensory perspective, psychological alignment, and access to information.
With this category of sensory perspective, we begin by considering the manner
by which the player apprehends the gamespace through sight, sound, and haptics. Is
the player provided the avatar’s first-hand perspective from within the gamespace (as
with Bioshock), an omniscient, god-like point-of-view from above or outside (as in The
Sims), or some other perspective that is in-between (as with Metal Gear Solid)? Does
the soundtrack suggest the avatar’s spatiotemporal position, or is the player given audio
cues to suggest a more removed perspective? Are haptics or other sensory data used
to simulate the avatar’s experience?
Please note: The terms “first-person” and “third-person” have become the
colloquial terms for distinguishing visual perspective, yet I’ve made a deliberate attempt
to avoid this terminology. As I’ll discuss further in chapter four, the application of these
terms to describe visual perspective is highly problematic—largely because the terms
are not meant to describe visual perspective at all—so rather than get bogged down in

274
Gerard Genette, Narrative Discourse: An Essay in Method, trans. Jane E. Lewin (Ithaca:
Cornell University Press, 1980).
275
Branigan, 87.

297
troublesome terminology, I’ll simply distinguish a subjective point of view through the
eyes and ears of the avatar from an objective camera and soundtrack sourced outside of
character (a more appropriate pairing would actually be “first-hand” and “second-hand,”
as these more accurately distinguish a personalized experience from a removed
perspective).
Of course, as Solid demonstrates, even these terms would fail us. While we can
feasibly locate an internal, first-hand perspective placed behind the very eyes of the
avatar on one end of the continuum and an objective, god-like POV completely detached
from any character on the other, there are many shades of subjectivity in between.
Starting from the most subjective and moving to the most objective, I might posit the
following off-the-cuff scale of POVs:
• the internal POV of Bioshock
• the over-the-shoulder perspective of Gears of War
• the trailing perspective of Ratchet & Clank
• the panning perspective of Metal Gear Solid
• the fixed-angle perspective of Resident Evil
• the side-scrolling perspective of Super Mario Brothers
• the fixed overhead god-view of Civilization
• the free-roaming god-view of The Sims

Resident Evil and Metal Gear Solid both provide three-dimensional, external
perspectives, yet while Resident Evil’s camera is based on fixed, static positions, rather
than being motivated by the avatar’s movement (aside from their leaving the frame
cueing a change in shot), Solid’s camera remains centered on the avatar’s position,
making this POV more closely attached to the avatar. Gears of War’s over-the-shoulder
perspective is technically outside the avatar, yet this POV is so subjective that Mark S.
Meadows refers to it as a “second-person” camera—not for the “you “ that second-
person typically implies, but rather to indicate that its subjectivity lies somewhere

298
between “first-“ and “third-person.”
276
Many other games’ points of view could easily be
added between the few I’ve chosen, yet this provides a small sampling of the variety
available.
However, literal points of view cannot be our sole concern. Consider that many
games do not lock the player into a single POV, but instead offer a choice of visual
perspectives. Hitman: Codename 47 allows the player to choose an over-the-shoulder
vantage as well as the avatar’s POV, yet these do not drastically affect the player’s
relationship to the character; the choice is largely a preference for whichever makes it
easier to interact with the gameworld. After all, many argue that an over-the-shoulder
POV is actually much truer to the avatar’s perspective than an internal camera, as the
latter creates a significant case of tunnel vision. Racing games like Gran Turismo
similarly offer the choice between a first-person driver’s seat view or a third-person view
from behind the vehicle, yet again, this doesn’t make the player feel that in one they are
the driver and in the other just a spectator. Further, many games require the player to
alternate visual perspectives at different points within the game, as when Metal Gear
Solid’s player switches from an overhead perspective to Snake’s POV to operate a
sniper rifle (or inversely, when the player of Halo 3 must switch from the Master Chief’s
POV to an over-the-shoulder to operate a flamethrower or gattling gun). Yet once again,
the shifting POVs do not drastically alter the player’s relationship to the avatar, as they
are primarily utilitarian in nature—each mode offers the most practical vantage point for
the situation at hand.
While we instinctively privilege the information that we take in through our eyes,
we shouldn’t neglect our other senses in this regard. Sound can be just as effective, as

276
Mark S. Meadows, Pause and Effect: The Art of Interactive Narrative (Berkeley: New Riders
Press, 2002), 75. Meadows does, in fact, invoke the “you” distinction here, yet his explanation in
this regard is quite sketchy. The discourse around his use of the term seems to rest much more
heavily on its lying “between” first- and third-person.

299
a continuous, subjective soundtrack may immerse us in the avatar’s position even when
our eyes tell us otherwise. When driving in Grand Theft Auto III, for example, the
game’s camera trails well behind the vehicle, leaving the player a fair distance behind
their avatar, yet the car radio’s domination of the soundtrack creates a counterbalancing
effect of subjectivity, aurally positioning the player inside the car, despite the visual
perspective indicating otherwise. And as Metal Gear Solid demonstrates, haptics can
serve a similar function, providing sensations that simulate the avatar’s own tactile
experience, regardless of visual point of view.

Psychological alignment: empathy vs. antipathy
Despite the embodiment complications wrapped up in cut-scenes and similar
exposition, one cannot overlook their ability to create a sympathetic protagonist with
whom the audience feels an intimate bond—after all, such was among the original
intents of these Hollywood conventions. We distinguished traditional narrative
“identification” with characters—the sort practiced by Hollywood cinema—as being a
matter of empathy, something different than the more literal identification players feel
toward an avatar. Yet that is not to say that such narrative empathy cannot still be
utilized as a means of fostering a stronger connection with those figures. Indeed, Snake
is a very sympathetic character, thanks in large part to the elaborate narrative
development provided by these episodes. The lengthy cinematics and extensive use of
character-developing radio conversations present Snake as an attractive and likeable
character, positioning him as the hero of the story at all times, encouraging the player to
become psychologically aligned with that character in much the same way that
conventional cinema fosters a connection with its own protagonists. Thus not only does
the game position the player as Solid Snake as a simple matter of mechanics (that is,

300
whether you like it or not, that’s the character in whom you are embodied), but it also
makes the player want to be Snake.
277

A very different manner of perspective, psychological alignment entails a more
intangible sense of how we are allied with our avatar: How “close” do we feel to our
avatar? How emotionally invested are we in that character? Are our own goals and
concerns consistent with his or hers, or do we feel disconnected from that character in a
metaphysical or spiritual sense? We might call this the player’s “emotional distance”
from the avatar.
Despite the numerous other cues that distinguish Solid Snake, Lara Croft, and
the Master Chief as separate from the player, the heroes of Metal Gear Solid, Tomb
Raider and Halo still remain highly sympathetic characters. We like them. And as a
result, we want to identify with those figures, and allow ourselves to feel complicit in their
actions by jumping into their shoes. In much the same way that the personas of
Hollywood matinee idols foster powerful connections with their legions of fans, such
featured characters—the celebrities of their own medium—equally foster an intimate
bond with the player, pulling them further into the diegetic world. It is for this reason that
the vast majority of videogame franchises built on recognizable avatars utilize
conventional techniques of character development, encouraging players to become
psychologically aligned with those figures beyond a mere embodiment during gameplay.
Establishing a narrative context and making our goals equivalent to theirs—assuring that

277
Indeed, fans initially expressed vehement disappointment upon learning that they must spend
a large portion of Metal Gear Solid 2 in the shoes of another avatar, largely because they all
wanted to be Snake once again; much of the sequel’s attractiveness lain in the promise of that
very opportunity. Snake still remained as a significant character in that story, despite the player
occupying another avatar, yet his presence alone was hardly enough to satisfy fans wanting to be
the character, demonstrating the draw to identify with Snake. When the game was re-released
with additional features a year later (as Metal Gear Solid 2: Substance), one of the big selling
points for the new edition was its allowing the player to complete the entire game as Snake,
finally satisfying this desire.

301
the avatar’s wants, needs, and desires are equal to those of the player—only furthers
this psychic connection.
278

Yet not all videogames deploy psychological alignment in so straightforward a
manner, as Halo 2 demonstrates in its alternation between two avatars on opposing
sides of a conflict. When first thrown into the shoes of the Arbiter, a member of the
Covenant army at war with the humans, the player of Halo 2 most likely feels very little
alignment with that figure, despite occupying an internal perspective from behind his
eyes. After all, the Master Chief is the hero of the Halo franchise, and every fan of the
first game was eagerly anticipating the opportunity to reprise the role of that figure.
Further, the series is still the Master Chief’s story, despite this change in point of view,
and if the Master Chief is the franchise hero, how is the player meant to relate to a
character on the opposite side of the war, whose goals seem completely opposed to the
hero? Metal Gear Solid’s sequel caused a similar antipathy, though for very different
reasons, thanks to an avatar that replaces Snake shortly into the game. Fans of the first
game came to the sequel eager to reprise the role of Snake, thus when begrudgingly
forced to play a new, unknown character—one whose effeminate and insecure
personality seemed at drastic odd’s with Snake’s popularly gruff demeanor—they
resultantly expressed varying degrees of antagonism toward their own avatar.
279

278
As such, many of the techniques previously discussed as moving the player toward the
external end of the spectrum might actually be seen as functioning in the opposite direction in this
light. Such examples demonstrate the need for a pragmatic examination of these issues within
any given text, as the effect of such characteristics may differ significantly from game to game,
depending on their deployment, and may even foster different positioning effects within the same
game.
279
Antipathy may also result simply as a matter of individual personality and tastes. Though
vastly popular with many players, a similar discrepancy might arise in the Grand Theft Auto
franchise, notorious for its violent, racist, sexist, homophobic, and otherwise generally offensive
content. While the player is certainly positioned as each game’s protagonist—a criminal working
his way up the underworld ladder via a series of violent and illegal activities—it is not unlikely for
some players to feel alienated or uncomfortable during missions that might offend their own
sensibilities, thus breaking their alignment with the protagonist performing those actions. Though

302
Conversely, SimCity affords a very external embodiment and point of view, yet
attempts a much higher degree of internal alignment. Despite being offered god-like
omnipotence with only the most cursory suggestion of an embodiment, the player is still
addressed as a diegetic member of the game’s populace through their role as mayor,
and encouraged to identify with the flurry of citizens walking the streets of their city. The
mayor might spontaneously bulldoze ten city blocks, lay down a freeway, and install a
museum without so much as breaking a sweat, yet all the while, they are addressed by
city advisers and concerned locals and encouraged to align themself with the individual
citizens that can be seen going to work, enjoying the city zoo, mourning at the local
cemetery, or protesting the lack of health care. As such, the player is encouraged to
psychologically align themself with the diegetic populace and feel like a member of the
community, despite all the evidence that suggests otherwise.

Access to information: focalized vs. unfocalized
During the early stages of Metal Gear Solid, after penetrating the Shadow Moses
compound, the player navigates Snake into a prison cell holding the hostage DARPA
chief, a character from whom Snake has been sent to collect information. Upon entering
the cell, a lengthy cinematic ensues in which Snake interrogates the DARPA chief,
revealing a slew of exposition is revealed, and the chief abruptly succumbs to a sudden
and inexplicable death. Of course, while the player surrenders control of Snake during
the cut-scene, they still recognize Snake as their embodiment in the story and the
perspective with which they are aligned, to one degree or another. However, as Snake
and the DARPA chief converse, the camera pans through the cell’s wall to reveal a

an admitted fan of the series, I have often found myself in that very position, sighing, “but I don’t
want to kill the innocent bystander!” and resultantly feeling quite detached from the avatar that
happily does so.

303
woman eavesdropping on their conversation from the cell next door. Snake is unaware
of her presence, with no means of knowing he’s being overheard, yet the player is given
this narrative privilege, allowing their knowledge to surpass that of Snake—an
occurrence unexplainable if only identifying with the protagonist.
While psychological alignment was concerned with our empathic connection to
character, narrative access is more an issue of information: What info are we provided,
and what is the means of such provision within the game? How does the game justify
the information to which we have access? Is our level of information consistent with that
provided our avatar, or do we know more or less than they do? Such questions are the
very things at stake when narratologists refer to the concept of “focalization”, a matter
several have addressed using a three-tiered typology.
Gerard Genette popularized the term focalization—which roughly equates to
“focus of narration”—by distinguishing between zero focalization (or nonfocalized
narrative), internal focalization, and external focalization, while Tzvetan Todorov
provides three similar tiers via simple formulas: Narrator > Character, Narrator =
Character, and Narrator < Character. In either case, they distinguish, respectively, the
omniscient narrator who knows more than the character, the subjective narrator who
says only what a given character knows (particularly evident in first-person narration),
and the objective narrator who knows less than what the character knows. But these
need adapting for our purposes, in two significant ways.
Firstly: traditional narratives see a concern with authoring; their stories are told
through the eyes of a narrator, typically in the first- or third-person, with an emphasis on
who is telling the story. The narratee (they who receive the narrative) typically remains
an invisible figure with little bearing on the story. Gaming, on the other hand, is
specifically concerned with who is receiving the narrative, as the player and avatar so

304
easily bleed together, with the player often being addressed as the character themself.
Thus their stories are told through the eyes of the narratee. For while we might
distinguish between a “first-person shooter” and a “third-person” side-scroller, all games
are, to a great extent, addressing the player in the second-person, as “you.” Consider
that although Resident Evil takes a “third-person” visual perspective, the game-over
screen does not read “Jill Valentine died;” rather it informs the player, “you died.” Stories
are told, emphasis on the experience of who is telling the tale; games are played,
emphasis on the experience of who is receiving the tale. So while these typologies
address the narrator’s relation to the character, we must switch to the narratee if we are
to apply them to gaming.
Secondly: For this model, we’re not so much concerned with whether the
narratee knows more or less than the avatar as with whether there is a discrepancy
between the two figures’ perspectives. So it seems reasonable, for this application, to
merge the objective and omniscient, leaving us again with two positions: the internal and
the external.
The focalized perspective is thus characterized by the player’s knowledge
remaining equaivalent to the avatar’s, best evinced by the amnesia trope utilized by
numerous videogames. In providing a protagonist with very little memory of his past,
Bioshock easily resolves the gap of knowledge that typically exists between characters
and audiences. Awakening from a plane crash, afloat in the ocean with no idea of what
is going on or how they got there, the player’s knowledge (or lack thereof) is roughly
equivalent to that of the protagonist. With no options but to explore the small island
nearby, the player begins with the same logical impulses as the protagonist—survival
and discovery—allowing the player to easily align themself with the hero.

305
Yet even outside such overt devices, videogames can still keep the player’s
information on par with the avatar’s by at least keeping plot-relevant details equivalent
between the two. In the main, Metal Gear Solid takes considerable efforts to do just this,
revealing most of the game’s expository information through events witnessed by Snake,
rather than scenes occurring away from the protagonist. Over the course of the game,
Snake overhears an inordinate number of conversations through which we learn the
details of the terrorist plot, background info on characters, and other story-building
material. Much of the material is not vital to Snake’s mission so much as it is narrative
enhancement—details to enrich the story and explain causal motivation—and as such,
conventional narration might just as easily reveal such information away from the hero.
Yet to keep the narrative focalized through Snake, the player is only provided exposition
when it is received within earshot of the avatar, thus justifying their learning of that
material—even if it means an implausible number of instances in which Snake happens
to be in the right place at the right time.
280

Yet there are cracks in that intent, as the cinematic in the prison cell
demonstrates, as Snake is overheard from next door. Such instances often seem

280
The most overt example of this heavy-handed means of delivering exposition is a flashback
that occurs shortly after Snake leaves the DARPA chief’s cell, a scene in which we witness the
terrorist leaders discussing a hostage’s inadvertent death during a torturous interrogation,
revealing that the torturer failed to gain the launch codes the hostage possessed before he died.
The exposition is vital to the narrative, as we later learn that the hostage of which they are
speaking is actually the real DARPA chief; the man whom Snake encountered in the cell was an
imposter. However, as this scene occurred earlier in the story, away from the hero, the game’s
designers had little means of allowing Snake to witness the event. For the sake of the plot,
Snake does not actually need to hear this information—it only serves to help make sense of what
happened in retrospect—yet rather than reveal the scene away from Snake’s presence, the
game’s designers choose to give it to Snake in a sudden psychic flash. Snake grabs his head in
pain as the scene is played in a distorted hallucinatory fashion, suggesting that the events the
player is witnessing are somehow being beamed directly into Snake’s brain. Soon afterward,
when Snake discusses the incident over the radio, Campbell reasons that the experience must
have been “psychometric feedback” from the terrorist’s resident telepath, Psycho Mantis—a
rather far-fetched, if not conveniently tidy, means of exposition. Yet regardless of its heavy-
handedness, the episode allows the player to justify why they’ve been privy to this information,
despite its having occurred away from Snake’s presence, before his arrival at the base.

306
unintentional—the designers are merely falling back on conventional cinematic
practices, and likely have no deliberate intention of placing the player in a position of
narrative privilege over their avatar. Being familiar with such conventions, players are
equally likely to pay no conscious heed of the slip. Yet this is not to say that they don’t
affect their relationship to the avatar nonetheless, as they still create a problem of
information.
Such moments are instances of suspense, a narrative technique that relies on a
separation between character and audience. Alfred Hitchcock—the cinematic master of
suspense—explains this by distinguishing suspense from surprise, using the example of
a bomb under a table: If a character in a scene sits at a table and a bomb suddenly goes
off, this is surprise—we are just as unaware as the character, shocked at what has just
occurred. However if that same character sits at that same table, and the camera
reveals a bomb beneath it, this creates suspense—we wait in anxious anticipation to see
whether the bomb will go off, whether the character will escape, and yearn to warn them
of their impending fate.
281
Suspense relies on an estrangement between audience and
character, on having a capacity of knowledge separate from the protagonist, as this
anxiety is contingent upon the audience knowing something that the character does not.
Equally, when the player is shown information that Snake does not have, it creates
suspense—someone is eavesdropping on our spy hero (and after all, a spy’s success
relies on his remaining undetected), and we anxiously wait to see what happens as a
result. A similar situation arises anytime that the player is provided information occurring
away from their avatar.
Better evidence of unfocalized narration occurs in our Halo 2 example. When I
step into the role of the Arbiter, I still know everything that I witnessed as the Master

281
François Truffaut, Hitchcock (London: Paladin-Granada Publishing, 1978) 79-80.

307
Chief even though the Arbiter does not have access to that information. When I go back
to being the Master Chief, I equally know what’s going on within the ranks of the
Covenant army, despite the Master Chief’s inability to possess that knowledge. In both
cases, I’m moved to a position of narrative privilege beyond that of my avatar, distanced
from that figure by virtue of knowing more than he does. A similar estrangement
becomes especially evident when replaying a videogame. If I play Bioshock a second
time over, my narrative access is suddenly much greater, as I know every twist and turn
the story will take and can anticipate all that will happen. I now know that Jack did not
coincidentally wind up in the city of Rapture, that he (or should I say “I”?) is actually a
“programmed” assassin, so while the avatar begins with complete amnesia, I’m now able
to fill all the gaps in his knowledge. Resultantly, my relationship to Jack—and to the
diegetic events of the game—is significantly different the second time through the story,
occurring form a much more external perspective.
Conversely, it’s also possible for the avatar to know more than the player, also
differentiating the two figures. In Metal Gear Solid 2, many elusive references are made
to the past of Raiden, the avatar that replaces Snake, with Raiden often refusing to
discuss his history. Finally, late in the game, he reveals his violence-riddled background
as a child soldier during the first Liberian civil war, details which become vital to the
game’s plot. Yet if the player is expected to be Raiden, how could they not have known
these details? Again, our narrative access is inconsistent with the knowledge of the
avatar.

Coherence: representation vs. abstraction
Coherence deals with the type of world presented by the game, and that world’s
comprehensibility in relationship to our own. Is the game meant to produce or resemble

308
a coherent, explicable world, or is the game abstracted from our understanding of time,
space, person, and action? Is the gameworld composed of comprehensible signifiers
easily linked to tangible signifieds, or do game elements resist association with standard
archetypes and ideas? Is the space of play one which players can intelligibly imagine
themselves to occupy, or does it does it demand an entirely new and unique sense of
meaning, defying comprehension as a plausible virtual space?
Keep in mind that when we discuss “representation”, it is not to be confused with
“realism”—the issue is not whether signifiers connect with a realistic or actual
component of our world. Stories about vampires, time-travel, interstellar spaceships,
and flying superheroes are all representational, despite their unreality. Nor is this a
matter of “believability” or “the suspension of disbelief”—that issue will be taken up
below. Instead, coherence is concerned with the degree to which signifiers are
understandable using the logic and paradigms of familiar experience.
Despite Metal Gear Solid’s over-the-top twists of plot, Bioshock’s science fiction
premise, and Grand Theft Auto’s cartoonish engagement with crime film clichés, all
present representational worlds to one degree or another—worlds that find easy
correspondence with intelligible notions of time, space, and logic. We comprehend their
fictions because they are rooted in a world similar to our own, containing human
characters, plausible action and motivation, and obeying some variation on the laws of
physics and causality with which we are familiar (aside from where their fiction
deliberately departs). Players can easily immerse themselves into such contexts, then—
and into the position of their avatar—thanks to that context’s understandability as a
coherent and explicable world.
Furthest away from such examples are abstract games like Tetris, Bejeweled, or
Pong, games that may seem irrelevant to the question of being positioned “inside” or

309
“outside” the game, of being embodied as a character or an observer. In Pong, we’re
merely playing the game, right? There’s no diegetic world to be inside or outside of!
Indeed, Tetris, Pong, and Bejeweled offer little, if anything, that is consistent with our
understanding of the world. Tetris and Pong loosely depict the logic of gravity and
physics, and Bejeweled incorporates tangible real-world objects (jewels), yet even these
aspects are significantly abstracted. As a result, the games demand to be engaged by
the player on their own conceptual terms rather than experienced by a character inside a
diegetic space.
Yet if we are to distinguish these games as “something else”, a type outside this
methodology, then where do we draw the line between abstract and representational
games, between games with or without an avatar or a traditional diegesis? We might
find ourselves hard-pressed to call Pong’s paddle an avatar, or the empty screen a
diegetic space, but does Pong differ all that drastically from Atari’s Tennis? Aside from
some very minor aesthetic changes (the background now looks like the overhead view of
a tennis court), the gameplay is identical, yet Tennis is clearly representative of a sports
context, positioning the player in the role of a tennis pro within the diegetic space of a
tennis court. Through this minor shift, we are being moved a little further toward the
internal end of the continuum—which suggests that Pong must exist along that
continuum as well.
The moment we are provided a point of interaction within the text, we’ve been
given some embodiment (read: a representation of ourselves) in the game. Even the
simple arrow-pointer of a computer desktop helps us to conceptualize ourselves as
active within a spatial field. When playing Bejeweled—a game in which the player
moves jewels around a two-dimensional board in order to make rows of identical
stones—the player can always recognize their point of interaction according to which

310
jewel is highlighted, and navigates the space by moving that highlighting up, down, left,
or right to select other jewels. There’s certainly no character with which to identify, but
the player is still represented within a space.
Slightly more representational, Frogger still remains a fairly abstract game, yet
the overall context helps make the game even more intelligible via representational
concepts. Without the representational cues, the object of Frogger is simply to move an
object across one spatial field while avoiding other objects passing through it, then move
through a second spatial field while doing exactly the opposite—stepping only on the
moving objects, and avoiding “empty space.” Simple enough, perhaps, yet not very
easy to remember, nor very compelling. However by adding some representational
cues, the game becomes much more readily understandable—the avatar is a frog, the
first field is a highway crowded with speeding traffic, the second field a river crowded
with logs, lily pads, and alligators. Not only is the game easier to understand (we know
that traffic is dangerous, and that one can drown in water), but the player now has a
better idea of how they—as embodied in the avatar—fit into this world, making the game
(potentially) a more immersive, internal experience.
Such representational contexts demonstrate one of the most common techniques
for creating a comprehensible world: providing a familiar scenario that recasts game
components as understandable objects. It is for this reason that early videogames—
forced into abstraction by graphical limitations and lacking the memory space for
narrative exposition—so often placed existing narrative contexts around the game
situation in order to absorb the player, from sports scenarios (Football) to existing movie
franchises (Star Wars) to iconic characters in cartoonish situations (Pac-Man). The
narrative context is a means of explaining how the player fits into the diegetic world, of
explaining the situation in which the player is meant to participate; if the player is willing

311
to accept that context, then they are agreeing to imagine themself in a fictional situation
in much the same way that a cinematic audience agrees to suspend their disbelief and
accept the reality of a film. This trend has, of course, evolved into today’s very
elaborately narrative-driven games, employing text- or dialogue-based exposition, cut-
scenes, celebrity voice-actors, and many other storytelling techniques. And just as we
previously noted cinematics’ ability to foster psychological alignment with the player’s
avatar (despite their potentially opposite effect in regards to avatar autonomy), cut-
scenes can equally offer a compelling way to provide narrative exposition, explain the
game situation, or establish an intricately detailed narrative world—all methods for
immersing the player into the text.
Between these two extremes of representation and abstraction are games like
those of the Civilization franchise. In Civilization III, we saw a game built on countless
real world signifiers, from general concepts such as technology, politics, and history, to
specified names and places like Abraham Lincoln, the Roman Empire, and the Eiffel
Tower. Yet the game also abstracts these concepts: a single citizen is used to indicate a
populace of thousands, a single soldier to indicate and entire platoon; the past, present,
and future of human technology is embodied by 48 scientific discoveries; and Abraham
Lincoln of the Americans somehow lives for thousands of years, all the while
manipulating every aspect of his civilization’s development (quite similar to SimCity’s
mayor). We might say that such signifiers are “abstract representations”—or more to the
point, that the game is iconic (not only in its use of individual signifiers, but in the entire
context of gameplay). Somewhere between representation and abstraction, iconic
games use representations (Abraham Lincoln), yet those representations become
abstractions of other things (all of America’s leaders, a broader notion of government
and leadership, or simply America itself). Such games cannot be readily classified in

312
either camp—indeed, the danger of such a game lies in the ideological assumptions that
could be made via the desire to accept Civilization solely on its representational (yet very
reductive) merits. Instead, the only way to appropriately consider the system of meaning
created by such a game is through its use of an iconic context.
Similar to this iconic take on world history, all representational games abstract
concepts to one degree or another—that’s why they’re “games” rather than
“simulators”—while all abstract games engage some representational concept intelligible
to human experience. Bioshock and Metal Gear Solid present first-aid stations and food
rations (respectively) that instantaneously heal gunshot wounds and other injuries—
abstractions of the healing/sustaining power of those items—while conversely, Tetris
and Pong still manage to incorporate representational notions of the physical behavior of
objects, despite those games’ relative abstraction.

Verisimilitude: plausibility vs. tenuity
Verisimilitude—as addressed in chapter two—is the ability of a text to create
something that has the appearance of being true or real. This category is concerned
with the plausibility and dimensionality of the diegetic world, and its ability to make the
player “suspend their disbelief.” Of what consistency is the fabric of the gameworld? Is
it given the impression of being dimensional and complete, or is it merely a bounded
arena inside the screen? Are the game’s boundaries veiled in such a manner that the
world is given a plausible sense of continuity, or are its limits apparent and
unconcealed? Do events and actions follow a logic that the player is willing to accept as
“real,” or do they beg to be received on other terms? In short, does the game create (or
even attempt to create) a credible, self-sustaining internal reality?

313
The credibility lent to a game like Metal Gear Solid comes in large part from its
ability to veil its own boundaries and allow the player to psychologically “fill in” any gaps
that may be present. The gameworld is inherently limited—no programmer can create a
life-sized world that expands around the globe—yet by using corridor walls, locked
doorways, mountainsides, and other natural boundaries, Solid manages to present the
appearance of a complete world. In truth, there is no “world” outside of Shadow Moses;
there’s not even a complete compound, as a map of the game’s spaces would reveal
large pockets of empty void that are never explored, nor were ever designed by the
game’s programmers. Yet the depiction of the gameworld that does exist allows for the
verisimilar illusion of a larger world outside; when Snake first emerges from the water in
the underground loading dock, there may, in fact, be nothing below the dock, yet the
player is given every reason to believe there is. Equally, when the player comes upon a
locked door, there is no reason not to believe that a room exists on the other side,
regardless of whether or not one does. It is this intention toward verisimilitude that
results in the very common trope of videogames taking place on an island—a
surrounding coastline creates a convenient natural boundary.
Yet verisimilitude is not merely about space and spatial boundaries. It further
concerns the events and actions within that space and the text’s ability to veil the
underlying necessities of gameplay. The representational narrative contexts discussed
above become essential to this end by incorporating nonsensical elements into the
narrative artiface and allowing the diegetic world to retain plausibility, despite game
conventions that threaten to break the narrative mask, revealing the gameworld’s
artificiality and potentially pulling the player out of the game.
Power-ups provide a vivid example. Based on their abstract and cartoonish
design, many games of the 8- and 16-bit eras were forced to use some abstract icon,

314
often floating in mid-air, to represent a health or weapon power-up, from Pac-Man’s
power pills and Super Mario Bros.’ magic mushrooms to Castlevania’s floating hearts
and Contra’s letter icons (“M” for machine gun, “L” for laser rifle, etc.). Yet as games
have become more representational, they’ve equally demanded stronger plausibility;
Bioshock refigures its power-ups into tangible and comprehensible items: weapons and
ammunition, first-aid stations and food, and genetic modifications in keeping with its sci-fi
premise. Such instances utilize the game’s context in order to explain away the power-
up, an underlying trope of the medium that is functionally needed for playing the game,
yet doesn’t necessarily correspond with our experience of the real world (Metal Gear
Solid exists somewhere between—it uses a variety of real world items that are put in
logical locations, yet they appear as crates and ammunition boxes spinning in mid-air,
waiting for Snake to walk over them). The player’s interface has even received
verisimilar justification, as when Halo incorporates the player’s health and ammo meters
into the digital readout inside the avatar’s helmet, providing diegetic motivation for that
information. And in truth, disguising the player as an avatar, opponents as characters,
and the very objectives of the game as the narrative itself are the most blatant examples
of fictional plausibility. But perhaps most problematic has been the resurrection element
necessary between multiple lives and saved games. Over the years, games have found
a variety of means for incorporating this function into the diegetic world: The Metroid
series features Samus entering a high-tech pod in order to make a record of her travels,
thus saving her game. Resident Evil allows its characters to record their story on a
typewriter, while Parasite Eve has its heroine use a payphone to make a report of her
progress with police headquarters. For its part, Metal Gear Solid allows Snake to radio
an “operator in charge of communication data processing” who is responsible for logging
Snake’s “mission data.” In all four cases, the resurrection element isn’t quite addressed,

315
but the save function is incorporated into the narrative and rationally explained within the
diegetic context, thus glossing over its illogical nature and providing the player with
enough plausibility to allow them to suspend their disbelief.
Metal Gear Solid utilizes believable human motivation, a consistent chain of
causality, careful exposition, and a range of other storytelling devices, all of which aid in
creating the plausible world of that game. Resultantly, even the most far-fetched turns of
plot—a telepathic adversary, a cybernetic ninja cloaked in an invisibility suit, and
characters born of advanced genetic experimentation and cloning—become
understandable via the logic we’re given. These concepts may be fictional, but they are
hardly nonsensical. Further, the game relies on narrative formulas and plot devices
familiar to our culture—the highly cinematic opening and the game’s overt references to
Hollywood films from James Bond spy thrillers to John Carpenter’s Escape from New
York demonstrate its engagement with more traditional narrative media. Keep in mind
that what people will accept on screen is not always equivalent to what they will accept
“in the real world,” and as a result, even the improbable plot becomes acceptable in the
player’s mind thanks to its similarity to numerous suspense thrillers of its kind. Seymour
Chatman refers to the verisimilar as the “artistically probable,” estimating that an
audience thinks through events and fills in perceived gaps not only through personal
experience, but also through the norms established by previous texts and the norms
written by society and culture at large.
282
It is for such reasons that people can have a
thorough understanding of the logic of time-travel and vampirism—despite their fictitious
nature—and similarly, why players accept that a gunshot to the chest will only cause a
minor diminution of health, and that a first-aid kit (or a food ration, or a “health pack,” or a
mystically floating heart icon) will quickly heal the damage. Having seen such instances

282
Seymour Chatman, Story and Discourse, 50.

316
enough times to build an instinctive familiarity, players naturalize these conventions,
allowing them to follow the fiction’s “logic” without giving it a second thought.
In contrast, Pac-Man and Frogger do very little to offer verisimilitude. In its
abstraction, there is little logic to the space of Pac-Man—the “pellets” that the title
character gobbles up are never explained in its original arcade version, nor is the context
of his being chased by ghosts, and as such, the player is not given the proper tools with
which to “fill in” narrative gaps or interpret that world as a plausible reality. Frogger, on
the other hand, does offer a more plausible reality as a result of its representational
context—the player’s understanding of the dangers of traffic, drowning, and alligators all
play into their comprehension of the game. Yet Frogger still offers no explanation as to
why the avatar cannot move further up or down the highway or river than the borders of
the screen allow (a waterfall, for example, might better explain why drifting off screen is
a deadly occurrence), why Frogger must reach his lily pad within thirty seconds, nor why
Frogger’s death causes him to reappear on the far side of the highway.
By these examples, we can see an obvious correlation between coherence and
verisimilitude; representational games lend themselves better to verisimilar experiences,
while abstract games more readily reject such plausibility—yet that is not to say that the
two always go hand-in-hand. The peculiar world of Super Mario Bros.—a world that
demonstrates an eclectic merging of both representation and abstraction—can help us
to distinguish the representational from the verisimilar. The mustachioed plumber hero
and his quest to save Princess Toadstool from the evil Bowser is certainly intelligible
through an everyday understanding of the world. Further, most of the imagery invoked
in the game has a basis in real-world referents—bricks, pipes, flowers, turtles, beetles,
fish, cannons, fire, and quasi-dinosaurs are all understandable through typical human
experience. Fire can harm you. Pipes lead to other places. Turtles recoil into their

317
shells. Cannons fire deadly projectiles. All these objects and concepts find root in
representational meaning. Yet simultaneously, much of the game is nonsensical to
human experience: Bricks hang suspended in mid-air, and a head-butt from underneath
can often make a giant mushroom emerge from within. Said mushrooms cause the hero
to double in size. Giant bullets travel slow enough for Mario to jump on top of them.
Acquiring a flower allows Mario to throw fireballs—along with causing a spontaneous
change in costume. And for no logical reason, Mario cannot backtrack; the screen
scrolls left to right, but not vice versa, thus an invisible wall always follows behind the
hero. These all require an understanding of the game’s own internal logic beyond the
typical associations signified by its contents—the world outside the game would not help
us to understand these feats and actions, nor would the “textual logic” offered in the
narratives of popular culture. With verisimilitude often failing, then, the player must
comprehend this world “as a game,” and is thus moved further toward external
positioning than the largely representational context would otherwise suggest.

Formal awareness: transparency vs. reflexivity
Metal Gear Solid contains numerous memorable moments, yet by far its most
talked about episode is the player’s quirky encounter with Psycho Mantis, a telepathic
villain that Snake must defeat early in the game. As Mantis levitates about the room,
taunting the hero, he demonstrates his psychic capabilities by choosing an assortment of
comments based on how the player has played the game. For example:
“You are a highly skilled warrior, well suited to this stealth mission. However,
traps are your weak point. Hmmm. You have not saved often. You are
somewhat reckless. And you are neglecting to save this game as well.”

He further comments on other games that the player enjoys by accessing their
Playstation’s memory card: “Now I’ll read more deeply into your soul. Ah…I can see into

318
your mind. You like Castlevania, don't you?” He then continues by instructing the player
to place the game controller on the ground so that he can demonstrate his telekinetic
powers, insisting that he can make it “dance” with his mind—a feat achieved by
activating the controller’s vibrating rumble pak. All of these acknowledge the player as a
presence separate from Snake, a presence outside of the game, yet most striking is the
strategy necessary to defeat the villain. As a psychic, Mantis can predict Snake’s every
move and thus dodge any attack, making him nearly impossible to hit. To overcome
this, the player must unplug their Playstation controller from the player one port—thus
severing Mantis’ link to their mind—and reconnect it via the player two port, at which
point Mantis will incredulously question why he can no longer read Snake’s thoughts.
283

Despite Metal Gear Solid’s immense degree of coherence and verisimilitude, going to
great lengths to establish plausibility and coherence, Metal Gear Solid equally flaunts its
artificiality during these self-conscious moments that playfully reveal the text’s status as
“just a game.”
With formal awareness, we are largely discussing the player’s conscious
perception of the text itself and the videogame’s attempts to veil or foreground its own
rhetorical strategies: Is the game’s form made transparent, allowing a sense of direct
access to the gameworld, or does the form draw attention to itself, acknowledging its
own artificiality? To what degree does the player accept the game’s content as a self-
contained internal reality? Is the player absorbed into the experience of the game, or
are they made conscious of their role as a player outside the text?
Demonstrated by the seamless presentation of videogames like Halo, Bioshock,
or Oblivion, where all the rules, structures, and aesthetics of the game are effaced by its

283
If the player is unable to surmise this strategy and perishes at the hands of Mantis’ attacks
several times, Colonel Campbell will finally suggest this strategy over the radio, indicating that
Campbell is also aware that Snake has replayed this battle numerous times.

319
coherence and verisimilitude, formal transparency allows the game-text to elide its own
discourse—to allow the experience to feel “real” by not drawing attention to its mediated
nature. Transparency thus refers to the form’s ability to appear nonexistent. In such
cases, the form is actually quite the opposite—it is opaque, veiling the mechanics of
gameplay and obscuring the text’s underlying structures (rather than providing the clear
and unobstructed window on such devices that “transparency” might indicate). Yet by
veiling its own form, the game allows the player to ignore its constructed nature and any
consideration of its status as a videogame, feeling as though they are given transparent
and unmediated access to the game’s content—its diegetic world. “Unmediated” is the
key descriptor here. The player is able to forget the medium—to forget the game
console, controller, and TV screen/monitor; the game’s designers, producers, and
programmers; the game’s packaging, marketing, and mass media attention; and
everything else “videogame” entails—and feel as though they are being given direct
entrance to the gameworld. The player of Bioshock is not meant to feel as though they
are experiencing a piece of software running on an Xbox 360; they are meant to feel as
though they are simply exploring the city of Rapture. Of course, this experience does
occur through a piece of software and a videogame console, thus in order for the player
to attain that feeling, the game’s formal components must remain “transparent.”
284

284
This is yet another convention that videogames have inherited from previous narrative forms,
as briefly described in chapter two. The Classical Hollywood Cinema equally employed a
narrative style that was dependant on its own effacement—viewers were able to suspend their
disbelief and accept the “reality” of a film’s diegesis in large part because the film’s form was able
to erase itself through such devices as continuity editing and motivational causality, making the
screen seem a mere window on the world, and the events therein seem straightforward and
natural. By eliding such formal properties as editing, composition, and production design, the
spectator could allow “the frame to fall away” and simply accept the reality of the film’s content.
Videogames have learned the same tricks. For more on Classical Hollywood style, see David
Bordwell, Janet Staiger, and Kristen Thompson, The Classical Hollywood Cinema: Film Style &
Mode of Production to 1960, (New York: Columbia University Press, 1985).

320
Representational narrative contexts and verisimilitude are among the most vital
components to such transparency. After all, the functional components that
verisimilitude helps to veil, components like power-ups, saved games, and extra lives,
are the underlying structures of a videogame—the very form transparency is meant to
veil. By justifying such elements within the game’s diegetic context, verisimilitude helps
foster formal transparency. Unobtrusive interfaces and intuitive control schemes can
equally foster such transparency, allowing the player to “forget” the hardware, to not
spend conscious thought on the gamepad or the requisite buttons that need pressed,
thereby allowing the formal component of the hardware to “disappear.” The player can
then better merge their own consciousness with that of the avatar, allowing for the
thought process of “I better duck!” rather than “I better press the X button!” On the other
hand, the difficulty or clumsiness of an awkward control scheme will continually remind
the player of the game’s form.
285
In much the same way that cinema’s continuity editing
remains “invisible” when it works, and becomes apparent only when its conventions are
transgressed, intuitive control schemes remain transparent, and only draw attention to
themselves when they defy the player’s instincts. Other formal components function in
the same manner, veiling the artificiality of the game’s diegetic world by not drawing
explicit attention to itself.
Yet videogames don’t always necessarily want to veil their artificiality. From the
very beginnings of the medium, videogames have equally enjoyed acknowledging the
player and the gameplay situation, using reflexive devices that move the player outside
the text rather than pulling them in. Direct address offers a straightforward example—
Mario waving to the player at the end of a level, or Sonic impatiently tapping his foot and

285
And there are some deliberate instances of this—when their avatar has been drugged by a
monster’s poison, The Legend of Zelda: Wind Waker reverses the player’s control scheme in
order to create a sense of disorientation, mimicking their avatar’s disposition through deliberately
unintuitive controls.

321
checking his watch when left standing still. Just as in cinema or literature, such
instances acknowledge the artificiality of the medium by “breaking the fourth wall,”
making the screen that separates the fictional world from the real world porous (and
these instances are doubly effective by virtue of the character in question being the
player’s own avatar). This effect becomes accelerated with the use of metatextual
content.
From tongue-in-cheek, self-aware humor to existential ruptures in the game’s
narrative fabric, metatextuality inherently moves the player outside the text by disrupting
immersion and acknowledging the text’s status as “just a game.” Sometimes such
material manifests as a mere passing joke: the protagonist of Space Quest, for example,
at one point finds an escape pod with a button marked “Don’t Touch” which, if pressed,
transports him into one of the publisher’s previous games, King’s Quest. The hero of the
Monkey Island series often pokes fun at heavy-handed twists of plot, and the Ratchet &
Clank and Jak and Daxter franchises frequently exchange self-conscious references to
one another. Yet in other games, self-consciousness becomes integral to the game’s
plot to such an extent that it defines part of the experience, making the player aware of
their own presence as a figure outside the diegetic space of the game. Such is the case
in Metal Gear Solid.
The Psycho Mantis episode loudly declares its status as “just a game” by
subverting formal transparency and instead breaking the fourth wall, drawing explicit
attention to the player’s existence behind the avatar of Snake—and several other
moments in Solid echo this reflexivity. In another instance, Snake must contact his ally
Meryl by Codec radio, yet is never given the proper frequency on which to reach her.
Observant players will notice, however, that a screenshot on the back of the game’s
casing reveals a radio conversation between Snake and Meryl—with the appropriate

322
Codec frequency clearly displayed on-screen. Colonel Campbell even instructs Snake
to do just this, saying that he cannot remember her frequency, but that “it should be on
the back of the CD case.” Aside from the obvious reflexivity of asking Snake to check
the packaging of the very game in which he exists, this also creates a particularly
unusual paradox, as the screenshot reveals a moment in the game after Snake has
discovered the correct frequency and radioed Meryl, yet this screenshot cannot happen
until Snake (or the player) has discovered the frequency by observing that very
screenshot, essentially witnessing an image of his own future. This occurrence is the
videogame equivalent of a movie character looking ahead in the script to find out what
happens next—a device popular in Mel Brooks spoofs, but exceedingly unusual in a
serious spy thriller.
In these examples, the game goes out of its way to draw attention to its own
artificiality, its status as a game. These occurrences cannot be explained within the
diegetic logic of the text; the player can only make sense of the experience by
acknowledging the gameplay situation—their extra-diegetic control over Snake and their
status as a player outside the text. Regardless of how representational or verisimilar the
game may be, such reflexive moments aim to move the player to a more externalized
position by acknowledging their existence behind the avatar. Such occurrences are
especially notable when considering that many games that employ such reflexivity
equally take great pains toward verisimilitude. While verisimilitude may have strong
correlations with formal awareness, these two components need nonetheless to be
considered discretely, as many games provide a plausible world while equally providing
a reflexive awareness of that world’s artificiality.

323
Pleasure: narrative satisfaction vs. mastery of the text
Lastly, we need consider the goals of the player, and the pleasures they take
from playing the game: Is the player concerned with progressing through the game’s
sequence of events, fulfilling the goals of the game’s protagonist, achieving the game’s
ideal ending, and realizing a satisfactory narrative? Or are they concerned with
achieving a high score, exploring every nook of the gameworld, finding the secret items,
and mastering the game text? Essentially, is the player enjoying the internal pleasures
of the avatar, or the external pleasures of the game itself?
The internal pleasures of a game are fairly similar to those of other narrative
media—accepting the reality of the text’s fiction and identifying with the protagonist in
order to immerse oneself in the text. In videogames, this manifests most deliberately in
role-playing—not only those titles specifically referred to as role-playing games, but any
game in which the player takes on the persona of the avatar, diminishing the line
between player and character. Yet this is also the pleasure at work anytime the player
takes concern with elements inside the story, with people and events that do not actually
exist in the player’s world. If the player takes legitimate concern with whether or not the
Master Chief defeats the Covenant or the Persian Prince undoes the havoc caused by
the Sands of Time—as opposed to simply “winning” or “finishing the game”—they are
engaging the internal pleasures of the game.
More unique to videogames, however, are the external pleasures of the text.
Certainly, some players of Super Mario Bros. identify with Mario and focus their efforts
on saving Princess Toadstool from the evil Bowser—engaging the internal pleasure of
narrative satisfaction. Yet many players could not care less about this rather
superfluous plot, concerned instead with conquering the game itself rather than
conquering Bowser. These players are concerned with progress and mastery, with the

324
external pleasures of mastering the very object that is the videogame. They might
attempt to progress through the game as quickly as possible, utilizing the game’s secret
warp zones and other beneficial maneuvers to advance to the end as quickly as
possible, thus demonstrating their deft gaming skills. Or they might try to accumulate as
high a score as possible—many early videogames presented point-accumulation as their
primary draw, and the prestige of entering one’s initials on the “high score” list of an
arcade game was a fundamental show of such mastery (with point-collection having
waned in popularity, a contemporary equivalent might be conquering a game on its next-
to-impossible “extreme” or “nightmare” difficulty setting). Or, players might attempt to
find every secret area of the game, every hidden object, and every quirky glitch and
exploit there is to be found.
A curious trope that began in 1980 with the release of Adventure for the Atari
2600, the inclusion of “Easter eggs”—secret extras hidden within the game—are
perhaps the best illustration of the external pleasures of a videogame. Quite famously,
Adventure’s designer, Warren Robinett, was upset over Atari’s policy of not crediting
game designers, and unbeknownst to anyone, placed his name in a secret room within
the game, a room that could only be found through a series of unlikely actions. When
the hidden room was finally discovered and word of its existence got around, its
enormous popularity eventually led to entire games being built around such secrets and
surprises.
Nearly all the titles in the Mario franchise, the Sonic franchise, and the Legend of
Zelda franchise are built around a vast range of secret rooms, objects, and levels that
encourage players to return to the game again and again in order to find them all. Other
games include secret codes which cannot be deduced from the game itself, but can only
be learned through word-of-mouth, game magazines, or internet message boards (much

325
like the Konami Code in Contra). The most recent wave of consoles has only
accelerated such pleasures with the advent of the Xbox 360’s achievement system and
the Playstation 3’s similar trophy system—both of which encourage players to earn
points by performing obscure and at times rigorous tasks in order to increase their
“gamerscore,” a running total that accompanies their online profile (the player of
Bioshock, for example, can earn the “Historian” achievement for finding every audio
diary, the “Toaster in the Tub” achievement for electrocuting an enemy in water, or the
“Seriously Good at This” achievement for completing the game on the hardest difficulty
setting).
Nowadays, one would be hard-pressed to find a game which did not include such
hidden secrets, as they’ve become an expectation and requisite inclusion. Grand Theft
Auto: San Andreas includes almost 100 “unlockables”, ranging from acquiring a stash of
weapons for spray-painting over 100 graffiti tags, to increasing your maximum health by
completing a series of ambulance missions, to receiving a pimp outfit for attaining a
100% relationship score with a potential girlfriend—there’s even a 100% completion
unlockable for finding all the other unlockables! And Metal Gear Solid offers such
unlockable and discoverable content as well. By completing the game with different
endings, the player can attain an “infinite ammo” bandana (allowing an endless supply of
ammunition whenever equipped), “optic camouflage” (essentially, an invisibility suit), and
a James Bond-like tuxedo, each to be placed in Snake’s inventory for future replays.
Also available are a series of “ghost pictures” which can be uncovered if the player
discovers a camera hidden in the game; by snapping photos of the proper places, the
player can discover an assortment of hidden images.
In all these cases, the unlockable and hidden secrets may entail some diegetic
benefit, however they cannot be properly explained in the game’s diegetic logic,

326
requiring some acknowledgement of the game’s artificiality. How do we explain Mario
jumping above the game screen in order to find a warp zone, or San Andreas’s CJ
inexplicably becoming fireproof because he completed a series of firetruck missions, or
Solid Snake gaining infinite ammunition from a bandana? Much like Space Quest’s self-
aware humor or Metal Gear Solid’s reflexive gimmicks, these hidden extras ultimately
treat the game in a metatextual manner, asking the player to engage not the gameworld,
but the game-text itself. The videogame becomes as an object to be mastered by the
player rather than a world to be conquered by the protagonist.
Yet this is only the case if the player chooses to concern themself with such
metatextual components. Metal Gear Solid’s player need pay no attention to the game’s
unlockable elements, and most likely remains unaware of their existence during their
initial playing of the game, as they are only available for replay. So while these items
must be considered within the game’s overall experience, they remain merely extra
features, secondary objects that add another layer of pleasure to the text. For despite
these features, the game still privileges its cinematic narrative above all else, indicating a
concern with narrative pleasure. If the player takes legitimate concern with whether or
not Snake saves Meryl or defeats the terrorists (as opposed to simply “winning” or
“finishing the game”), they are still engaging the internal pleasures of the game. As
such, this final category seems to offer the most flexibility within any particular game, as
the player can often choose which pleasures they wish to engage. I can play through
Grand Theft Auto: San Andreas by identifying with CJ, following his character’s story and
striving for narrative satisfaction, or I can conquer the text itself by scouring the game for
all its hidden secrets and striving for “100% completion.” Here, the game’s balance of
the internal and external is not entirely dependant on the text itself, often relying as much

327
on the player’s own preferences and style of play as the specific cues provided by the
game.

Ultimately, the player’s positioning in Metal Gear Solid—or indeed, in any given
game—will inevitably vary to some degree from player to player. I might be able to chart
out my own experience playing a wide range of games and present those as the
“average,” yet other players would undoubtedly dispute those claims on the grounds of
their own play sessions. Players can engage the internal and external qualities of the
game independently, in turns, or simultaneously, depending on their own preferences
and disposition. The player’s experience remains a dynamic and intangible quantity that
can hardly be gauged or catalogued.
However, we can account for the textual cues that influence the player’s
positioning. While neither my nor anyone else’s experience playing Metal Gear Solid
can be held up as an exemplar for the standard encounter with that game, we can note
that the excessive cut-scenes, copious amount of dialogue, and other efforts toward
developing the character of Solid Snake all encourage the player to recognize Snake as
a distinct individual (external positioning) while the overhead camera’s attachment to
Snake’s movements, his sympathetic treatment, and numerous perspectival cues all
support a fair degree of player identification with Snake (internal positioning). The
coherent narrative and great efforts toward verisimilitude, along with the player’s
receiving of information through Snake’s experience, creates an immersive, believable
world (internal positioning), yet the game’s numerous moments of reflexivity and self-
consciousness, as well as its metatextual unlockables, remind the player of that world’s
artificiality (external positioning). We may not be able to reach a fixed or finite account
for the “standard” game experience, but that is not to say that we cannot account for the

328
experiential cues the game provides. The above categories are designed to account for
just such cues.

The Ends of the Spectrum
Despite the game’s seeming contradictions, Metal Gear Solid has been received
as among the very best games of its time:
“From beginning to end, it comes closer to perfection than any other game in
PlayStation's action genre. Beautiful, engrossing, and innovative, it excels in
every conceivable category.”
286

“Konami's espionage game is a fantastic game, straight and simple. … It's the
kind of game the PlayStation is made for.”
287

“Solid is a gaming masterpiece that can safely join the elite ranks of the greatest
PlayStation titles of all time. … Metal Gear Solid elevates video gaming to high
art.”
288

“If you've ever wanted to see the very best of the Sony PlayStation, Metal Gear
Solid is your game.”
289

Continuously ranked among the informal lists of “The Greatest Games of All Time” and
having influencing not only its own vast franchise (sequels, prequels, spin-offs, and even
a remake) but the entire lineage of stealth-based games which have followed, Metal
Gear Solid is a touchstone in videogame history. Yet if the player’s positioning in the
game is as fragmented as its textual cues would suggest, how has it managed to
achieve such high regard? Why does this not create the schizophrenic experience that
one would assume from such a dynamic, and how does such a game still remain so

286
Randy Nelson, review of Metal Gear Solid, IGN, posted on October 21, 1998,
http://psx.ign.com/articles/150/150569p1.html.
287
Doug Perry, review of Metal Gear Solid, IGN, posted on October 21, 1998,
http://psx.ign.com/articles/150/150569p1.html.
288
MajorMike, review of Metal Gear Solid, GamePro.com,
http://www.gamepro.com/sony/psx/games/reviews/236.shtml (posting date undetermined,
accessed June 29, 2006).
289
Victor, review of Metal Gear Solid, The Electric Playground,
http://www.elecplay.com/reviews_article.php?article=485 (posting date undetermined, accessed
June 29, 2006).

329
engrossing? How can a videogame manage such a fluid and seamless experience
while wavering so drastically between these poles?
To answer these questions, we need look to some other games. Before
addressing a game like Metal Gear Solid, one which lies so squarely in the middle of the
positioning spectrum, let’s first consider a few games that lie at the far ends of that
gamut and seem the limit cases of this approach. By examining these games,
comparing their modes of embodiment and probing the manner in which the player
engages with the gameworld in each, the relationship between the two ends, and how
they manage to co-exist so harmoniously, becomes strikingly clear.
On one end of the continuum, we might consider Half-Life. Approaching the
ideal of complete immersion, Half-Life represents a particularly strong push toward a
fully internal embodiment in the avatar. To create this near-seamless experience, Half-
Life features a sprawlingly continuous, highly-detailed (for the graphic capabilities of its
time), three-dimensional world encountered entirely from the avatar’s point-of-view.
Though one would have to stop short of describing the experience as “virtual reality,” the
intention is similar: to present an impression of direct experience rather than one of
mediation. To accomplish this, the game attempts to elide its mediating body—the
avatar—by eliminating all but the most inherent elements of a character outside of the
player’s own consciousness. By using the internal point-of-view, the player is literally put
into the body of protagonist Gordon Freeman and receives the entire narrative through
his eyes, able to move freely about the diegetic world, looking in any direction they
choose, all from the vantage of the hero himself. Resultantly, the player never sees the
avatar as a character outside of their own presence—Freeman could very well be
anyone, allowing players to easily merge their own identity with his (though the name
does at least suggest a male gender). Further, Gordon Freeman exhibits no sentience

330
of his own—he never acts, moves, or speaks outside of the player’s control, and remains
completely static unless acted upon by the player. Cut-scenes and other cinematic
techniques are eliminated so that all narrative events are presented through Freeman’s
POV, and Freeman’s voice is never heard, as dialogue is written in such a way that non-
player characters address Freeman without needing or expecting a response,
eliminating the problem of hearing the avatar speak on its own. The game’s interface is
designed so that the heads-up display can be explained as part of Freeman’s HEV
(Hazardous Environment) suit, with the health and ammo meters on-screen justified as
the readouts inside the character’s helmet, eliminating any suggestion that the player is
receiving extra-diegetic information not available to the hero. Similarly, the game strives
for full verisimilitude by veiling game mechanics and other structural components behind
diegetically-justified devices, using the game’s context in order to explain away
underlying tropes of the medium that are functionally needed for playing the game, yet
don’t necessarily correspond with our experience of the real world; “power-ups,” for
example, get refigured as tangible real-world items—weapons, ammunition, and first-aid
stations—and those items get placed in logical places within the gamespace.
290
Lastly,
the game plays in real-time; there are no lapses in time and space between levels (save
for one instance in which Freeman is knocked unconscious), so the diegetic world
retains spatiotemporal continuity. Twenty hours playing the game is equivalent to twenty
hours of diegetic time, as experienced by Freeman, making the player’s experience of
the narrative equivalent to that of the avatar. The name “Gordon Freeman” essentially

290
The game’s manual takes pride in such verisimilitude, contrasting its approach against titles
like Metal Gear Solid due to the latter game’s employment of useable items that appear as metal
boxes floating above the ground, rhythmically rotating in mid-air: “You won’t find Half-Life’s
arsenal spinning around in space. Instead, you’ll find items in more realistic locations, such as on
gun racks, in the hands of your allies, or on the bodies of your dead enemies.” Half-Life Platinum
Collection Instruction Manual, (Los Angeles: Sierra Entertainment, Inc. and Valve, L.L.C., 2002),
19.

331
becomes the only hard-and-fast indicator of a character outside of the player,
encouraging very little distinction between these two figures.
By comparison, consider the approach of The Sims, where every character is
depicted as an autonomous being, with the player offered no avatarial presence within
the game whatsoever (not even in the largely metaphorical sense that we saw in
SimCity). Provided a family of Sims (simulated people) in a suburban household, the
player is charged with helping that family negotiate the trials and tribulations of everyday
life—finding a job, paying the bills, building relationships, getting married—however the
player is never put into the role of actually being one of the characters. While the player
can use the point-and-click interface to tell their Sims to perform a wide variety of actions
(make lunch, go to work, feed the baby), the Sims don’t always agree to do so, often
protesting directly to the player when not in the mood for such an action. Further, when
left to their own devices, the Sims will act of their own volition, in accordance with their
various wants and needs (if a Sim is hungry, they may decide to make lunch on their
own; if tired, they may go to bed or nap on the couch). The player is relegated to the
role of a god-like figure looking down upon the Sims’ world and influencing their lives;
one could describe the Sims as dolls or as pets, but certainly not as avatars.
The deity comparison is apt (hence the term “god game,” a moniker often
associated with The Sims and other world simulators of its kind). Provided an
omniscient point of view of the gamespace, the Sims player can navigate the diegetic
world effortlessly and without limitation—a free-roaming camera allows the player to
zoom out to view a household’s entire lot at one time, or to zoom in to a closeup on a
character’s face or a particular object, and the camera can equally be manipulated to
view the household from any angle at any moment. Further, The Sims allows the player
near-omnipotence over the gameworld. Without any specified avatar through whom to

332
act, The Sims allows its player relatively unbridled action—the player can pause, slow,
or speed up time; add, move, or delete objects on a whim; create new characters and
delete old ones; influence the behavior of multiple characters simultaneously; and
generally oversee the lives of all the Sims in the narrative universe.
So these two games—Half-Life and The Sims—would appear to exist at
complete opposite poles, one a seamlessly immersive, internal experience, the other a
wholly external experience, lacking any avatar whatsoever. Yet let us complicate this
comparison.
On its surface, 2001’s Black & White (as well as its sequel, 2005) appears
strikingly similar to The Sims. Like The Sims, Black & White positions its player as an
omnipotent and omnipresent observer looking down on the gameworld: an island
populated by a tribe of villagers who go about their business and remain autonomous
from the player, despite being influenced by the player’s hand. The player is similarly
able to manipulate the landscape and its populace by pointing and clicking with the
mouse icon, while the villagers go about farming the land, constructing homes, and
going about their daily lives. Like The Sims, the player is able to influence these
characters, yet none are used as avatars. The villagers also utilize direct address by
speaking directly to the player, in much the same fashion as disgruntled Sims
complaining over their hunger, loneliness, or boredom; these characters even hold
complete conversations and voice specific requests rather than employing the
abstracted “Simlish” spoken by Sims. Despite these similarities, however, Black & White
does provide the player with an avatarial presence: their god-like omnipotence is justified
by being positioning as a literal deity within the gameworld.
Unlike The Sims, the player is acknowledged as a legitimate presence within the
diegetic space. When villagers look to the sky and speak to the player, they reverently

333
address them as their god, praising their munificence or begging for mercy, depending
on whether the player chooses to be a benevolent or malicious god. The player’s point
of interaction with the gameworld—the mouse-pointer icon—appears as a hand which
can interact with the diegetic space by picking up villagers, moving boulders, uprooting
trees, or casting miracles through mystical gestures—a literal hand of god. The player is
even in competition with other deities on the island, tasked with recruiting the faith of as
many followers as possible as their influence spreads across “the five lands of Eden.”
Perhaps most intriguing is that, as a god, the player is offered a “pet” animal that
wanders the island and acts as a divine representative, a proxy who stands in the deity’s
stead—a literal avatar for our avatar! Thus while offering a similar dynamic to that of
The Sims—providing no tangible embodiment for the player and offering an omnipresent
POV over an autonomous populace—Black & White actually offers a twofold
embodiment: both an avatar (the deity) and an avatar for that avatar (the pet).
In truth, the pet functions in much the same manner as the villagers—a mostly
autonomous being operated by the game’s artificial intelligence. The player can mold
the pet’s behavior through praise and punishment, yet is offered little opportunity to
actually “inhabit” that character, thus for the most part, the player remains embodied in
the deity figure itself, apprehending the gameworld from the god’s perspective. But if
such is the case, Black & White would not be offering the external perspective of The
Sims, despite the two games’ drastic similarity. By justifying the player’s experience as
coming from the perspective of an avatarial presence—the deity protagonist—Black &
White would instead be exceedingly similar to Half-Life’s internalized gameplay, offering
the consistent and immersive point-of-view of an avatar rather than the externalized
perspective of an outside observer. The game provides the first-hand POV of the deity
avatar.

334
How might we rationalize this occurrence? As we’ve seen, both Half-Life and
The Sims provide markedly different experiences from a more traditional “avatar game”
like Metal Gear Solid—a game with a tangible and recognizable avatar in the spirit of the
original meaning of the word. Such avatar games utilize a figure that embodies an
overseeing presence (originally a deity, now a player) while also existing independently
of that presence, essentially combining both internal and external traits. Alternately,
games like Half-Life and The Sims stake their ground at the polar ends of the scale by
offering near-seamless experiences—one offering the extreme of internal positioning,
with the other just as faithful in its externalized gameplay. While Solid lies squarely in
the middle of the positioning spectrum, combining an amalgamation of internal and
external traits, Half-Life moves away by eliminating all traces of external gameplay, short
of only the most inherent traits of a player outside the gameworld; concurrently, The
Sims moves in the opposite direction, maximizing the game’s external traits by
eliminating any hint of an avatarial presence.
291
Yet as Black & White demonstrates,
these games seem to have departed in opposite directions and paradoxically met at the
end of their journey.
By this account, we might posit that while “internal games” like Half-Life and
“external games” like The Sims provide vastly different experiences from traditional
avatar games—those games which provide a tangible body that is both an embodiment
for the player and a character unto themselves—they are also highly similar to one
another.

291
While speaking to a class at the University of Southern California, Tim LeTourneau, senior
producer of The Sims 2, indicated that his one regret over the design of the game was its
inclusion of numerous scenarios that occur while the sims are at work, asking the player to make
an either/or decision on their behalf. LeTourneau stated that he regretted this feature because
The Sim 2 “is not an avatar game,” and thus players should never be put into the position of
making a decision from the character’s point of view.

335

In the above illustration, the dark shading represents the degree to which a distinct
avatar is present and noticeable—strongest in the center with games like Metal Gear
Solid, games that negotiate the two ends through a combination of internal and external
cues. Yet note that on both extreme ends, the presence of the avatar fades. In internal
games like Half-Life, this occurs as the player is placed inside the perspective of the
avatar itself, making that body all but disappear as it merges with the player’s own
consciousness. In external games like The Sims, it occurs as the player is moved to a
god-like position of omnipotence, detached from any tangible or explicable presence and
thus left “outside” the diegesis.
The blurring of the internal and the external provided by Black & White would
seem to support just such a theory, appearing extremely similar to the external Sims, but
addressing its player in a similar manner as the internal Half-Life. Indeed, if The Sims
were to be modified by merely addressing the player as a literal deity to its Sim
residents, the game would seem to make the same leap as Black & White. How would
we explain this occurrence? To sort the matter out, let us consider another pair of
games, also positioned at opposite ends of this scale.
Most striking about the design of Myst is that the player experiences the game
through their own eyes, as themself. Like Half-Life, the game is fashioned entirely from
an internal point-of-view, suggesting the player to be experiencing the tale first-hand,
without a mediating body. The key difference, however, is that Myst neglects to provide
any sense of an avatar. There is no description of a character that the player is meant to
inhabit within the diegetic space—no tangible body on-screen, as with Metal Gear

336
Solid’s Solid Snake, nor even an ambiguous avatar referenced by name, as with Half-
Life’s “Gordon Freeman.” Instead, the game seems to address the player in a fashion
similar to The Sims: simply as “the player.” In fact, the game’s manual repeatedly
emphasizes that the player should experience the game as themself: “The key to Myst is
to lose yourself in this fantastic virtual exploration, and act and react as if you were really
there.”
292
Resultantly, the player does not have to accept another identity; the first-
person/third-person confusion typical of most avatars—that is, the dual identification of
“that’s Solid Snake, but that’s also ‘me’”—is seemingly absent. The game’s manual
even establishes how the player got there:
You have just stumbled upon a most intriguing book, a book titled Myst. You
have no idea where it came from, who wrote it, or how old it is. Reading through
its pages provides you with only a superbly crafted description of an island world.
But it’s just a book, isn’t it?

As you reach the end of the book, you lay your hand on a page. Suddenly your
own world dissolves into blackness, replaced with the island world the pages
described. Now you’re here, wherever here is, with no option but to explore…
293

The game disc itself is positioned as the titular book, thus as the player sits down to
“read” Myst, the narrative picks up in the very next moment, just after the player has
been magically transported to the world within its pages. The user’s manual—a few brief
pages on how to move about the game space and operate the keyboard controls—
reveals no further information on the game’s story. Thus with no introductory exposition
provided, the player begins the game with no idea of their goals, of what they are
supposed to do, simply because the narrative of the game begins with that very
scenario. Further, the game does not present the player with situations that would seem
unfamiliar to the common individual. “Myst is real. And like real life, you won’t die every

292
Myst User’s Manual (Novato, Calif.: Broderbund Software Inc. and Cyan Inc., 1993).
293
Ibid.

337
five minutes. In fact, you probably won’t die at all.”
294
There are also no monsters
nipping at your heels, no weapons to wield, no timers or countdowns to beat. There is
simply a mystery to be solved, a story to be uncovered—a scenario with which most
gamers (or cinemagoers, or readers of fiction) are quite familiar. In this way, there is no
gap of knowledge or experience between the player and the avatar, no sense of knowing
more or less than the protagonist does. The world might seem strange, wondrous, or
mysterious, but this is the reaction one would expect from such an unusual situation.
The natural urge then becomes to explore the world.
There are actually two stories in Myst, a back story and a game story, a hallmark
of what J. C. Herz calls the “free exploration” game, where these two stories are
superimposed.
One is the sequence of events that happened in the past, which you can’t
change but is a very good story. The other is the sequence of events that
happens in the present (e.g., you are wandering around trying to solve puzzles),
which is a lousy story but highly interactive. As you hurdle obstacles in the
present, you can see a larger picture of the past.
295

The game space is actually a vast, abandoned world. There are no characters there.
There’s nothing happening there. Essentially, there is no story in the diegetic space.
But the player begins to unravel another story, a vast, intricate tale which seems to open
up entire worlds—one which is not happening, but rather has already happened—and
the illusion is then created of being a part of that story.
All narrative games have a pre-designed story, to some extent or another. Part
of this dual positioning is in being simultaneously inside the narrative, experiencing that
story, and outside the narrative, watching or discovering that story. Myst’s design makes
the external side of the spectrum a part of the game itself—the protagonist inside the
game (“you”) is actually outside yet another story, the story to be discovered. The

294
Ibid.
295
Herz, 150.

338
avatar’s story is simply one of exploring the island, solving puzzles, flipping switches,
and collecting items. As this is a rather boring tale, it shifts attention to the other story,
the story inside the story, and if the player is concerned with that narrative, then they are
again placed inside the gameworld. It’s all a sleight of hand to immerse the player in the
game. For in reality, the story of Myst is simply that of a player running about the
diegetic space and completing tasks.
In Tetris, on the other hand, the game itself becomes the task. Here, on the
external end of the spectrum, is a game with absolutely no narrative, thus no character
with which to identify. There is no semblance of an avatar, even in the most abstract
sense, as there is no context to position the player in the diegetic space. The game
situation is merely one of arranging blocks as they fall from the top of the screen,
attempting to complete horizontal rows so that the blocks disappear before the next
barrage fills the screen. Most puzzle games provide at least some marginal target of
identification—the player of Bejeweled, for example, can at least recognize the
highlighting that surrounds one jewel as their point of interaction inside the gamespace,
the specific spot from which they can move left, right, up, or down. Yet Tetris fails to
provide even this semblance of presence. The only possible target of identification is
each individual set of blocks, as these become the primary focus of the player’s attention
and the only element of this world that they can manipulate.
Jesper Juul notes that games with no avatar or specific point of identification
have appeared throughout the history of the computer game; many—such as Atari’s
Missile Command—have been extremely popular. Missile Command’s game screen is
primarily an open sky, with a strip of land creating a horizon at the bottom. As cities are
attacked by missiles, the player must intercept and destroy the incoming projectiles
using rockets from three missile batteries. The player thus does not see “themself” on

339
the screen, but merely the results of their actions. Yet even here, the player has
opportunities to recognize themself in the narrative context. As Juul notes, “it would be
possible to create a ‘job description’ for the player—a soldier controlling missiles: a
typical hero.”
296
The player is thus an off-screen presence, but still within the war-time
narrative. The player could alternately create an even more direct identification with the
missile batteries themselves, seen at the bottom of the screen, or imagine themself
within those batteries. But Tetris provides no such context, and further, it seems
relatively impossible to even construct any such context for the player. Marie-Laure
Ryan notes that Tetris “represents the lowest degree of narrativity, because ‘fitting
blocks of various shapes into slots as they fall from the top of the screen’ is hardly
interpretable as the pursuit of human interests in a concrete situation.”
297

However, despite the lack of any narrative framework within the game, Tetris still
creates a character: the player themself as the actant. The player becomes the
individual performing a task, and through direct involvement with the game construct, the
player creates their own narrative arc (of a sort). Some even claim this arc as having
legitimate dramatic content, as with Janet Murray’s explication of Tetris as a
metaphorical enactment of the overtasked lives of Americans.
298
Whether literal or
metaphorical, the story is not vicariously lived through an avatar, but rather undergone
directly. The player’s actual playing is thus the story; the narrative is a first-hand
experience of the player performing a task.
It would seem, then, that Tetris is actually quite similar to Myst, as Myst also
purports to provide the player direct access to the gameworld (as the character in the
game is simply “you”), and in both cases, the player themself is behind the interface,

296
Jesper Juul, “Games Telling Stories?: A brief note on games and narratives,” Game Studies,
vol. 1, issue 1 (July 2001), http://www.gamestudies.org/0101/juul-gts/.
297
Ryan, “Beyond Myth and Metaphor.”
298
Murray, 144, as addressed in chapter one.

340
completing tasks, accessing the diegetic space not through identification with an avatar,
but rather through their own direct point-of-view. It would seem then that the push
toward a completely immersive gaming experience—toward the internal end of the
spectrum—is merely a push to recreate the original gaming situation, a player directly
interacting with the task, with the game construct itself. If such were the case, we might
conceptualize the positioning spectrum not as a line…

…but rather as looping back upon itself.

But this is not quite the case. For the player in Myst does not, in fact, experience
the game as themself, though this may be the conceit. Instead, the player accesses the
diegetic world through a projected self, an ego ideal of sorts within the game. This
projected self is easily recognizable in “avatar games;” we see Solid Snake, Lara Croft,
or Super Mario. The medium’s dual-positioning comes from recognizing that avatar as
ourselves, and yet simultaneously as someone else. But even in those games that
employ an internalized point-of-view, the player is still utilizing a projected self. Just as
the player of Half-Life can envision Gordon Freeman running about the halls of the Black

341
Mesa Research Facility, just as the player of Black & White imagines a deity hovering
over the villagers of the island below, the player of Myst still conceptualizes someone
within the gamespace. The first-hand POV becomes the “you” avatar.
So while the “story” of Tetris finds the player directly performing a task, Myst has
the player controlling a projected self performing a task. It recreates the Tetris
relationship, but with a mediating body providing access to the diegetic world.

Or, we could simply say that “internal” games move the actant/task relationship one step
down the chain.

In Myst (and Half-Life), the “you” is within the diegetic world performing diegetic tasks; in
Tetris (and The Sims), the “you” is in the real world performing the task of the game. But

342
because all videogames have some trace of both, thus can never fully reach either end
of the spectrum, the player is always playing in both modes simultaneously.
The subtlety of this shift—the movement from you to “you”—helps explain the
ease with which videogames can slip between the internal and the external, confusing
address between the player and the avatar. It is for this reason that Colonel Campbell’s
tutorial remarks to Snake—“Hold down the R2 Button to enter Weapon Mode”—can play
without a hitch. Were the player to step back for a moment and consciously consider the
narrative context in relation to these instructions, they would certainly recognize the
illogical nature of this exchange, however the movement from player to avatar remains a
slight enough shift that the game can elide the difference between these two figures.
All four of these titles—Half-Life, The Sims, Myst, and Tetris—offer an immediacy
of some sort, purporting to provide some form of direct experience rather than the more
outwardly mediated experience of functioning through a tangible and recognizable
avatar. Half-Life and Myst offer (seemingly) direct access to the diegetic world inside the
game, via an avatar integrated into the player’s own consciousness; The Sims and Tetris
offer (seemingly) direct access to the game itself, unmediated by any explicable avatar.
At both ends of the spectrum, these videogames seem to promise direct experience and
unmediated access to content, and this parallels observations made by Jay David Bolter
and Richard Grusin in their book Remediation: Understanding New Media.
Bolter and Grusin argue that immediacy and hypermediacy are concurrent
concerns that run throughout all “new media,” and that have equally run throughout
previous media dating back at least to the Renaissance.
299
With immediacy, they are
addressing a transparency of the medium, seemingly offering direct access to content;

299
They also discuss their titular interest of “remediation” as a third concern, yet this is irrelevant
to our purposes here.

343
with hypermediacy, an opacity and privileging of the medium, essentially foregrounding
the form itself. Bolter and Grusin discuss these competing impulses in numerous
arenas—digital art, the internet, virtual reality, television, digital photography, ubiquitous
computing, and indeed, videogames—yet ultimately they determine that these two
impulses share the same end-goal:
Hypermedia and transparent media are opposite manifestations of the same
desire: the desire to get past the limits of representation and to achieve the real.
They are not striving for the real in any metaphysical sense. Instead, the real is
defined in terms of the viewer’s experience; it is that which would evoke an
immediate (and therefore authentic) emotional response.
300

They go on to note that hypermedia can be received as an authentic experience “not in
the sense that it corresponds to an external reality, but rather precisely because it does
not feel compelled to refer to anything beyond itself.”
301

This would seem quite appropriate to the videogame experience: a simultaneous
fascination with engaging the diegetic world of the game (transparency) and engaging
the game-text itself on its own terms (hypermediacy)—of being positioned as both
internal character and external player. Equally, Myst’s design and premise feign
unconditional transparency, particularly in its first-hand point of view and its “you”
protagonist, while in its complete abstraction, Tetris makes little attempt to refer to
anything beyond itself. Half-Life strives for immersive transparency in its seamlessly
verisimilar experience, while The Sims, although taking reference in the “real world,”
creates its own quirky and fantastic internal reality that exists on its own terms.
Meanwhile Metal Gear Solid, like the large majority of videogames, neatly provides for
both impulses. The immersive qualities of its cinematic presentation, narrative context,
and efforts toward verisimilitude all demonstrate its transparent qualities; the quirky

300
Bolter and Grusin, 53.
301
Ibid., 53-4.

344
playfulness of reflexivity and metatextual pleasures of the game’s “hidden secrets”, on
the other hand, demonstrates the joy of hypermediate engagement with game text itself.
With this chapter, I’ve proposed a dual-positioning model for videogames,
claiming that the videogame avatar affords the player a simultaneous positioning in two
places at once: as a protagonist inside the game-text, subjectively experiencing the
diegetic world as a participant, but also as a spectator outside the game-text, provided a
god-like point-of-view external to the diegesis—essentially combining the ideal of virtual
reality (complete and seamless immersion) with the voyeurism of cinema (being a
spectator looking through the screen). I’ve contended that all games exist between
these two poles, negotiating this dialectic, with no game engaging only one over the
other. The Metal Gear Solid example that opened this chapter demonstrates the
common technique of blurring these two positions—player and protagonist—by
acknowledging both figures, but simultaneously eliding the distinction between them.
The effectiveness of such blurring is demonstrated in games like Metal Gear
Solid and its sequels, which utilize strikingly reflexive moments—moving the player to
the external end—while still maintaining their immersive qualities. This franchise is
commonly cited for its engagingly cinematic method of storytelling, and is a perfect
example of a series lauded for “making you fell like you’re in a movie.” Meanwhile, Half-
Life 2—like its predecessor, a perfect example of a game striving for a completely
internal experience—actually ends on a strikingly reflexive note: amidst an exceedingly
climactic moment, the avatar’s first-hand POV suddenly freezes as the diegetic world
becomes paused, and a doorway of light opens in the middle of this image, allowing a
figure to step through and address the avatar/player directly, suggesting the story’s
entirety to have been a simulation (i.e., a game) of some sort.

345
I noted that we might conceptualize the positioning spectrum as circling back
upon itself rather than as a straight line. In truth, the proper shape would best be
described as a spiral—looping back from whence it came, but not quite landing in the
same place. As such, it’s no wonder that games like Black & White can so easily utilize
conventions from both ends, using the god-like external positioning of The Sims and the
internal first-person POV of Half-Life simultaneously. It’s no wonder that the internal
tasks within Myst function quite similarly to the external task of Tetris itself. And it’s no
wonder that games like Metal Gear Solid can so readily confuse address between player
and avatar. These are all achieved because the jump between internal and external is in
fact quite easy to mask (and we will explore the reasons for this even further in chapter
four).
This sleight of hand, this slippage between player and avatar, has become a
defining characteristic of the contemporary videogame and a vital component of the
medium as we know it. Film scholars have often noted that one of Classical Hollywood’s
defining characteristics, and equally one of its biggest achievements, was its ability to
erase its form, to elide the medium itself through formal transparency, thus allowing their
stories to seem “real” and unmediated. As we’ve seen, videogames have no problem
with mediation—this is the avatar’s very function—so the medium instead strives to
erase the gap between these two positions, to elide gaming’s dual-positioning in order to
allow the player to still feel unified. The magic of the medium has become its ability to
continually put the player in two positions at once—as a spectator outside the diegetic
world as well as a participant within it—but to erase any distinction between the two.

346

Chapter Four: PLAYER

They were all dead. The final gunshot was an exclamation mark to everything
that had led to this point. I released my finger from the trigger…and then it was
over.

My screen reveals a letterboxed wide-shot gracefully tracking into a snowy
Manhattan, while superimposed text informs me of the setting: “New York. The present
day. One hour ago.” The game’s “camera” follows police cars swarming through
desolate streets, then ominously tracks up a tall skyscraper to reveal Max Payne—the
game’s titular hero—standing atop the building, gun in hand, a beaten mess. As the
camera pulls in on Max’s bruised and bloodied face, the above words are spoken in
voiceover as Max begins to recount the twisted story that has led him to this point. As
Max begins his flashback, the game segues from this cinematic opening into a noir-ish
graphic novel style, depicting events through painterly stills—complete with word-bubble
dialogue—while Max’s terse, hard-boiled narration relays his memory of the night that
everything fell apart.
To make any kind of sense of it, I need to go back three years -- back to the night
the pain started.

I was still in the force back then. NYPD, Manhattan, Midtown North Precinct.
Hell’s Kitchen.

347
On-screen, we see Max refuse a fellow officer’s offer of undercover work as he flashes a
silver wedding band, insisting that “Michelle and the baby come first,” and demonstrating
his cleaned-up lifestyle by presenting his last cigarette. The scene transitions to an
idyllic suburban street as Max approaches his front-door with a cheery “Honey, I’m
home!” His narration continues:
Life was good. The sun setting on a sweet summer’s day, the smell of freshly
mowed lawns, the sounds of children playing -- a house across the river, on the
Jersey-side. A beautiful wife and baby girl. The American Dream come true.
But dreams have a nasty habit of going bad when you’re not looking.

The sun went down with practiced bravado. Twilight crawled across the sky,
laden with foreboding.

The last panel of this faux comic book sequence depicts the entryway to Max’s house,
draped in foreboding low-key lighting that contrasts the sunny atmosphere of the
previous images. The game then transitions back into a cut-scene, showing the same
room, now in 3-D, along with the words “New Jersey. Three years ago.” As I watch Max
entering the house, his voiceover concludes: “I didn’t like the way the show started.
They’d given me the best seat in the house—front-row center.”
And with this, Max becomes my avatar. I control Mr. Payne from an over-the-
shoulder perspective, and having gotten the hint that something is amiss, I begin to
investigate the house with weapon drawn. As soon as I move forward, I notice the
picture frames on the wall in disarray, along with graffiti spray-painted over the
wallpaper—and Max obviously notices as well. An exclamation mark appears over his
head as he looks at the disarray, uttering, “what the hell?”—a device which not only
voices Max’s concern, but ensures that the player notices such spatial details by
directing their attention. I move toward the graffiti—a V underneath a syringe—and
press the “action” button, which cues Max’s voiceover: “Something ugly had been

348
tattooed on the wall, a map of things to come. It was a poison syringe, a magic tag full
of diabolical meanings.”
As I continue upstairs, I can hear the sounds of a baby crying, men’s muffled
voices, gunshots, and a woman’s scream. All the while, Max continues to vocalize his
concerns, exclaiming “Michelle!” while I try to navigate him towards the intruders. A
gunman bursts from the baby’s bedroom; Max announces “freeze! NYPD!” yet the
gunman opens fire. I take him down with a few shots—along with a second hoodlum—
and enter the room to find blood adorning the walls, the baby lying dead in the crib. “No,
no, please God, no!” Max cries, while the eerie twinkling of a music box continues to play
in the background. I move ahead into the next hallway, and can hear another man
grumbling behind the closed door ahead.
I push through the door, and another gunman immediately opens fire. I return
fire, and his body twists through the air in “bullet-time” slow motion before landing in a
heap. Max cries, “No, no, God, please, no, Michelle!” as his head turns to the left, and
after turning my perspective to follow Max’s gaze, I find his wife slain on the bed. The
game returns to a cinematic as Max kneels on the floor beside her, and the scene ends
with an illustrated panel of Max holding his dead wife, overlapped with a final cry of “No!”
Looking solely at these opening moments, we can already see that there is
something very strange going on with the narration of Max Payne. Max is addressing
us, the audience, as he tells his story in the first-person—saying “I’m Max Payne, this is
my story.” But simultaneously, the game itself is addressing us in the second-person—
saying “you’re Max Payne, and this is your story,” leaving us with the very curious
phenomenon of being told the story that we are playing by the very character that we are
playing. And just for good measure, we also have cinematic and graphic novel
sequences which essentially show Max in the third-person, functioning independently of

349
any sense of a narrator (the provider of the discourse) or narratee (the receiver of the
discourse). As a result, we’re receiving intermingled first-, second- and third-person
address—all in reference to the same character!
Max’s glance at the spray-painted graffiti as he enters the house demonstrates
all three in a singular moment. As I control Max walking through the house, Max is
treated in the second-person, via my control over the avatar (“you are walking through
the house”). While I walk, however, Max looks at the graffiti on his own, a third-person
instance (“he looks to his left”), and his separateness from me is evinced in that his
glance is meant to draw my attention, to indicate that Max has noticed something which I
may have not.
302
Pressing the action button then cues first-person exposition from Max
himself, so that the button-press is essentially requesting information from my future
self—explanation of the graffiti’s significance as told by Max himself, narrating from that
rooftop at the story’s end. In that single moment, then, all three narrative persons are
applied to Max at once.
This pastiche of narration results in several peculiarities, and they begin with
those few opening words that introduced our hero: “I released my finger from the
trigger…and then it was over.” The very first words of the game inform me that the story
is over. I haven’t even begun playing yet, and the story is already over! The conceit of
the contemporary videogame is that by inhabiting an avatar within the gameworld, the
player is given agency within the diegesis, provided some manner of free reign to act of

302
These two actions—Max’s movement and his glance at the graffiti—are distinguished,
fundamentally, in that Max’s movement is reliant on my activity via the interface, requiring that I
steer Max in a one direction or another, while his head-turn is not. I have agency in Max’s
movement, and am able to recognize it as the result of something that I have done; his look
toward the graffiti, on the other hand, cannot in any way be credited to my activity. As such,
when I steer Max through the house, I recognize that movement on-screen as the computer’s
translation of my action, showing me what I am doing in this space—“you are walking through the
house”—however when Max glances at the graffiti, I cannot trace this action to anything that I
have done, leaving this an entirely third-person instance independent of my agency within the
game.

350
their own volition, to move about with freedom and perform their own narrative actions,
thus participating in the creation of the story while experiencing it first-hand. After all, the
whole point in playing a game is to achieve a goal, to attempt an outcome that is
uncertain. Yet by opening with the ending, Max Payne denies any such uncertainty.
The sleight-of-hand of linear games—those games which provide a fairly rigid
story that the player is only meant to realize through participation—is to provide the
impression that the player maintains autonomy in the diegetic world, that the player has
some control over the story, despite all evidence to the contrary. But Max Payne
harbors no such illusions—it immediately lets the player know that this story has already
occurred, and that Max has indeed reached the final scene. The story is finished, and
thus the player is experiencing not a present-tense reality, but a flashback. The
narrative is fixed. This, of course, directly opposes the notion of the interactive text. The
progressive credit given to videogame narrative has been the potential of an infinite
number of story variations (even if those variations are merely playthroughs which end in
death or failure rather than the game’s successful ending). However the flashback
structure of Max Payne does not allow for such variations. If Max dies midway through
the narrative, then who’s narrating the tale? It cannot be Max narrating from the final
scene, for if he dies, then the final scene never occurs. Any ending other than the
game’s ideal ending would create a temporal paradox—thus by ending with the final
scene, the game tells us right away that only one ending is, in fact, the “correct” ending
to the story. The player’s job is then not to participate in the creation of the story, but to
realize its correct sequence of events.
An equal peculiarity is in our relationship to our avatar. Max talks to us, tells us
his story by recounting these events in the first-person in vivid and charismatic detail,
creating a relationship between Max-as-narrator and the player-as-audience.

351
Throughout the game, cut-scenes and other narrative interstitials continue to be narrated
by Max, just as in this introduction, playing into the game’s hard-boiled mystique and
noir-ish agenda, and this narration even extends into gameplay so that Max becomes
our guide through this narrative world—as I navigate the subway station in the ensuing
level, Max continues to provide exposition in voiceover: upon finding a dead transit cop,
Max interjects, “Death was in the air at Roscoe Street. I would have to find Alex fast”;
when I pick up a bottle of painkillers (the game’s version of a health power-up), Max
informs me, “the pills would ease the pain”; and so on.
And this would not be at all unusual—were it not for the fact that we’re supposed
to be Max. Numerous videogames include a character who acts as the player’s guide,
mentor, commander, or some other figure who provides relevant exposition that helps
guide them through the game (we saw a fairly standard example in chapter three with
Colonel Campbell, who provides exposition by radio throughout Metal Gear Solid). Yet
in this case, that expository figure is our own avatar. The game suggests that I am Max
Payne, that playing allows me to inhabit both the body and persona of Max, while Max
simultaneously addresses me in the first-person by relating his story. Not only does it
becomes impossible to imagine the game as a present-tense scenario, but it becomes
impossible to conceive of myself as experiencing the story from the point-of-view of Max;
I must exist both outside the narrative and within the narrative simultaneously, playing in
the present as Max, while also observing a past-tense narrative as an outside spectator.
These phenomena occur because the dual-positioning of videogames frequently
necessitates at least two different modes of narration. When addressing the player
externally as an observer of the story, suggesting the avatar to be an autonomous
character outside of the player, the game must employ third-person narration (or, if the
character themself functions as a narrator, first-person). Yet when addressing the player

352
internally as the avatar, suggesting that the player is the person to whom the story is
happening, videogames inherently become second-person narratives.

Whereas chapter three addressed the nature of avatars and the player’s
relationship to that pivotal figure, we conclude in chapter four by looking at the player’s
relationship to the videogame as a whole. In many ways, this will be a continuation of
chapter three’s concern with player-positioning—and it will, in fact, continue to address
the vital role that the avatar plays in this regard—yet this chapter is more dominantly
concerned with the player’s relationship to videogame discourse and the manner by
which they are addressed by the game-text. To discuss the videogame player, then, this
chapter will be primarily concerned with the narration of videogames—the discursive
manner(s) by which videogames present their worlds, deploy their systems, and engage
their audience.
But to do so, we’ll first need to do some terminological housekeeping. Sadly, the
critical theory of the past several decades has left some vital narrational terms in quite a
bit of disarray, so before we can fruitfully address videogame narration, we’ll need to first
clean up our use of the narrative person.

353
This is Your Life: Videogames and Second-Person Narration
Videogame enthusiasts, I’m about to blow your minds: first-person shooters do
not offer first-person narration.
Seriously. Half-Life, Halo, Doom, Deus Ex, Call of Duty, Battlefield 1942,
Goldeneye 007, Quake, Unreal, Wolfenstein 3D, Rainbow Six, Marathon, No One Lives
Forever, F.E.A.R., America’s Army—none of these offer a first-person perspective.
Coined to reference those videogames that provide the visual point of view of
their avatar rather than showing the avatar itself, the term “first-person shooter” has
become one of the most common phrases in the videogame medium, and arguably its
most readily identifiable genre. And yet the term itself is actually a misnomer.
Unfortunately, several decades of critical theory and popular discourse has
severely confounded our use of the narrative person so that we now commonly
associate first-person narration with an optical point of view. In its original literary model,
the narrative person was solely concerned with deixis—with the discursive orientation of
the sender and receiver in relation to the content (i.e., am I conveying information about
me, about you, or about someone else?). And verbally, it’s very straightforward—we
can simply tell by our use of pronouns. But for some reason, in translating these terms
from a narrating voice to a narrating camera, we’ve completely thrown out any attention
to the discursive situation and have focused instead on the distinction between
subjectivity and objectivity, POV shots and everything else, and equated these with the
first- and third-persons. This is a horrendous translation, as evinced by the complete
lack of a second-person, so the first thing we need clarify is that optical point of view and
the narrative person are two entirely separate issues.
We all remember learning about the grammatical person in middle school English
(you do remember, don’t you?). Hinging on our use of personal pronouns in reference to

354
a subject, the grammatical person is only concerned with three questions: who is
conveying the information, who is receiving it, and who is it about? These questions are
our only concern because personal pronouns are deictic words—they only gain meaning
in the context of discourse, the context of who is addressing whom about whom.

When the speaker discusses themself, they employ the first-person (“I”), when they
discuss the receiver, they employ the second-person (“you”), and when they discuss
someone outside the discourse altogether, they employ the third-person (“he,” “she,” or
“they”). Thus each grammatical person is able to address the same subject; the choice
of which relies entirely on discursive context.
Now consider Half-Life’s manner of discourse in conveying the story of Gordon
Freeman. Freeman is the player’s avatar, the game is relayed entirely from Freeman’s
visual point of view, and it’s this POV that earns the game its moniker of “first-person
shooter” (that, along with the game’s significant amount of shooting). Yet when one
plays Half-Life, Freeman does not address the player by saying, “I’m Gordon Freeman,
this is my story, this is what happened to me, and this is what I saw.” Rather, the game
itself addresses the player, saying, “you are Gordon Freeman, this is your story, this is

355
what is happening to you, and this is what you see.” “First-person shooters” like Half-
Life actually provide the cleanest and most definitive example of second-person
narration in contemporary media; in offering such a deliberately internal experience,
Half-Life elides nearly any trace of Freeman as a character outside of the player, and
resultantly, all the information conveyed throughout the game is done so under the
context of “you” being Freeman, couched under the discursive statement “this is
happening to you.”
The term “first-person shooter” and the industry parlance of distinguishing first-
and third-person perspectives is especially regrettable because they completely elide the
very thing that makes videogame narration unique from the way that previous media
narrate. Unlike film, television, literature, theater, or any other narrative medium,
videogames position their audience as the central actant in their narration, rather than
their narrator or a third party, and as such, the videogame medium’s misguided reliance
on the terms first- and third-person is actually very unfortunate. How, then, have we
ended up with this term “first-person shooter,” a term that would suggest quite the
opposite of a second-person experience?
We can credit several reasons, but primarily, it’s been the result of the strange
lineage of the narrative person, descending from the original grammatical person,
through literature, through cinema, and eventually into videogames. The concept has
essentially suffered from fidelity-degradation—as it has descended through multiple
media, it’s fallen victim to the perils of translation in what amounts to a decades-long,
critical theory game of telephone, as slight shifts in each translation have accumulated to
create an entirely different meaning.
Originating in linguistics to address verbal discourse of any kind, the first
translation was in adapting the grammatical person to discuss modes of literary

356
narration. And this translation seems an unproblematic one. Similarly hinging on
pronoun usage and discursive context, the literary narrative person simply applied
deictic address to a more specific manner of verbal discourse—storytelling—resulting in
first-person narration (stories about “me”), second-person narration (stories about “you”),
and third-person narration (stories about “him” or “her”). Yet although this was an
accurate usage of person, complications were already afoot. The first problem was that,
traditionally, the narratee all but disappears in storytelling, thus the second-person
largely fell by the wayside. There have been some examples of literature told in the
second-person over the years, yet they’ve been few and very far between. By and large,
literary narration prefaces the narrator and elides the narratee. Some literary narrators
make explicit reference to themselves as a character, yet even those who don’t still
clearly exist, as someone is providing the words on the page. Yet we rarely get narrative
references to “you,” and even when we do, they tend to be in passing rather than making
the narratee significant to the story. As a result, the narrative person quickly became an
unofficial binary—first-person vs. third-person, distinguishing whether the narrator is a
diegetic character or someone outside the action.
303
This problem was further
compounded as the narrative person became conflated with a separate critical term—
“point of view,” an ambiguous concept used to reference a wide variety of issues, from
narration to subjectivity to emotional involvement to mood to empathy to visual
perspective.
304
The seeds of confusion were sown, and the narrative person was primed
for further troubles.

303
As evidence of this, think back to your grade school English classes—when discussing first-
and third-person narration, did your teacher even mention the possibility of second-person?
304
Gerard Genette is among those who have argued against “the confusion between the question
who is the character whose point of view orients the narrative perspective? and the very different
question who is the narrator?—or more simply, the question who sees? and the question who
speaks?” Genette, 186.

357
As I noted in chapter one, film theory of the mid-to-late-20
th
century frequently
looked to literary theory as a means of academic legitimacy. Using the “film-as-text”
metaphor to garner critical esteem for their burgeoning field, scholars systematically
appropriated all manners of literary concepts to justify academic attention to the cinema.
The narrative person was among these, frequently justified via an “enunciation theory” of
the cinema deriving from the work of French linguist Emile Benveniste (who counted the
I/you/he-she distinction among his discussions) and other linguistic models.
This translation was a much heartier endeavor than applying deixis to literary
narrative, as film theorists had to cope with the particularly challenging task of moving
from the narrating voice of literature to the narrating camera of cinema. The most
common translation was to address first- and third-person narration as analogous to
subjectivity and objectivity (and often in relation to Benveniste’s concepts of discours
and histoire). Mark Nash provides a prototypical example in his 1976 examination of
Vampyr, arguing the film to exhibit marks of narrative person that we can equate to
personal pronouns.
305
The more reductive variation that has reached—and stuck—in
the popular consciousness is to address first-person narration as analogous to an optical
point-of-view shot, and third-person as analogous to everything else. Yet as David
Bordwell points out, such readings suffer from a flagrant and undeniable flaw—the lack
of even a possibility for the second-person.
[T]he comparison breaks down when [Nash] fails to find filmic equivalents for the
second-person function. …[W]hat would a second-person image look life? Nash
can find no instances because we have no idea of what could be the pertinent
marks. …[I]f in analyzing a film we cannot distinguish first-person discours (i.e.,
shots bearing personal marks of the author) from second-person discours (i.e.,
shots bearing personal marks of the viewer) then the category of person has no
equivalent in the cinema.
306

305
Mark Nash, "Vampyr and the Fantastic," Screen 17 (1976).
306
Bordwell, 23-4.

358
Despite such a convincing and straightforward argument, however, the POV/non-POV
model has persisted in both critical and popular discourse for applying the narrative
person to the cinematic image. And as a medium frequently struck by cinema envy (and
practically begging to be discussed in culturally familiar terms), videogames were quick
to take up the terminology.
As Wolfenstein 3D and Doom rose to popularity in the early 1990s, the term
“first-person shooter” was coined to describe those and other games in which the player
apprehends the world from the visual perspective of their avatar, with gun-filled hands
extended in front of them. With it, “first-person” and “third-person” became the common
parlance for distinguishing a POV interface from those perspectives that feature an on-
screen avatar. But much like Bordwell’s criticism of Nash, we see the same, rather
obvious problem arise: if we’re going to use the terms first- and third-person in this
manner, then what would we posit as the second-person equivalent? What would a
second-person interface look like, and how would it be any different from a first-
person?
307

This conundrum—the difference between a first- and second-person camera—
has been the fundamental problem for applying the narrative person to visual images.
While the first- and second-person are obvious enough in verbal language, the average
layperson has a difficult time distinguishing the two in other modes of discourse without
definitive pronouns at work. The narrator and narratee become nearly invisible without
explicit verbal narration, and we rely on the roles of the narrator and narratee to

307
Mark S. Meadows has suggested an over-the-shoulder shot as second-person—seemingly
because it’s not quite a POV, and not quite a detached perspective, and thus numerically falling
between the first- and third-persons. A fair attempt at bringing reason to our illogical use of these
terms, to be sure, but clearly lacking any connection to “I”, “you,” or “he/she.” This reasoning
essentially repurposes the terms altogether by removing it from any reasonable connection to the
grammatical person and moving it instead toward degrees of subjectivity (first being the most,
third the least, and second falling in the middle). Meadows, 74-6.

359
distinguish between the provision and reception of discourse, as well as the object’s
relationship to either—without the sender/receiver relationship, we cannot assign deixis.
As a result, many of us simply can’t tell the difference between the first- and second-
person (and the scant history of second-person narration makes this all the more
difficult). Distinguishing whether or not the object is connected to the discourse isn’t that
tough (solidifying the third-person), but when we see an image projected on screen, how
do we distinguish between the perspective of the sender and the receiver? Common
sensibly, we might ask: is there any difference?
There is, but it does not hinge on visual point of view, nor does it hinge on
subjectivity vs. objectivity—after all, literature offers plenty of examples of subjective
third-person narrators as well as objective first-person narrators, so this distinction fails
even before the translation to the screen. To discuss the narrative person on a screen,
we need put POVs aside and focus instead on the original concern of the grammatical
person: discursive context—who is providing the information, who is receiving it, and
who is it about?
Classical Hollywood filmmaking offers us the most blatant rendition of third-
person narration on-screen, thanks to its relentless disavowal of the discursive situation.
Utilizing conventions of internal coherence, spatial and temporal continuity, verisimilar
causality, and formal transparency, the classically realist film strives for a seamless
impression of reality and an elision of its own making, appearing “natural” and thus
drawing little attention to its conventional nature. Resultantly, classical filmmaking
effaces the marks of enunciation, elides any sense of artist or audience, and resultantly,
any sense of "I" or "you" in the discursive exchange. What we witness on screen is
simply happening to "someone else"—to a third-person. And significantly, this does not
change when we view a subjective shot.

360

In these subsequent shots from Casablanca, we’re shown the character of Rick looking
off-screen, followed by a shot of the doorman taken from Rick’s perspective. Yet we
hardly read the cut between the two shots as a sudden shift in narration; when we cut to
Rick’s point of view, we do not imagine that Rick has become a temporary first-person
narrator. Rather, Rick remains in the third-person throughout; as a discursive statement,
we would best verbalize these shots as “here is Rick, and this is what he sees.” Keep in
mind that third-person narration has a long history of letting us in on the subjectivity of
characters, whether it be their thoughts, their concerns, or their visual perspective. As
such, POVs and non-POVs (subjective and objective viewpoints) have little relation to
the narrative person.
The primary reason that our use of the narrative persons on-screen has become
so misguided is from attempting to force them all onto this model—onto screen narration
which doesn’t actually invoke a sender or receiver of the discourse. If we want to talk
about first-person narration on screen, we should actually be looking at recent films like
The Blair Witch Project and Cloverfield, films that suggest their fictional characters to be
providing the discourse itself—the very footage that we are watching. Unlike Rick’s POV
in Casablanca, where Rick is never suggested as the source of the discourse,
everything that we see in Blair Witch and Cloverfield—the entirety of the discursive
information conveyed by the text—is suggested to be sourced in the characters

361
themselves, who traverse the diegetic space with cameras in hand. And whether those
characters put the camera to their eye or turn it on themselves doesn't change that. The
films still feign first-person narration, an "I" statement. Just as a literary first-person
narrator relays a story while also existing within that story, so too the protagonists of
Blair Witch and Cloverfield operate as the providers of the discourse while also existing
within the discourse. Similarly, autobiographical films and documentaries whose
filmmakers take part in the action on screen (such as Michael Moore’s Roger and Me or
Sicko, both films in which Moore chronicles his own investigations) evince an “I”
statement, and whether or not the artist appears on camera doesn’t change this—either
way, we still recognize the source of the discourse as part of the content.
To discuss second-person narration on-screen, however, we’d best look to
videogames, as every videogame addresses the player by providing an avatar and
saying "this is you." It might provide the POV of that avatar, as in Half-Life or Halo, or it
may show the avatar on-screen, as in Max Payne or Tomb Raider, but regardless, the
game asserts the receiver of the discourse (the player) to be part of the content. All
videogames do this—they present a fictional situation and state “this is what you are
doing,” and this applies to abstract games like Tetris and Bejeweled just as readily as
representational games like Max Payne and Half-Life. When we play Tetris, the game is
essentially saying "Blocks are falling, and you are controlling those blocks." This is not
true, of course—you're not controlling blocks, because there ARE no blocks. None of it
really exists. But the game presents the fictional situation of blocks falling and asserts
“you,” the player, as the actant in that fiction. So we can apply this model of second-
person narration even to texts that we would not necessarily identify as narrative ones.
In fact, to look at the history of second-person narration, we’d have to look to
texts that we do not traditionally think of as narrative ones. Staking the theoretical

362
claims of ludology—an academic discipline of game studies that is variously defined, yet
is commonly set in opposition to narratological approaches—Gonzolo Frasca suggests
that we should examine videogames as simulations rather than narratives, arguing “even
if simulations and narratives do share some common elements…their mechanics are
essentially different.”
308
While we could challenge this sentiment, I think we can
understand the distinction that Frasca is trying to make—videogames are certainly
different from narration as we’re used to it. But the problem is not in narration; the
problem is that, relatively speaking, second-person narration has seldom been used in
the past. We’re not used to it as a viable mode of narration, thus we don’t immediately
recognize it as narration at all.
In literature, there have been a few examples over the years—works like Italo
Calvino’s If on a Winter’s Night a Traveler (which opens by positioning the reader as
having just sat down to read that very book) or interactive books like Choose-Your-Own-
Adventure novels. Orally, there have been a few as well—role-playing games like
Dungeons & Dragons, in which the dungeon master narrates a diegetic world and asks
players to make decisions, are probably the best example. But if we want to look at the
history of second-person narration in mimetic form—via showing rather than telling—
we’d have to go to simulators. When a pilot climbs into a flight simulator, that simulator
essentially says “you are in the cockpit of an airplane, flying over Seattle” when in fact
the pilot might be sitting in a hangar in Houston. The simulator is narrating a fiction,
positioning the user in an alternate diegetic space, making the simulator a manner of
second-person narrator—a discursive device that positions its receiver as the actant
within its diegetic content, a narrator whose content is always premised under an implied

308
Gonzolo Frasca, “Simulation vs. Narrative: Introduction to Ludology,” in The Video Game
Theory Reader, eds. Mark J.P. Wolf and Bernard Perron (New York: Routledge, 2003), 222.

363
statement of “this is what you are doing.” In this regard, Frasca’s argument becomes a
convincing one; rather than argue that simulation is fundamentally different from
narration, we’d be better suited by arguing that simulation (at least, the particular manner
of simulation that is comparable to videogame discourse) is a manner of second-person
narration, and that such narration is fundamentally different from traditional (that is, first-
and third-person) modes of narration.
309

Tellingly, most screenic examples of second-person narration could reasonably
be characterized as simulations. Among the scant few cinematic examples we might
find is the roller coaster sequence from 1952’s This is Cinerama (a compilation of
footage made to showcase the new widescreen format of Cinerama) in which the
camera simply rides along with the roller coaster. Unlike Rick’s POV in Casablanca, the
roller coaster sequence is not prefaced as a character’s point of view; instead, the
audience is merely positioned as the rider. While Hollywood frequently practices these
sorts of POV gimmicks, they are typically framed by showing a character climbing
aboard a roller coaster (or whatever the context may be) so that we know whose POV
we are seeing, allowing the subjectivity to be couched within classical Hollywood
transparency. The spectator is thus allowed to participate vicariously through a
character’s point of view, couched under the third-person understanding that this is
happening to someone in the film. Yet This is Cinerama merely positions the audience
as the participant themself. Rather than implying “this is what John sees as he rides the
roller coaster,” the film simply addresses the viewer by saying, “You are on a roller
coaster! Whoo-hoooo!” And in doing so, this segment simulates the experience of
riding a roller coaster.

309
I should clarify that conceptually, I would agree with the vast majority of what Frasca has to
say about simulation as a model for videogames; my only dispute is in his narrow positioning of
narration, essentially equating to “the traditional way that people tell stories.”

364
We could also look to amusement park exhibits like Star Tours or Back to the
Future: The Ride, which equally position the audience in a separate diegetic space,
saying “this is what is happening to you.” Both attractions feature an expository
prologue that occurs as attendees wait in the queue, establishing the ride’s fictional
context—that the attendees are space tourists flying to the forest moon of Endor or
volunteers for a time travel experiment, respectively—then proceeds to seat attendees in
their respective vehicles (a spaceship or time machine), featuring a large screen to
simulate their vantage point. The contraption bumps and rolls along as the vehicle
seems to fly through a wide variety of tense situations and close calls, the attendee all
the while positioned as a passenger on the journey. Much like a flight simulator, these
rides equally position the viewers in an alternate diegetic space, simulating a madcap
journey through outer space or the spacetime continuum by suggesting “this is
happening to you.”
But in the media that we traditionally think of as narrative media, second-person
narration is practically non-existent. This is largely because traditional narration occurs
“on rails”—it’s linear and predetermined, having only one path to follow. When the story
involves someone else—the first-person “I” or the third-person “he” or “she”—we can
justify this linear nature as the result of those characters’ actions and decisions. But
when the receiver is injected into the narration, linearity becomes problematic. In the
second-person, linear storytelling would require the narrator to force actions, decisions,
emotions, and an identity onto the narratee, ones that the audience may very likely not
accept as their own. Unless the speaker is telling the receiver an actual story that
happened to that person, it becomes difficult to insert the audience into a linear story, as
linear narration asserts details of both action and character—tells us what the actants
did and why, as well as who they are and how they felt—making linear fiction and

365
second-person narration relatively incompatible. A narrator can easily tell us their own
actions and emotions; a narrator can easily tell us someone else’s actions and emotions;
but assigning those details onto the audience is much trickier. Resultantly, outside of
interactive contexts, second-person narration has commonly relied on two necessities:
empty character—making no reference to who “you” are, thus tailoring the role to any
audience—and an “on rails” context—one that does not offer choices of action, thus
alleviating decision-making. It’s quite telling that This is Cinerama’s roller coaster
sequence, Star Tours, and Back to the Future: The Ride all position their audience as
passengers in vehicles, contexts that offer no activity but to ride along, and that
accommodate any audience. Equally telling is that If on a Winter’s Night a Traveler
opens with the singular detail that Calvino can guarantee about his reader: that they’ve
just begun reading that very book.
The alternative, or course, is the route chosen by Choose Your Own Adventure
novels, Dungeons & Dragons, and videogames: interactivity—narrating a situation, then
asking “what will you do?” Keep in mind that such interactivity often need not involve a
complete range of all possible actions—a Choose Your Own Adventure novel or
hypertext adventure might position their audience at a fork in a road and simply ask: “will
you go left or right?” Practically, there should be plenty of other choices in this
situation—one could simply stop where they are and wait in the middle of the road, or
turn around and head back the way they came—yet even the simple choice between two
possible actions allows the receiver to take possession of the action. “You” make the
choice, and therefore it is your activity—even if it is a rather heavy-handed one, born of a
vastly limited set of options.
This solves the problem of activity, but not necessarily that of character. Each of
these examples involves protagonists whose identity clearly do not match the player

366
themselves—the first entry in the Choose Your Own Adventure series, The Cave of
Time, positions “you” as an adolescent visiting their Uncle Howard; Dungeons &
Dragons games position “you” as a knight, sorcerer, thief, or some other medieval
fantasy figure; and Half-Life positions “you” as the very specific character of Gordon
Freeman, a theoretical physicist working in a secret underground research facility. How
are these perspectives any different from Rick’s POV shot in Casablanca—the
perspective of a third-person, an accounting (or recounting) or what “he” saw (rather
than what you see)? Because unlike Rick, these protagonists are a special type of
character—avatars. Rick is a character with whom we empathize and vicariously follow
while still recognizing as “someone else”—no one in the audience would point to Rick as
their own embodiment. But this is not the case with an avatar. Rather than a mere
empathetic protagonist, the avatar “stands in” for the receiver, providing a figure that the
audience does indeed point to as themself.
Recall our discursive situation from chapter two—content conveyed via a means
of transmission between a sender and receiver, four components common to every
instance of discourse:

In a textual model, then—instances in which the discourse is contained within a
discursive artifact—the content and means of transmission are what make up the text:

367
These models, however, merely describe the broadest discursive situation, that which
applies to any instance of discourse. In narrative discourse—a particular manner of
discourse—we actually have to expand this model. We still have those four basic
components, yet within the text itself, we need introduce a second, interior discursive
instance: the activity of the narrator.
The narrator is the pivotal figure in narrative discourse—after all, this is who does
the narrating, the very act for which a narrative is named—and this individual should not
be confused with the author or artist themself. The narrator is a figure inside the
fiction—inside the magic circle. When we step inside the magic circle, we receive the
story not from the author, but rather from the narrator, a figure who tells the fiction as
though it were real, asserts a truth claim that we agree to via a willing suspension of
disbelief. Outside the magic circle—outside the text, outside the fiction—we know the
proceedings not to be real, but when we engage the fictional text, we engage not an
author, but a narrator: a storyteller who asserts details of people, places, and events as
though they all have existed.
The narrator is essentially a role that the author assumes for the sake of the
fiction. This is most blatant in first-person narration, as when Arthur Conan Doyle
speaks through Dr. Watson to narrate The Hound of the Baskervilles. Such narration is
essentially a case of role-playing—Doyle assumes the role of Watson, and relates his
story of Sherlock Holmes solving a mystery as though it were a real event that he
witnessed and in which he participated. Yet while first-person narration may be the most
blatant case, second- and third-person narrators serve an equal function, even if those
figures are not as plainly delineated.
Clearly the figure who narrates the futuristic thriller Do Androids Dream of
Electric Sheep? must be separate from author Philip K. Dick, because the narrator

368
speaks of the story’s futuristic events in the past-tense. The narrator must exist even
further into the future than the story itself, so that narrator couldn’t possibly be Dick.
Instead, just as Dr. Watson, the narrator is a fictional role that the author assumes in
order to tell a fictional story. Despite having no clear embodiment, the third-person
narrator is still a separate figure from the author, and we can apply this same logic to
any fictional narrative. The fictional narrator offers a recounting of events from a fictional
world, and by presenting those events “as though they were real,” the narrator must exist
within that same fictional world (typically from a time posterior to the story’s events),
making that presence—regardless of how amorphous or ambiguous it might be—
separate from the author themself.
In mimetic narration, the “narrator” is typically an even more ambiguous entity.
As with Doyle’s Dr. Watson, first-person narration still provides a clear sense of a
discursive source—The Blair Witch Project and Cloverfield each provide diegetic
characters that we identify as the fictional source of the film in its entirety, narrating
figures that we clearly recognize as the source of the discourse. But even more so than
Dick’s futuristic narrator, classical, third-person films like Casablanca or Star Wars elide
any sense of a “telling” of events. In most mimetic narration, the narrator all but
disappears, absorbed into the text itself and leaving behind only a faint trace of a
narrating presence, indicated by the logic, coherence, and organization of the narrative.
Yet as Christian Metz observes, we do still sense this presence:
The impression that someone is speaking [in a narrative film] is bound not to the
empirical presence of a definite, known, or knowable speaker but to the listener’s
spontaneous perception of the linguistic nature of the object. …The spectator
perceives images which have obviously been selected (they could have been
other images) and arranged (their order could have been different). In a sense,
he is leafing through an album of predetermined pictures, and it is not he who is
turning the pages but some “master of ceremonies,” some “grand image-maker”
who (before being recognized as the author…) is first and foremost the film itself
as a linguistic object…or more precisely a sort of “potential linguistic focus”

369
situated somewhere behind the film and representing the basis that makes the
film possible.
310

Even in the most classically transparent of narratives, we retain a sense that someone
has organized these events—even beyond the literal authoring process. This is the
presence that overcomes the disparate forces working together to produce the text,
creating a clear sense of seamless narration and offering the consistency and coherence
of a singular storytelling presence. This is the lingering, amorphous trace of the
narrator.
311

But to whom are all these narrators speaking? Who is it that Dr. Watson
addresses? To whom does Dick’s futuristic narrator address their story? Who do we
conceive as receiving Blair Witch and Cloverfield’s fictional footage, and who receives
the ambiguous address of Casablanca or Star Wars? Obviously it can’t be us, the real
world audience, because all of these narrators exist in fictional worlds, while we do not.
These narrators are obviously communicating to someone, even if, like many narrators,
that someone is never enunciated.
The author communicates to the audience, yet this is an exchange that occurs
outside the fiction, with the text itself as the discursive instance. But inside the text—or
more specifically, inside the fiction—the narrator communicates with a different figure:
not the audience, but the narratee.
As Seymour Chatman and others have observed, the narratee is a figure we are
only beginning to distinguish: “He is much less well known [than the narrator]; indeed,

310
Christian Metz, Film Language: A Semiotics of the Cinema, trans. Michael Taylor (New York:
Oxford University Press, 1974) 20-1.
311
Many will describe this narrating presence as the “implied author,” who is separate from a
(potential) narrator within the text (such as a narrating voice). This a valid distinction, one that
we’ll take up later when addressing levels of embodiment, however we need not worry over this
right now.

370
his very existence has only recently been recognized.”
312
Yet in every narrative
instance, this figure most certainly exists. Every discursive instance must involve both a
sender and receiver, thus if a narrator is to narrate, they must narrate to someone (even
if that someone is only implied or assumed). Just as the author assumes the role of the
narrator inside the fiction, so too we, the audience, assume the role of narratee. We
suspend our disbelief, enter the magic circle and accept the narration “as though it were
real,” by assuming this role. Thus we do not receive the fiction directly, but rather via an
intervening figure.
As such, narratologists will frequently present the narrating situation as occurring
something like this:

The author communicates with the audience, but only does so through the text, and thus
through the figures of narrator and narratee. Inside the text—inside the magic circle,
inside the storytelling context—the narrator communicates with the narratee. The author
and audience then, communicate, via a manner of role-playing—the author assumes the
role of narrator, the audience assumes the role of narratee, and the fiction occurs.
Yet with their dominant concern for the act of narration and the telling of the story,
traditional narratives have long prefaced the narrator at the expense of the narratee.
Storytelling’s transparent manner of delivery tends to elide a receiver—textual stories
are usually told as though to no one, a blank manner of narration that is simply delivered
into the ether. As Seymour Chatman notes:

312
Seymour Chatman, Story and Discourse, 253.

371
[E]very tale implies a listener or reader, just as it implies a teller. But the author
may, for a variety of reasons, leave these components unmentioned, indeed, go
out of his way to suggest they do not exist.
313

As such, it’s not surprising that narratology has made little examination into the
narratee—while some narratives annunciate the narrator and others let them fade into
the text, very few call attention to this “narratee” fellow.
At least, such was the case before videogames arrived. As a predominantly
second-person mode of narration, videogame discourse hinges not on the provider, but
rather the receiver of information—the one who plays—because all videogames are, to
some degree, suggesting “this is happening to you.” Thus while traditional stories have
long been concerned with the narrator, videogames are instead concerned with the
narratee. Using videogames—along with the scant few other examples of second-
person narration which have been produced over the years—we can begin to gain a
clearer picture of the narrative person, particularly in regard to mimetic narration. And
rather than focus on pronouns, point of view, or subjectivity vs. objectivity, we need
merely consider whether either member of the discursive situation is drawn into the
narration.
Third-person narration is that which elides both the sender and receiver, focusing
solely on the telling and evoking no sense of an “I” providing the discourse or a “you”
receiving it.

The clearest example of such narration is the classically realist model—eliding both
parties, and resultantly providing a sense that the story is simply providing itself, that the
information emitting from the screen or from the page simply “is.” Even in the literary

313
Ibid, 151.

372
model, where an undeniable telling is evinced by the very words on the page, the third-
person offers no explicit suggestion of who speaks those words, or to whom they might
be intended. Note that third-person narration can indeed be heavily subjective, thus
drawing explicit attention to the discourse itself—by providing commentary or stylization,
a third-person narrator supplies irrefutable proof of narration. Yet without first- or
second-person inflections, we’re given no context for who provides or receives that
commentary, or who we ought credit for that stylization.
First- and second-person narration, on the other hand, are those which attend to
the members of the discursive situation. The first-person draws out the sender of the
discourse. By annunciating an “I” or explicating the source of information, the first-
person privileges the narrator, pointing to their existence by explicit self-reference.

Such is the case in The Hound of the Baskervilles, The Blair Witch Project, and any
other text that projects the narrating figure into the diegesis, acknowledging some entity
responsible for the telling. Alternately, second-person narration draws out the receiver
of the discourse. By evoking a “you,” second-person narration privileges the narratee by
explicitly pointing to that figure’s existence.

Such is the case in a Choose Your Own Adventure novel, Half-Life, or any other text that
projects the receiving figure into the diegesis. So while both figures are requisite of
narration—of any sort—it is only first- and second-person narration that point to the

373
narrator and narratee (respectively) in a tangible sense, that draws out these figures by
acknowledgement of their existence.
314

In videogames, we merely refer to the narratee by another name: the avatar. Or
perhaps more precisely, the avatar is used as an embodiment for the narratee, a
discursive proxy for the audience, serving a similar function for the player as a first-
person narrator does for an author. Just as Arthur Conan Doyle dons the role of Dr.
Watson in narrating the Sherlock Holmes stories, allowing Doyle to address the
audience by stating “this is what happened to me” (“me” being Watson), so too the
player dons the role of Max Payne or Gordon Freeman, allowing the text to address the
player by stating “this is what is happening to you” (“you” being Max or Gordon). Just
like a first-person narrator, a second-person avatar is a special kind of protagonist, one
that is inhabited by a member of the discursive situation. In either case, the narrator or
avatar is a role that is adopted by someone outside the fiction in order to maintain the
fiction, an act of role-playing performed so that a story can be exchanged between “you”
and “I” as though we both existed in a world where that story were true. Both are
manners of role-playing—they’re simply performed by opposite members of the
discourse.
After all, the protagonist in second-person narration, most frequently, is not
actually you, but rather “you”—a hypothetical you, a figurative you, a character (whether
tangible or implied) that the narrator uses as a stand-in for the actual receiver. This is
the role of the avatar. Relying on the metacommunication of fiction, the second-person
narrator says “this is happening to you,” while also implying “this isn’t really happening to
you.” In truth, it’s merely happening to a fictional character; that the character is

314
And although it is rarely seen, it would be entirely possible for a narrative to evoke both first-
and second-person narration simultaneously—where a story to open “I led you to the door, and
we both went inside…” which narrative person would we credit?

374
referenced as “you” is simply a matter of how the character is discursively
contextualized. This is why the character of Max Payne can be addressed in the first-,
second-, and third-persons, all within the same text—each address occurs in a different
discursive context.
It is because traditional narration rarely invokes a receiver of the discourse,
relying instead on first- or third-person address, that the narratee, as Chatman noted, is
a relatively under-examined figure. Truth be told, videogames are the first medium to
significantly foreground the narratee, a possibility allowed by its interactive nature.
Unlike traditional media, interactive media are perfectly suited for second-person
address: they allow the receiver to interact, to role-play, to be the “you” in the story, a
feat previous media allowed only by vicarious association more suited to the third-
person. While previous media made it very difficult to address a fictional “you” without
employing a role that required no input nor decision-making—a role of literally being a
mere receiver—the interactivity of videogames allows for the necessary participation
required of true second-person narration.
Entire books have been dedicated to the complex roles and functions of the
narrator—a considerable portion of the field of narratology is dedicated to that figure.
Such works have addressed this individual from nearly every angle imaginable—the
variability with which they may be annunciated, their reliability and the validity of their
perspective, their subjectivity and objectivity in relation to the content, their relationship
to character(s), their temporal positioning, their narrative authority, and many, many
further issues. The narrator is the source of this discourse itself, and thus ultimately,
every observation we might make about narrative discourse must be bound up in the
narrator.
The narratee—as embodied in the avatar—is no less vital.

375
Fiction and Function: The Dual Roles of the Avatar in Videogame Narration
In chapter three, we discussed the player as being positioned in two places at
once: internally, as a participant in the fiction, as well as externally, as an outside
observer watching that fiction. The game’s mode of narration is directly tied to this
issue—yet whereas positioning dealt with the player’s reception of the discourse,
narration instead deals with its delivery. In the second-person mode, the game is
addressing “you,” suggesting the events at hand to be “your” story, thus positioning the
player internally by moving them inside the diegesis. Third-person address (and first-
person when applicable), on the other hand, suggests the events to be happening to
“someone else,” and resultantly positions the player externally by addressing the
protagonist and the player as separate entities.
315
When we discuss narration, then,
we’re addressing many of the same issues from chapter three—we’re merely looking
from the opposite end of the discourse.
The avatar is the key component to the functionality of videogames, operating as
both a member of the gameworld and as a systemic component of the game’s interface,
functioning as a mediating body between the player outside and the world inside. The
avatar facilitates the player’s access to the fiction of the diegesis as well as the systemic
operations of the game and its discourse, allowing them to exist in two places at once,
both inside and outside the videogame. Keep in mind that when I use the term avatar,
I’m speaking specifically of the two-beings-in-one figure that I elaborated on in chapter
three; the word itself is used in relation to a number of different contexts—chat rooms,
webpages, virtual reality—yet these are often a different manner of figure. When we
address videogame avatars, we’re discussing a complex figure that descends from the

315
When looking at the manner by which the player is positioned in relation to a character, the
distinction between first- and third-person address is irrelevant. Either way, the character is “not
you,” thus the player is positioned externally.

376
Hindu deployment of the word—a figure that is both the embodiment of an individual that
exists on another plane of existence, and a discrete being in its own right. It is a figure
that operates both as a fictional character and as an ergodic vehicle.

The Avatar as Fictional Character
On the one hand, the avatar is a diegetic character, developed as a protagonist
much like any other. Super Mario, Lara Croft, Solid Snake, Gordon Freeman, and the
Master Chief are all distinct individuals in much the same way that Rick Blaine, Luke
Skywalker, Jane Eyre, David Copperfield, and Willie Loman are. They are assigned
personalities, physical characteristics, distinct mannerisms, and other markers of
individuality—they are developed as unique persons.
When I describe an avatar as a “character,” I do so in the more progressive,
post-structural sense of that word, attending to the details and attributes of identity as
well as the “psychological essences” of personality, and treating characters as sentient
beings that can be examined in human terms. Yet this is no simple task, as the
components of character—both identity and personality alike—are hardly quantifiable.
Rather than account for the character as an entirety, then, we must assemble our sense
of character from the details provided, as Roland Barthes demonstrates when writing in
S/Z:
When identical semes [signifiers, descriptions, connotations] traverse the same
proper name several times and appear to settle upon it, a character is created.
Thus, the character is a product of combinations: the combination is relatively
stable (denoted by the recurrence of semes) and more or less complex (involving
more of less congruent, more or less contradictory figure); this complexity
determines the character’s “personality,” which is just as much a combination as
the odor of a dish or the bouquet of a wine.
316

316
Roland Barthes, S/Z, trans. Richard Miller (New York: Hill and Wang, 1974), 67.

377
Barthes is speaking of literary character, thus focusing more closely on verbal
descriptors, yet we still gain the sense of an abstract entity that we piece together
through characteristics, assembling our sense of a unique and complex being. Yet
where do we locate character? Beyond a simple name (“Lara Croft”) or image (the
pixilated figure that we identify as such), what is it that “makes” a person? At what point
can we say that we “know” a character? And how do we address a specific being when
characters are able to change over time? Character study can quickly become an
exercise in existentialism.
It is for this reason that Seymour Chatman describes characters as “a paradigm
of traits”: entities that materialize from the diegesis and are amalgamated from a wide
range of narrative data collected by the audience.
317
“[C]haracter is reconstructed by the
audience from evidence announced or implicit in an original construction and
communicated by the discourse,”
318
Chatman tells us, and as such are an assemblage
of qualities and personal details drawn from overt description (“Lara Croft is brave”) or
implied by suggestive details (her cool demeanor in the face of danger), the implications
of action (the adventurous feats that she performs), the reactions of others (the
admiration of other characters), and numerous other deployments of narrative
information. This is the process of “characterization”—constructing a sense of a unique
being through their component “characteristics,” i.e., their traits. Compiled, these details
construct a unified individual, yet that individual equally remains “open-ended, subject to
further speculations and enrichments, visions and revisions,”
319
for two reasons: (1)
though a cohesive totality, we can never know the entirety of a character, and (2)
characters may change and develop over time.

317
Chatman, Story and Discourse, 126.
318
Ibid, 119.
319
Ibid.

378
Chapman describes traits as parametric to the chain of events—they exist over
time, continue in part or in whole through the text. Traits may become significant in the
chain of events (such as casual motivation) and they may emerge from the chain of
events (as symptomatic or expressive of personality), yet they do not have strictly
determined positions in the chain of events—that is, while we can isolate the discursive
location of character X committing murder, we cannot isolate a discursive location for
their murderousness. Events have strictly determined positions in the story (their order
is deliberate) and are discrete (with a clear-cut domain between beginning and end).
Traits are not so limited—they “may prevail throughout the work and beyond” (precisely
why characters may continue through numerous sequels).
320

Traditionally, traits are assembled entirely by the author, and thus contained by
the text. Yet in interactive contexts, characters may also be assembled by the player,
only gaining full characterization through the playing—that is, character may emerge
from the game-text, or from the derived instance. In chapter three, we saw Solid Snake
developed as a very distinct character, with the vast majority of his characterization
rooted in the text; Snake’s personality, appearance, demeanor, mannerisms, and every
part of his identity are all written into Metal Gear Solid. Yet in chapter two, my Redguard
Crusader emerged in large part through my own design. I chose his race, profession,
gender, appearance, and name during the game’s opening, and his skills and attributes
developed through a combination of my activity (if I “sneak” a lot, his Sneak skill
develops, whereas if I charge into combat more often, his weapons skills would develop)
and the attributes I opted to upgrade over the course of the game. Further, I decide
whether he will be a noble character that helps others or a wicked individual who steals
and murders. And we can attribute such characterization even in games that do not

320
Ibid, 128.

379
offer the explicit customization of role-playing games. In Half-Life, the protagonist never
asserts himself in any way beyond the player’s activity, and as a result, Gordon
Freeman’s “character” is largely a projection of the player (whether they realize it or not).
Some of the game’s exposition may help to shape Freeman, and the game’s events
themselves may equally contribute, yet the majority of Freeman’s traits are those that we
project onto him, whether through our gameplay (Freeman kills everything in his path) or
our subconscious imagining of who he is (I imagine that he’s a defiant and cynical
person).
But regardless of where the characterization is sourced, such personal details
allow the avatar to function as a discrete, self-contained, fictional being—a paradigm of
traits that we conceive as a sentient individual. Yet while this model alone can provide a
fair accounting of the characters in other media—and perhaps even certain characters in
videogames (non-player characters)—it is not enough to account for avatars.
The traits which comprise a given character are typically many and varied,
defining all manners of identity and personality, yet for the sake of an example, we’ll
focus on a single trait: gender. In chapter two’s opening example of Oblivion, I chose a
male Redguard Crusader as my avatar, and within the game’s fiction, this marks a
significant difference from choosing a female Redguard Crusader. In the opening
moments of the game, the crude remarks that the dark elf in the adjacent cell makes will
differ based on the avatar’s gender, and even in instances where the dialogue remains
consistent regardless, players could easily read different interpretations based on the
changed context (for example, a male NPC’s dismissive remarks about a male avatar’s
combat-worthiness may be read simply as reflecting that character’s inexperience, yet
with a female avatar, we might read more into his disregard). The player’s relationship
to the avatar—whether immersive role-playing or more detached observation—may be

380
affected as well. As a character, then, the avatar’s gender (or any other trait, for that
matter) bears significance. However, as far as the rules and mechanics of Oblivion are
concerned, the male and female versions of that avatar makes absolutely no difference
whatsoever; each functions in a precisely identical manner, offering no difference in the
way that the player can navigate the gameworld and no difference in the unfolding of the
plot (aside from relatively trivial changes in dialogue—the dark elf makes crude remarks
regardless, the differences are really just a matter of logical context).
Many scholars have attempted to discern some transgender phenomenon in
such contexts—boys playing Tomb Raider, girls playing Halo—and in one sense (the
character sense, a narrative sense), there’s certainly something to this.
321
Not
surprisingly, these tend to come from media traditions used to addressing fictional
entities which have no function outside of their identity. But functionally, is there any
difference between inhabiting Lara Croft in Tomb Raider and Indiana Jones in Indiana
Jones and the Emperor’s Tomb? Between playing James Bond in 007: Nightfire and
Cate Archer in No One Lives Forever? Or between choosing the male or female
variants of a Redguard in Oblivion?
This is by no means to suggest that we should disregard the character status of
avatars or their markers of identity and personality. Narratively, we indeed find great
concern in our avatar’s identities, and pragmatically, we cannot ignore these traits, even
when primarily concerned with functional details—while multiple avatars may provide the
same functionality, their individual identities do matter. Yet falling back on the simple
traditions of character from other media—those which treat characters as psychological
beings, a paradigm of traits that we discern from the text—neglects the fact that avatars

321
For example: Marsha Kinder, Playing with Power (Berkeley, CA: University of California Press,
1991).

381
are not mere characters. They are indeed characters, yet they’re equally something
else.

The Avatar as Ergodic Vehicle
While the avatar is a manner of character, it is also unlike any other character—it
is inhabited by the player, and discursively serves the role of the audience, the very
person to whom the fiction is narrated. The avatar is equally a hollow vessel for the
player to inhabit, an empty body allowing the player access to the gameworld and the
functional processes of the videogame. In this regard, we would better describe the
avatar not as a fictional character, but as an “ergodic vehicle.”
Rooted in the Greek words for “work” and “path” and borrowed from physics,
Espen Aarseth uses “ergodic” to describe a certain manner of literary text, a term he
adopts over the more troublesome “interactive.” Aarseth decries the term “interactive”
when applied to such narrative work as Choose Your Own Adventure Novels, text-based
adventure games, and hypertext, arguing the word as inappropriate since the reader can
never claim authorial control. He chooses instead the word “ergodic,” where “nontrivial
effort is required to allow the reader to traverse the text,” in that the reader is either
encouraged or forced to participate in the progress of the narrative.
322
The reader
essentially becomes integrated into the functional processes of the text—such as
creating a linearity from a nonlinear work—so that their effort completes the structure,
and thus the narrative.
323

322
Aarseth, Cybertext, 1.
323
In later writing, Aarseth directly links the term to videogaming, describing ergodic work as that
which “forces a narrative structure (in the form of a single “correct” sequence of events) on its
players.” Of course, not all videogames force a single “correct” sequence of events on their
players—both Oblivion and Civilization offer events that can occur in such widely varying
sequences that their possibilities become near infinite, dictated only by the whims of players and
the emergence of play. Yet Aarseth’s point remains valid, as the player is still bound by the range

382
Aarseth’s concern, then, is that videogames do not see the interactivity
characteristic of the original model of discourse—two parties interacting so that neither is
solely sender nor receiver, making discourse the product of two interacting participants
rather than a single authoring one. While the player can operate within the text’s system
of rules (whether with other players or with the system itself), motivating emergent
possibilities within secondary levels of discourse, the player is given no opportunity to
contribute to the primary discourse of the text itself. The textual nature of videogames
guarantees that the primary discourse is fixed, thus while players can send input back
into the text, that input never returns back to the original senders.
324

The playing of a videogame remains a derivative experience in that the iteration of a
single match derives from a fixed set of rules, rooted in the game-text. The player may
interact with the text, yet they have no opportunity to influence that rule-set (at least, not
without hacking, thus creating a different text, and equally, a different game).
An “ergodic” text, then, is one which allows significant participation on the part of
the audience (participation that I am still willing to term “interactivity”) but participation
that does not exceed secondary levels of narration. The primary discourse—the
properties of the text and the underlying system of rules—remains fixed, and through the
emergence of play, the player participates in the creation of a derivative instance.

of possibilities dictated by the text’s designers. Epsen J. Aarseth, “Aporia and Epiphany in Doom
and The Speaking Clock: The Temporality of Ergodic Art,” in Cyberspace Textuality: Computer
Technology and Literary Theory, ed. Marie-Laure Ryan (Bloomington: Indiana University Press,
1999), 34.
324
To some degree, MMOs may mark the beginnings of a change in this regard, however we’ve
yet to see any significant examples of players influencing the underlying shape of the diegesis to
a degree which might challenge the notion of who authors the game.

383
In one manner or another, the videogame player traverses a fixed text—whether
literally traversing the land of Cyrodil or the Shadow Moses compound, participating in
the generative worlds of Civilization or The Sims, or abiding the rule-based spaces of
Tetris or Bejeweled. And the avatar is the figure that allows players to do so.
Because all these games offer fictional worlds—they are not real in any sense,
but merely artificial spaces projected by the text—the player cannot directly participate in
them. Instead, they must be offered an ergodic vehicle—a virtual presence (sometimes
tangible, other times implied) through which to experience the text. The mediating self of
the avatar thus operates as a vehicle in the diegetic space, allowing the player access to
the game’s tasks, and in this sense, the avatar is not so much a self-contained
individual, but rather an empty body in a narrative world, able to be inhabited by any
number of distinct personalities. The avatar becomes a means for activity, a device that
allows “doing” (hence videogames’ usual concern with action).
In this regard, the avatar is not so much a character as an “agent.” Writing in the
Poetics, Aristotle made a crucial distinction between agents—those who perform
actions—and characterization, arguing that agents are the only figures necessary to a
narrative world. For him, agents are the figures who do things, who propel the plot
forward. Characters, then, he describes as agents “of a certain kind”—agents given
characterization, agents possessing personality, identity, and traits extraneous to the
basic requirements of action.
325

In many circles, the notion of “agents” is regarded as an antiquated and overly
reductive notion, a remnant of formalist approaches that tend to treat characters as mere
textual products of what the plot requires of them. Chatman sums up the position:

325
Aristotle, Poetics, trans. Malcolm Heath (New York: Penguin Classics, 1996), 11.

384
Formalists and (some) structuralists…argue that characters are products of plots,
that their status is “functional,” that they are, in short, participants or actants
rather than personnages, that it is erroneous to consider them as real beings.
[Character study], they say, must avoid psychological essences; aspects of
character can only be “functions.” They wish to analyze only what characters do
in a story, not what they are—that is, “are” by some outside psychological or
moral measure.
326

This approach disregards any sense of individuality or existence, instead treating the
agent as a functional necessity of the text—a device rather than a being—and the flaws
of such an approach became apparent when confronted with modern narratives
dominantly concerned with characterization, where character could not be solely
accounted for by the necessities of plot. Yet for our purposes here, the formalist notion
of agents is actually quite appropriate.
As an ergodic vehicle, the avatar is a textual device more than anything, a figure
that we can reduce to its functional utility in the operations of the text. The avatar’s
functions are not only in service of the plot, or course—the avatar equally facilitates the
player’s ability to navigate the gameworld and “do stuff” in the game—yet these are also
textual processes, thus in either regard, the avatar is an apparatus serving a functional
purpose within the system of rules, a means to an end.
When a player first begins Super Mario Bros. 2, the choice between the game’s
four avatars—Mario, Luigi, Princess Toadstool, and Toad—is commonly determined by
those characters’ individual identities; with no other information to go on, players not
surprisingly choose whichever character they’d most like to be. Yet after minimal
familiarization with the game, players quickly realize that each avatar offers different
abilities—Luigi can jump highest, Toadstool can float for short periods of time—and

326
Chatman, Story and Discourse, 111. The canonical example of this approach is Vladimir
Propp’s study of Russian fairy tales, where Propp accounts of characters as mere products of
what the tale requires them to do “as if the differences in appearance, age, sex, life concerns,
status, and so on were mere differences, and the similarity of function were the only important
thing” (Chatman’s terms). Vladimir Propp, Morphology of the Folk Tale (Austin, TX: University of
Texas Press, 1968).

385
individual identities are quickly thrown out the window. As a fiction to be engaged, we
do indeed care who we are within a videogame, yet as a task to be accomplished,
identity frequently becomes a lesser consideration. This becomes especially apparent
when replaying—once a player is already familiar with the fiction, narrative immersion
often fades, making the game more of a task to accomplish than a fiction to experience.
Resultantly, the avatar takes an ergodic role, and in this role the protagonist becomes
empty, hollow, lacking in personality or distinction—they lack character.

A Psychological-Utilitarian Model of Avatars
In other words: as a character, the avatar is a fictional figure, while as a vehicle, it
is a functional one. And we might look at Max Payne in either manner.
In one respect, Max is a distinct individual developed through numerous markers
of identity and personality. He’s an undercover cop, a moody wise-ass, a soul
tormented by the murder of his family; he’s a figure who has changed over time, once a
happy and optimistic family man, yet now a cynical and melancholic loner; he’s a middle-
class, middle-aged, heterosexual white male living in modern-day New York. All of
these qualities shape a clear and specific individual—a character.
Yet equally, we can conceive of Max merely as the player’s ability to run, jump,
duck, dodge, and shoot. In this respect, Max is a textual product of what the game
requires of him, and a textual device for allowing the player access to the gameworld. It
is Max that allows the player to traverse the house in the opening flashback, or the New
York subway station that comprises the game’s first level; it is by using Max that players
can shoot the bad guys, open doors, and pick up items; and it is through Max that
narrative events unfold and players access the details of the story, allowing the plot to

386
move forward. In this regard, Max is a vehicle that accommodates players of all types—
not a unique person, but a formal device.
Max is someone that we can describe in personal terms: cynical, terse, driven,
righteous, fearless. And yet, while Max is developed far more than the average avatar,
players still seem to associate him most with the one action that makes him unique from
other avatars—his ability to use bullet-time slow-motion. So while Max features a great
deal of narrative characterization, functioning as a paradigm of traits, he is also an
amalgam of options and allowances for the player as they operate within the game’s
system of rules—a paradigm of abilities. Within the game’s fiction, we must address
Max as a unique and specific individual, yet within the game’s system of rules, we can
reduce him to a simple narrative function: “hero,” a role which can be easily inhabited by
any number of players, with little concern for the avatar’s uniqueness as a character.
We can divide these dual operations of the avatar into two realms—one
psychological, one utilitarian—and in many ways, this mirrors a very standard division in
traditional narrative theory: that between character and plot (or character and action).
While all narrative involves some degree of both, Tzvetan Todorov distinguishes two
broad categories of narrative, each placing a different emphasis between the two:
character-centered or psychological narration, in which “everything is subservient to the
psychology of the character,” and plot-centered or apsychological narration, in which
“characters are subservient to the action.”
327
The former mirrors the post-structural
conception of character—characters as unique and complex beings in their own right.
The latter instead reflects formalist notions of agents or actants as mere products of

327
Tzvetan Todorov, The Poetics of Prose, trans. Richard Howard (Ithaca, NY: Cornell University
Press, 1977), 68. Todorov actually discusses narrative (rather than narration), but since my
concern is with the manner of discourse rather than the complete product, I’ve taken the liberty of
reducing his terms accordingly.

387
what the tale requires of them, devices which propel the plot forward.
328
Rather than
positioning these as opposing types of narrative, however, we can deploy these as
different approaches to character—one psychological, one utilitarian. Whereas
psychological focuses on the experience, utilitarian focuses on the action; in the former,
actions become symptoms of personality, influences on development, or stimuli in role-
playing immersion; in the latter, they exist for their own sake, whether as pleasurable in
their own right, functions of the plot, or necessities of gameplay. To tackle avatars, what
we need is a combined methodology—a psychological-utilitarian approach—which
would address both of these roles.
329

As characters, we’d be vitally concerned with the differences between Gordon
Freeman (Half-Life), the Master Chief (Halo), Cate Archer (No One Lives Forever), and
Duke Nukem (Duke Nukem 3D), each videogame protagonists with different identities
and different personalities, and thus by no means interchangeable. Freeman is a
solemn MIT-educated theoretical physicist, Master Chief an all-business seven-foot-tall
genetically-enhanced super-soldier, Archer a sly and wise-cracking female superspy
from the sixties, and Nukem a hyper-masculine, politically-incorrect muscleman of the
Stallone and Schwarzeneggar variety. Yet as agents, the differences between the
four—each the avatar in a “first-person shooter”—is relatively negligible. All four function
in fairly identical fashions, allowing the player access to the gameworld via their
subjective point-of-view, a nearly identical range of projectile weapons, and an ability to
move by running, jumping, ducking, and strafing.

328
We often see this dialectic addressed in more casual terms, distinguishing character pieces
(melodramas, romances, coming-of-age tales) from action pieces (westerns, mysteries,
adventures); low concept stories concerned with emotion, psychology, and the human condition
(those in which “nothing happens”) from high concept tales driven by plot and action (those in
which lots of stuff happens).
329
Certainly, there may be exceptions—the paddle in Pong hardly qualifies as a character, while
the inhabitants of The Sims would barely meet the criteria of agents (as there’s no plot to
propel)—yet the vast majority of avatars combine the two.

388
So while narrative address dictates the avatar’s role in terms of where it positions
us (inside or outside), we’re more concerned here with how the avatar positions us. As
a character, the avatar positions us psychologically—it creates a mental or emotional
relationship between the player and the fictional context. In this regard, the avatar is a
narrative figure, and its functions serve the processes of narration, of projecting a
fictional world. We may be positioned internally, by assuming the avatar’s identity (role-
playing), or we may positioned externally, by relating to the avatar as a person separate
from ourselves, yet either way, the avatar operates as a live being. As a vehicle, on the
other hand, the avatar positions us spatiotemporally—it indicates our point of interaction
in the videogame, and provides a range of actions that we can perform there. In this
regard, the avatar is a textual figure, and its functions provide the player access to the
formal operations of the game. We may be positioned internally, so that we conceive of
ourselves as navigating the game’s events—feel as though we are performing actions in
a diegetic space—or externally, so that we conceive of the avatar as a functional
mediary that operates on our behalf within the formal system of rules (essentially, as a
gamepiece), yet either way, the avatar operates as an ergodic vehicle, a vessel that
provides us access to the game’s tasks. In either regard, then, the avatar operates as a
positioning device—it positions the player in relation to the videogame content.
These positions—psychological and spatiotemporal—are not always in sync,
however, and Max Payne provides a curious example.

The Schizophrenic Avatar
We’ve already seen an instance of this jumbled narrative address during the
game’s opening flashback, used to reveal Max’s tragic past. Frequently, this sort of
expository back story is treated in cut-scenes or some other third-person technique,

389
developing the avatar as an independent character separate from the influence of the
player, yet in this instance, the player is made to “play through” the flashback. We
navigate the house, discover the intruders, dispatch the bad guys, and discover the slain
wife and daughter all via Max, thus spatiotemporally, we’re positioned as Max himself—
addressed in the second-person, suggested to be navigating the house and performing
the relevant narrative actions. Yet the psychological experience of this flashback hardly
matches its functional operations.
As Max arrives home to discover intruders in the house and his wife and
daughter murdered, he is obviously meant to be very distraught. However, because the
player has just been dropped into this world, they likely have little emotional connection
to these characters or events, thus in order to convey the psychological experience, the
game must resort to having Max speak for himself while the player is controlling him,
essentially forcing words into the player’s mouth. As we navigate Max through the
house, we hear his emotional outbursts (“Michelle! Dear God, no!” and so on), a
technique that does not encourage us to identify ourselves as Max, but rather to
empathize with him. This is the sort of scene that makes us say, “Gee, I really feel bad
for him,” but in saying that, we’re implicitly acknowledging that he is not us. So while
we’re spatially positioned inside, as Max, we’re psychologically positioned outside, as a
spectator, and the narrational cues evince this. “You” run up the stairs, “you” shoot the
bad guys, and “you” discover the slain wife and daughter, yet “Max” has a bad feeling
upon arriving home, “Max” screams his wife’s name upon hearing gunshots, and “Max”
reacts in horror upon discovering his slain family. The first-person voiceover only
enforces the character’s position as separate from the player.
This sort of narration is not uncommon to the medium. Many videogames (I’m
tempted to say the vast majority of videogames) rely on these sorts of third-person

390
techniques of character development in order to convey emotion—show us a character
and make us empathize with their plight, rather than make us feel as though this is our
own plight. Yet while the experience is that of someone else, “you” are simultaneously
positioned as the person performing actions.
This largely occurs due to the “second-order design problem” of games. As
Salen and Zimmerman noted in chapter one, game designers cannot directly model the
behavior of players; they only design the rule-based context that gives rise to player
behavior. As such, character development poses a special problem for videogame
design: as a character, the avatar is very specific, while as a vehicle, it’s very generic.
Max is meant to be a very specific individual, with specific character traits, and yet, he’s
equally meant to be inhabited by millions of different players, all with their own unique
personalities and styles of play.
Such is the reason that what little character development might occur in
videogames typically happens in cut-scenes, fixed segments of narrative in which the
player is not inhabiting the avatar, not providing any input to the game, and thus not
necessarily identifying with the protagonist (particularly if the player is more concerned
with mastery of the game over a satisfactory narrative resolution). The problem this
poses for any sort of emotional or psychological identification is demonstrated in that
players can—and often do—skip the cinematics and yet still follow the basic action-
driven narrative (the goal is to get from here to there; the “why” established in the cut-
scene is unimportant). Unless somehow enforced by the system of rules, players have
no obligation to behave in accordance with any character trait established in such
scenes.
Resultantly, when avatars are “developed,” their characteristics are typically
attributes—strong, fast, intelligent, skillful—or proactive traits—courageous, inquisitive,

391
determined—but rarely anything complex enough to require a psychological
understanding of the individual, such as tormented, insecure, or desperate. Even role-
playing games—those which allow the player to shape their own unique avatar into a
role that they may inhabit, thus allowing the player to shape the character any way they
wish—still relies on skills and attributes over personality. Oblivion allows the player to
choose among attributes (strength, intelligence, agility), skills (athletics, marksmanship,
illusion magic), and other qualities such as race and profession which directly affect the
player’s abilities within the game, yet personality and psychology are given little heed.
330

The reasoning seems obvious enough—attributes and skills are the one area in which
character traits and ergodic abilities can overlap. The character’s “traits” essentially
become the things the avatar allows the player to do in the narrative world, so that
character development simply becomes exposition for why the player can perform
particular feats. Further, attributes and skills are more easily quantifiable within a
system of rules—the game can determine a particular level of strength, or of marksman
accuracy, and reasonably enforce them (as opposed to quantifying insecurity or
optimism). Yet these are hardly the sort of personal qualities that Chapman had in mind
when describing characters as a “paradigm of traits.”
Instead, Chapman’s model seeks to address characters as figures that—unlike
the “agent” model—cannot be wholly quantified, what he terms an “open” theory of
character. He wishes to view characters in the same manner we view people—as
complex, “round” beings, rather than “flat,” two-dimensional agents:
[A] flat character is endowed with a single trait—or very few. …Since there is only
a single trait (or one clearly dominating the others), the behavior of the flat
character is highly predictable. Round characters, on the contrary, possess a

330
Even the one quantifier of morality—fame and infamy points—are only dictated by acts
witnessed by the public and judged good or evil; one can play as a terrible murderer and still
steer clear of infamy, so long as no one sees you kill.

392
variety of traits, some of them conflicting or even contradictory; their behavior is
not predictable—they are capable of changing, of surprising us, and so on.
331

A rounded character is a psychologically complex figure with the potential for
unexpected actions, yet videogame protagonists must behave predictably because—
outside of third-person segments like cut-scenes—they only “behave” in the manner the
player chooses; they are incapable of being open-ended or surprising us. Further,
avatars rarely change traits; they don’t evolve and their personalities seldom change, as
such a change would be difficult to enforce within the player—could we imagine a World
War II game in which a courageous protagonist gets traumatized mid-narrative and is
suddenly filled with a fear of combat? How could the developers prevent the player from
wanting to charge into battle as usual? (Note that Max’s change in character—from
optimistic family man to cynical loner—occurs before the game.) In this regard, non-
player characters, the equivalent of supporting roles, actually have more potential for
development than the protagonist, as these are the characters whose traits can be
enforced.
The problem lies in Chapman’s distinction between events and traits. He notes
that events exist along the narrative chain, having strictly determined positions within the
story (“X happens, then Y as a result of X, then Z as a final consequence”), while traits
are not subject to such limitations, thus must exist outside the action.
Unlike events, traits are not in the temporal chain, but coexist with the whole or a
large portion of it. Events travel as vectors, “horizontally” from earlier to later.
Traits, on the other hand, extend over the time spans staked out by the events.
They are parametric to the event chain.
332

Yet as ergodic vehicles, avatars only exist within the narrative chain. As an agent, Max
does not exist outside his quest to clear his name and seek his vengeance. Like Propp’s

331
Chapman, Story and Discourse, 132. This echoes the work of E.M. Forster, from whom
Chatman borrows his terms.
332
Ibid, 128-129.

393
fairy tale heroes, he is merely the product of his actions. Therefore to instill him with a
trait like “lonely,” “happy,” or “nervous” is futile. While the game seeks to establish that
Max is tormented, there’s no method to instill this feeling upon the player inhabiting him.
The only way to enforce the trait is to incorporate it into the action: if the game’s
narrative establishes this trait as the reason he joins the DEA and goes on his quest for
vengeance against those who killed his wife and daughter, it then becomes a part of the
chronological chain; his torment occurs in the beginning, right before he begins his
quest. Logic would dictate that he would remain tormented throughout the game, yet for
the purpose of the game’s chain of events, there is no reason for him to be tormented
once the second event—going on the quest—occurs, so it’s still doubtful that the player
inhabiting the avatar of Max would inherit his trait of torment, concerned instead with the
immediate events at hand. The only traits “enforceable” by the narrative are those that
are inherent in every action, those ever-present in every narrative event, thus avatar
traits become proactive characteristics, as action is the very nature of the medium.
In order to maintain “torment” as a continuing trait, the game must then re-
establish it later in the chain of events—and the game’s manner of doing so is an
especially revealing gimmick. While “torment” is no longer needed as a motivating force,
the narrative evokes it later as a conflict which must be overcome via two flashback
dream sequences which revisit the game’s opening tragedy. Here, the torment that
previously acted as motivation has now found a new position in the chain of events, this
time as an obstacle; Max psychological struggle with his inner demons manifests for the
player as a need to escape his tormented psyche.
These sequences occur “in Max’s head” rather than any literal space. Both
feature a haunting soundtrack that combines sounds from the opening tragedy (a baby
crying, the screams of Max’s wife, the eerie, twinkling music of a child’s toy), as well as

394
warped visual perspectives—though the game is not played from Max’s POV, the game
warps the player’s perspective to replicate the subjective experience. But more
significantly, both levels take the form of mazes. Both begin with Max arriving home,
reminiscent of the opening discovery of the slain wife and daughter, however the
house’s hallways quickly turn into long, winding passages that loop and double back
upon themselves, then continue into a black, vacant limbo, where a thin trail of blood
beneath the avatar’s feet represent the pathways, and moving too far from that trail
results in a plunge into the empty nothingness below. The house’s furniture and even
the rooms themselves get rearranged to create a sense of disorientation, with visual
details such as wallpaper and picture frames switched between rooms to compound the
effect. So although suggested as a subjective emotional experience meant to immerse
the player in Max’s psychological position (rather than his spatiotemporal position), these
segments still rely on traditional plot-driven techniques of survival, escape, and spatial
progress. Further, despite occurring in Max’s head—not the diegetic world, bur rather
the spatial equivalent of “nowhere”—these segments must create a virtual space, a
maze, reducing the emotional “experience” to a simple quest, with the tropes of
subjectivity essentially added for atmosphere. In both cases, the player may, but need
not, feel any connection to the character of Max; the primary goals are still standard
objectives that any player can plug into. Psychologically, Max remains “someone else,”
while the player merely inhabits his body in order to progress from point A to point B.
Does this all mean that a psychologically second-person experience is a futile
endeavor? Should videogames simply accept that they are unable to deal in character
without relying on traditional first- and third-person techniques? Hardly.
In stark contrast to Max Payne, consider the second-person experience of
Planetfall, a hypertext adventure published in 1983 in which the player finds themselves

395
as the lone human on an uninhabited planet. Early in the story, the player befriends a
child-like robot named Floyd who follows the player throughout much of the game and
provides a great deal of comic relief. Floyd is an exceedingly endearing character, and
the player’s only companion on the desolate planet, and resultantly, players generally
become very attached to Floyd. So late in the game, when Floyd sacrifices himself in
order to save the player, players generally experience a significant emotional reaction—
a psychological experience—as they must now continue on without their companion.
Numerous players even attest to having openly wept at this moment.
Or perhaps we might consider System Shock 2. As the game begins, the player
awakens aboard a derelict space vessel overrun by mutants, and is soon contacted by
Dr. Polito, seemingly the only other human alive on the ship. Polito assists the player via
radio for a considerable portion of the game, guiding the player through a series of tasks,
yet upon arriving at Polito’s office, the player discovers that she has, in fact, been dead
for quite some time, and that the voice on the other end of the radio has been a
malevolent artificial intelligence manipulating the player’s every move. As one review
puts it: “You aren’t just scared, you feel betrayed and alone. Very alone.”
333
Again, the
player does not empathize with someone else’s sense of betrayal and isolation, but
rather experiences those emotions themself.
In both examples, emotional affect and psychological experience does not rely
on third-person techniques of developing a protagonist with which the player is meant to
empathize—in fact, neither game provides much of an avatar at all. Planetfall’s text-
based gameplay basically describes the environment and goings-on around the player,
and System Shock 2 merely provides an internal POV so that the player only sees the
avatar’s hands extended in front of them. Neither game provides any description of the

333
“The Top Ten Scariest Moments in Gaming,” Game Informer 174, October 2007, 36.

396
avatar, or even a name, much less any indicator of the protagonist’s personality or
emotional state. Rather than developing the avatar as an autonomous being (as does
Max Payne), they instead develop supporting characters that the player comes to know,
then uses those relationships to varying affect. In this manner, emotional reactions
occur organically from the player’s own experience. In Max Payne, we feel bad because
Max lost his wife; in Planetfall, we feel bad because we lost Floyd. The experience is
not happening to the narrator (“I”), and not happening to a character (“he” or “she”), but
is instead happening to the player (“you”).
Equally, the avatar’s spatiotemporal/utilitarian functions need not only occur in
the second-person, as we might compare Max Payne’s second-person spatiotemporal
positioning with numerous videogames that deploy “detached avatars” (for lack of a
better term) that the player consciously recognizes as mediating agents rather than a
marker of their own spatial positioning. Such is the case in The Sims, where players are
in no way suggested to be the titular beings of the game. Rather than control the Sims,
recognizing those bodies as indicating their spatial position in the gameworld, players
merely give commands to the Sims by clicking on them and creating a queue of actions.
Both psychologically and functionally, the Sims are always treated in the third-person, as
beings separate from the player. The player still interacts via those characters—the
Sims still function as ergodic figures that allow the player access to the game’s
operations—yet the player’s point of interaction is clearly coming through those figures
as mediating bodies, rather than allowing the player to seamlessly function within them.
Lemmings similarly offers third-person functioning, with players giving commands to a
troop of lemmings who blindly obey whatever orders they are given, and the player
charged with seeing the lemmings safely through a variety of treacherous levels. Similar

397
to The Sims, the player clearly does not inhabit the creatures, yet still “plays the game”
through them, making their utilitarian functions a third-person affair.
Videogame narration, then, is fully reliant on the use of the avatar as mediator,
and when the game addresses a player, it positions them, via the avatar, in at least one
of four ways:
(1) Identification (second-person psychological), in which the player experiences
a fiction via role-playing;
(2) Empathy (first- or third-person psychological), in which the player experiences
a fiction by relating to a separate character;
(3) Function (second-person utilitarian), in which the player locates their spatial
position in the game within the body of the avatar; or
(4) Mediation (third-person utilitarian), in which the player operates or commands
a game body that they recognize as separate from themself.
334

More often than not, our relationship to an avatar is some combination of all four.
These four options, however, are merely ways that we relate to an individual avatar. Our
embodiment in a videogame is actually much more complicated than this.

Frames of Embodiment: The Experience of Narration
In Grand Theft Auto: San Andreas, the player dons the role of Carl “CJ” Johnson
as he navigates several cities within the fictional state of San Andreas. CJ is the lone
protagonist of the game, and the sole character through which the entire game is
experienced; there is no portion of the game in which the player “leaves” CJ to inhabit a
different person. Yet this is not to say that CJ is the player’s only avatar.

334
Note that the avatar can only act as a sentient being as a character, thus first-person address
can only occur in that capacity of the avatar.

398
Playing as CJ, the player might enter a bar in the city of Los Santos wherein
they’ll find an arcade game called Go Go Space Monkey. If the player positions CJ in
front of the game and interacts with it, they will find themself playing the arcade game,
controlling a spaceship that must blast away oncoming enemies in a two-dimensional
side-scrolling parody of 1980s arcade classics like Defender and Gradius. And while the
player might feel as though they are playing Go Go Space Monkey and that the
spaceship is their avatar, this is only partly correct. In truth, CJ is still the player’s avatar,
while the spaceship is actually CJ’s avatar—an avatar for our own avatar. After all, CJ is
the one in the bar playing Go Go Space Monkey, not the player. And yet, functionally,
CJ has disappeared from the situation; for a moment, Grand Theft Auto: San Andreas
seems to fade away, replaced by Go Go Space Monkey, so that the player need only
concern themself with the tiny spaceship. Fictionally, the player still controls CJ, who in
turn controls the spaceship; functionally, the player simply controls the spaceship.
But even without the framing context of a videogame-within-the-videogame, we
see similar situations elsewhere in the game. When the player takes on the mission
“Supply Lines,” in which they must eliminate five targets with a remote-controlled
airplane (outfitted with a tiny machine gun), they’re similarly operating through two
different avatars—the player controls the RC plane, yet this activity is mediated by CJ
within the fictional context. Much like Go Go Space Monkey’s spaceship, the RC plane
is fictionally controlled by CJ, even if it is functionally controlled by the player, so that
even without the framing mechanism of a game within the game, the player still controls
a secondary avatar mediated by their primary avatar.
Yet why limit this to the remote-controlled instance? After all, isn’t this the
situation with every vehicle in the game? When CJ hops into a sedan and begins driving
around the city, the fiction suggests that the player is controlling CJ, and CJ is driving the

399
car. Functionally, however, the player seems to simply drive the car themself, with CJ
along for the ride. When the player presses the X button on their Playstation 2
controller, does X make CJ put his foot down on the accelerator, or does X merely make
the car accelerate? We’d have to say that the answer is both.
In all these cases, the player is operating an avatar for their avatar, instances
that provide clear cases of dividing the avatar’s character and vehicle functions.
Functionally, we merely navigate the “active” avatar—the avatar that we actually
conceive as doing something. But fictionally, we are still embodied in the primary avatar
(CJ), who in turn operates a secondary one (the spaceship, RC plane, or sedan).
Numerous games provide similar instances: Kameo: Elements of Power, Eternal
Darkness: Sanity’s Requiem, Blue Dragon, Black & White, and Okami all similarly
provide avatars for our avatars, positioning the player within one primary body (or in the
case of Black & White and Okami, a deity presence), but then affording secondary
avatars for that figure. And we can easily add driving games, RC missions, or any other
situation involving operable machinery to that list. In any of these instances, which
avatar do we identify as the mediating body?
As these many examples demonstrate, our experience of a videogame can
hardly be contained within a single avatar. While we can examine our relationship to
any given avatar by the tension between internal and external positioning described in
chapter three, videogames are frequently experienced not through a single embodiment,
but rather through multiple, alternating embodiments that operate together in a complex
framework to construct the experience of the game. And we need not only consider
avatars-for-avatars. Halo 2, Resident Evil 2, and Lego Star Wars all have us alternate
between multiple avatars at different times, playing one avatar at one moment, then an
entirely different character in the next. Baldur’s Gate, Earthbound, and nearly every

400
entry in the Final Fantasy franchise provide one clearly identifiable character as our
primary avatar, yet allow us to control numerous other characters in our party at the
same time. Wasteland, Freedom Force, and Rainbow Six don’t even bother with the
façade of a primary character, offering multiple avatars to control at once with no
preference or emphasis on one character over another. In any of these cases, we are
not merely positioned inside and outside, but in numerous contexts, each of which frame
our experience through different embodiments.
As an instance of narration, we do not receive the diegetic world of a videogame,
nor the events that happen therein, “all at once.” Instead, we receive that world
discursively, projected to us by the game-text through a continuous deployment of
information. There is always more information to be had, further data that has not yet
revealed itself, whether it be a dramatic twist of plot, what exists around the next corner,
or merely the move that our opponent will perform next. Equally, we do not receive that
data through a singular discursive context. Instead, narrative information is deployed
through multiple contexts, each providing a different circumstance of discourse—a
different frame of narration—through which we understand and engage the videogame.
Such frames are necessary to any narrative act, and narratologists have
recognized this phenomenon in narration of all sorts, using a variety of different terms to
describe it. Edward Branigan discusses the “levels of narration” at work in cinematic
fiction as a pattern that mediates the spectator’s understanding of the text;
335
Colin
MacCabe describes a “hierarchy of discourses,” each offering relative “truths” that we
combine in order to understand the plot;
336
Ben Brewster addresses similar concerns in
discussing the “hierarchies of relative knowledge” that result from changes in

335
Branigan, 86.
336
Colin MacCabe, Tracking the Signifier: Theoretical Essays: Film, Linguistics, Literature
(Manchester: Manchester University Press, 1985), 34.

401
viewpoint;
337
Gerard Genette discusses “narrative levels” contained within one another
that frame deeper levels of discourse;
338
Seymour Chatman describes the multiple
“parties to the narrative transaction”;
339
and so on. Each acknowledges, in one way or
another, that we do not simply receive narrative information through one consistent and
explicit account, but rather that narration involves varying degrees of information and
identification, occurring through multiple contexts. In any act of narration—fictional or
not, interactive or not—we are positioned by a variety of these frames, and the narrated
experience is a composite of those contexts.
We already addressed two of these frames earlier in this chapter by
distinguishing the author and narrator as separate figures, noting that the transmission of
information from the author to the audience is mediated by the discourse between the
narrator and narratee.

This provides us two different discursive frames—that outside the fiction, as well as that
inside. When Arthur Conan Doyle tells of Sherlock Holmes’ adventures, he does not
speak to his audience directly, but instead addresses them through Dr. Watson, who
tells his story to some invisible, unacknowledged listener within the fiction. Equally,
when Grand Theft Auto projects its player into the fictional state of San Andreas, it does
so not directly, but through the mediating avatar of CJ.

337
Ben Brewster, “A Scene at the Movies,” Screen, vol. 23, no. 2 (July/August 1982): 9. Cited by
Branigan, 74.
338
Genette, 227-8.
339
Chatman, Story and Discourse, 147.

402
Yet even these two frames of narration are not enough, so before addressing
videogame narration, we’ll first consider a simple narrative instance from our Hound of
the Baskervilles example to demonstrate. In this particular moment, the narrating
Watson relates a conversation that he had with a young boy who had been sent to
deliver a telegram, in which the boy explains that he had to leave the telegram with the
recipient’s wife, as the intended addressee was unavailable. In this instance, we see
four frames of narration at work:
• Sir Arthur Conan Doyle tells a story…
• …through the novel The Hound of the Baskervilles…
• …in which Dr. Watson tells a story…
• …in which a young boy explains the delivery of a telegram.
This diegetic event—the delivery of a telegram—is mediated by no less than four
discursive contexts, each identifiable by a different narrating presence.
The literal author initiates the discourse. This is Doyle himself, the actual person
who wrote the novel, and as we’ve noted, this is a distinctly separate figure from the
narrating voice of Dr. Watson, the presence we commonly know as the narrator proper—
the person who relates the fiction as a true and actual happening. Even when that
narrating presence lacks the distinct identity of Watson—whether an unidentified third-
person voice or a transparently documenting camera—it still exists as a separate context
from the authoring.
Yet between these frames lies a mediating context that frequently goes
unnoticed—the organizing principle that comprises the text, the most ambiguous and
variedly-discussed figure of the narrating situation. This is the figure that Chatman,
borrowing a concept from Wayne Booth, terms the “implied author”: “the agency within
the narrative fiction itself which guides any reading of it,” “the source of the narrative

403
text’s whole structure of meaning.” As Chatman points out, the real author retires once
the text is complete, yet “the principles of invention and intention remain in the text,” and
it is from those principles that the audience reconstructs the story.
340

He is “implied,” that is, reconstructed by the reader from the narrative. He is not
the narrator, but rather the principle that invented the narrator, along with
everything else in the narrative, that stacked the cards in this particular way, had
these things happen to these characters, in these words or images. Unlike the
narrator, the implied author can tell us nothing. He, or better, it has no voice, no
direct means of communicating. It instructs us silently, through the design of the
whole, with all the voices, by all the means it has chosen to let us learn.
341

Because textual discourse is mediated by a discursive object, Doyle is not present to the
audience, but rather exists only by implication of the novel (this is why we can still
receive Doyle’s story, despite his having died many years ago). Alternately, Watson—as
a fictional and thus nonexistent intelligence—cannot simply speak, but must be
preceded by a mediating presence that presents him; he exists only within the novel.
Herein lies the implied author—the formal system that produces the narration. For the
implied author, as Chatman concludes, “is nothing other than the text itself in its
inventional aspect.”
342

A text’s very organization implies a coherent discursive presence, and Chatman
suggests that we distinguish the implied and real authors by dividing what the text does
from what the author intends. The text becomes a self-existing thing independent of its
creators—a particularly useful approach for discussing videogames, considering that the
medium is chock full of emergent phenomena that may be unintended by its designers.
Further, the implied author solves the problem of “authorship-by-committee”—while
countless artists may work on a particular text, the very fact that the end result is
organized and deliberate attests to a single authorial presence—even if such a singular

340
Chatman, Coming to Terms, 74-5.
341
Chatman, Story and Discourse, 148.
342
Chatman, Coming to Terms, 86.

404
presence does not, in truth, exist. The implied author, as separate from the real author,
accounts for this sense of unified agency. In short, then: Doyle creates the novel, the
novel in turn creates Watson, and Watson tells the story.
These three entities—the author, the text, and the narrator—comprise the bare
minimum necessary for a textual instance of fiction. Fiction guarantees that the narrator
must be separate from the author; textuality guarantees that the context that sources the
narrator must be equally separate.
343
Yet we may also see additional, embedded frames
of narration from deeper within the text, as occurs in our Baskervilles example. These
are what Chatman terms “frame transmissions,” instances in which the narrator recounts
a separate telling within their own telling—a story within the story—as when the boy
describes the delivery of the telegram. Such episodes frame a separate narrational
context within the diegesis—which is in turn framed by the text, which is in turn framed
by the authoring act.
344

These narrating figures—the presence, intelligence, or organizing principle to
whom we can credit each context, the respective source for each discursive frame—
have been the traditional manner by which we distinguish frames of narration. And this
should not be surprising—these are, after all, the active figures in narration, the narrators
themselves. Yet as James Kinneavy told us in chapter one, every discursive exchange
must entail two participants—both a sender and a receiver—thus each narrator’s very
existence must entail an equivalent narratee, a listener existing within that same
discursive frame.

343
In other words, were one to simply tell a true story in person to a live audience, that narration
would not need entail an implied author (a text) nor a separate narrator (a teller to whom the
fiction is truth).
344
Chatman, Story and Discourse, 254.

405

In the Baskervilles example, only one of the embedded narrators explicitly evokes a
narratee, as we know that the boy is speaking to Watson.
345
However logic would tell us
that each sender must entail a separate receiver: Watson addresses someone (even if
we never learn who that figure is); the “implied author” of the text addresses an implied
reader (the hypothetical figure presupposed by the text); and Doyle communicates with
us, the reader (although only indirectly, through the novel). Baskervilles elides the very
existence of these receiving figures, disavowing the discursive situation so that the
narration seems to transmit as though to no one. While Watson, the text, and Doyle
himself are all clearly telling a story (conjointly, via their respective contexts), we’re never
given an impression of who they are speaking to, nor that we ourselves are specifically
being addressed.
In each discursive context, the narratee exists merely as an ambiguous presence
to whom the next frame is projected:
• In the real world, we’re provided a text (as our real selves);…
• …in the text, we’re given Watson as a narrator (as the hypothetical reader
assumed by the text);…
• …in the narration, we’re told of Watson’s encounter with a young boy (as
a fictional listener);…

345
And note that the diegetic Watson is a different figure from the one speaking, a past self of the
narrating Watson.

406
• …and within that encounter, we learn what happened when the boy
delivered the telegram (listening via the diegetic Watson—the Watson in
the story, rather than the one telling it).
If the notion of being “positioned” in these roles seems disputable, then you could
alternately consider that we receive the boy’s story in the same way that Watson hears
it, that we hear Watson’s story in the same way as his fictional listener, and that we
receive the text in the same way as its intended audience. Regardless of whether we
wish to conceive of ourselves as being those figures in any sense, we still receive that
narrative information through their vantage. By either model, our reception of that unit of
information (the telegram delivery) is mediated by all four frames of narration, just as
Doyle’s telling of it occurs through all four speakers, and aside from our actual selves
outside the text altogether, each narratee is—in one way or another—a proxy for the
audience.
In other words: each narratee, within each frame of narration, is an avatar.
The only difference between our fictional positioning as Watson’s narratee and
our fictional positioning as Max Payne or CJ is that in Baskervilles, the narratee is never
given deliberate embodiment. Because the narratee has no active role, and because
classical narration generally elides that figure, the fictional listener exists only by
implication of the discourse itself. Embedded narration becomes the only exception to
this rule because in such instances, the narratee is safely contained within the diegesis
and thus identifiable as a “he” or “she” rather than a “you” (as when Watson listens to the
boy’s story).
346
It is only in second-person narration that this figure is given a tangible

346
It’s telling that after 107 pages discussing narrators, Seymour Chatman chooses to address
“frame transmissions” in his brief, ten-page section on the narratee—traditionally, these are
among the scant few instances in which we can clearly identify that figure.

407
embodiment—existing not merely as an ambiguous presence, but as a deliberate
personage projected into the narrated content.
Second-person contexts are those that draw out and provide embodiment for the
audience, manifesting the narratee in much the same way as the first-person manifests
the narrator. Just as Watson projects himself into the story, existing in the narration as
well as the diegesis (the reason we see Watson in two frames, as a speaker in one and
as a listener in the next), and the young boy, equally, uses the first-person so that he
exists within the diegesis (the frame narrated by Watson) as well as the embedded
diegesis (the frame narrated by the boy, in which he delivers a telegram), so too second-
person narration provides a means for projecting the audience into the fiction, and as
such, the previously-elided narratee—along with their embodiment in the fiction—
becomes crucial.
347
When a Choose Your Own Adventure novelist employs the second-
person, the narrator gives embodiment to the receiving “you”—to the narratee. The
narratee, in turn, is a role meant for the implied reader (the intended audience of an
adolescent reader), and that role is one assumed by the actual reader. The actual
reader, then, gets projected through each frame, and into the diegesis as the fictional
protagonist of the story. Avatars perform the same service as that “you” pronoun.
So while videogames similarly entail multiple discursive contexts, our model must
privilege the narratees rather than the narrators—or more specifically, the embodiments
offered to the narratee—and in this regard, Edward Branigan will serve us well.
Branigan describes the levels of narration at work in a fictional text as a pattern which
mediates the audience’s understanding of that fiction, and rather than focusing on the

347
Max Payne, then, serves conflicting purposes—when the designers of Max Payne employ
Max as a narrator, they use Max as a proxy for their own storytelling, yet when they employ Max
as an avatar, they use him as a proxy for the audience. Resultantly, Max stands for both the
sender and the receiver of the discourse!

408
sources of narration (though he does discuss these) Branigan is more concerned with
the roles that the audience plays in these contexts.
Levels of a text are postulated in order to explain how data is systematically
recast by the spectator from one perceptual context to another. …Generally the
spectator engages the text in multiple ways, assuming a variety of roles for
different contexts at different times.
348

Equally, Susan Lanser—upon whom Branigan bases his model—explicitly addresses
the framing contexts of narration as a hierarchy of roles by which a reader participates in
a literary text.
349
For both, the information acquired through these roles is shaped into a
narrative pattern, and it is by integrating this data that the audience constructs (or
reconstructs) the diegesis.
350
And strikingly, when we conceive of these discursive
frames via the audience rather than the artist, they begin to appear quite similar to
models of gameplay. Numerous critics describe gameplay in terms of such parallel
contexts, whether Gregory Bateson’s concern with the meta-communication of play
(when something is real in one context, yet not in another) or Katie Salen and Eric
Zimmerman’s descriptions of the magic circle and the cognitive framing of play. Yet the
most markedly similar is Gary Alan Fine’s discussion of fantasy role-playing games, as
table-top RPGs (much like videogames) also project the player into a diegetic space
through the alternation of framed “selves.”
Fine relies considerably on the work of sociologist Erving Goffman, one of the
originators of the sociological concept of “framing”—a process wherein our perceptions

348
Branigan, 86.
349
Susan Sniader Lanser, The Narrative Act: Point of View in Prose Fiction (Princeton, NJ:
Princeton University Press, 1981), 131-48.
350
Christian Metz addresses these separate receivers in a different fashion, relying instead on
the metacommunication of fiction to make his point: “Since it is ‘accepted’ that the audience is
incredulous, who is it who is credulous and must be maintained in his credulousness by the
perfect organisation of the machinery (of the machination)? This credulous person is, of course,
another part of ourselves, he is still seated beneath the incredulous one, or in his heart, it is he
who continues to believe, who disavows what he knows.” Christian Metz, The Imaginary
Signifier, 72.

409
are influenced by frames of interpretation and reference—and views fantasy role-playing
in a manner similar to Branigan and Lanser’s approaches to fictional narration.
The world of fantasy gaming as a framed world is both typical and unique. It is
unique in the particular experiences it creates and the rules necessary to create
those experiences. At the same time it is typical of other forms of social life in
that it permits the rapid shifting of frames and requires the enactment of several
framed selves. …[The player] recognizes the existence of several selves that
must be juggled, hidden, or exhumed when appropriate.
351

Each of these frames, Fine describes, has a different set of meanings associated with it
and is defined by a different self, with the player’s experience resulting from an
oscillation between frames—not a far cry from our frames of narration. We will use a
similar model for addressing how videogames project us into their worlds.
Rather than address levels or frames of narration—which would seem to
privilege the figures who do the narrating—we will instead address the player’s levels of
embodiment. In chapter three, we addressed two “yous” in a videogame—the you inside
and the you outside. Here, we’ll examine a few more.
Every videogame will construct its own discursive or cognitive framework, from
the very simple to the head-scratchingly complex, and as such, I can’t realistically offer a
“universal” model of videogame embodiments (equally, the Baskervilles example was
convenient for providing one frame within one frame within one frame, though narration
is frequently much more complex than this). Yet as a starting point, we’ll consider four
basic embodiments, each corresponding to a different frame of narration, from which we
can construct our models.
We’ll start from the outside and work our way in, beginning with our outermost
embodiment in the game—as the audience.

351
Fine, 203.

410

This is the actual “you,” outside the game-text altogether and part of a mass audience,
the you free of artificial contexts. I call this the “audience” because this frame operates
before and outside of fiction or gameplay, outside of immersion or interactivity. This is
our position outside the magic circle, our true embodiment as a person engaging the
tangible artifact of the videogame itself. Before our embodiment in an avatar, before
entering the system of rules or participating in gameplay, we are first audience members
in the world at large.
Discussing RPGs, Fine refers to this context as the “primary framework”:
First, gaming, like all activity, is grounded in the “primary framework,” the
commonsense understanding that people have of the real world. This is action
without laminations. It is a framework that does not depend on other frameworks
but on the ultimate reality of events.
352

Rather than “audience,” Fine refers to this embodiment in more straightforward terms:
simply our role as “people.” For Fine, this is the inescapable frame of reality (our
“paramount” reality, as he puts it), our anchor of sanity that reminds us of the true nature
of our actions—that none of this is real.

352
Ibid, 186.

411
This embodiment is common to both players and non-players alike. Those who
merely watch someone else play the game, for example, and therefore find no further
embodiment within it, are still part of the audience; such a person’s involvement simply
does not exceed this level. In this frame, we do not act, nor play, nor participate; we
merely observe the game-text, a product that addresses all of us equally. At deeper
levels, we may have different experiences (and resultantly conceive of the game
differently), yet at this outermost embodiment, we are all part of the same mass
audience. Not every member of the audience actually plays, thus not every member
enters the “magic circle” and moves beyond their physical embodiment in the real world.
This marks a fundamental difference between traditional frames of narration and
videogame’s mode of address—while traditional narratives address a passive narratee,
a role that the passive audience can occupy, videogames address an active participant.
Non-playing audience members may still become invested in the content of the
videogame (just as cinematic spectators fixate on a film’s plot or sports fans invest in the
outcome of a game they are not playing), yet these are not the “implied readers” of the
videogame, not the playing audience presupposed by the game-text. Because the
game-text addresses an active player (a role that we’ll address within the next frame),
the passive audience is not “embodied” in later frames. While such an audience
member may invest in a fictional context (much like a cinematic spectator) or appreciate
the stakes of a contest (much like a sports fan), they do not themself enter into those
contexts, remaining instead in the real world outside. Before and outside of
participation, then, we are first passive observers, mere outsiders observing the contexts
within.
In this, we can see the unique position a videogame player is given. When we
observe a traditional narrative, the audience is clearly separate from the characters, and

412
equally, when we watch a sporting event, the fans remain separate from the players.
Yet a videogame player is able to sit outside the screen, watching from a position of
privilege as events unfold within, even as they themself participate in those events.
Observing the game through the screen, we are positioned as spectators even while
addressed as participants—we are audience to our own actions. It’s not surprising,
then, that story-based games rely so frequently on cinematic cut-scenes, or that sports
games replicate the tropes of a television broadcast, or that racing games feature
replays from multiple angles—these are all replications of traditional audience address,
positioning the player as audience even while putting them inside those contexts.
This is also the level at which we receive the meta-communication of play, the
implicit statement that none of this is real. While deeper levels attest to some alternate
reality, whether the game situation (“I must score a goal”) or the fictional context (“I must
save the world”), as audience, we are always aware of the un-reality of the game.
Further, at this level, we are informed not only by the game-text itself, but by peripheral
information sourced outside the game entirely. We bring in our cultural knowledge, our
awareness of the world, our familiarity with other games, and our expectations and
assumptions. We may be familiar with the game from past playings, word-of-mouth, or
walkthroughs downloaded from the internet. We may be familiar with the conventions of
the genre, or influenced by the game’s ad campaign, or swayed by the excessive hype
of a blockbuster title. As an audience, we process excess information which is not
inherent to the game itself, but rather obtained from the world at large—in the real world,
the game-text is but one source of data. We cannot simply “shut off” knowledge rooted
outside the magic circle, and it is here that we might account for its influence.
When we first move inside the game, then, we become a player.

413

This is the “you” that has entered the magic circle, absorbed into the artificial context of
gameplay.
353
Here we do not yet enter into the diegetic world, operating instead within
the constraints of a system of rules that we accept for no reason other than the play they
provide. The player is the embodiment that actually “plays the game,” that interacts with
the formal rule-set. This figure is not trying to rescue the princess, defeat the villains,
find the treasure, or save the world; this figure instead attempts to win the game.
Fine describes this framework as the game context, a context separate and at
odds with the world the audience knows to be real:
Second, players must deal with the game context; they are players whose
actions are governed by a complicated set of rules and constraints. They
manipulate their [avatars], having knowledge of the structure of the game...
Players do not operate in light of their primary frameworks—in terms of [reality]—
but in light of the conventions of the game.
354

353
Note that in my earlier diagram of the videogame situation I used “player” to describe the
figure outside the videogame, whereas here, I use it inside the game; this is the unfortunate result
of using the term in slightly different ways, yet not wishing to unnecessarily coin new terms. In
the former context, I used “player” as the generic term for the videogame audience (as opposed
to spectator, reader, theatergoer, etc.). Here, however, I use “player” to distinguish a particular
role which that person dons (as opposed to their role as audience, or as a fictional character).
354
Fine, 186. Fine uses the term “characters” rather than avatars, as his work precedes the
contemporary usage of the latter term, yet his account of role-playing characters happily
accommodates the notion of an avatar.

414
At this level, the player does not inhabit the avatar, but rather manipulates it from
outside; the avatar is a tool of the interface, a means of activity, an ergodic vessel
providing a utilitarian function. Rather than enter the diegesis, the player operates at the
systemic level, bound by the rules that comprise this context and aware of the
videogame’s formal properties. Much like a basketball player obliged to the court’s
boundaries or a chess player aware of the rules that govern their knight, the player
consciously operates within a rule-set. Yet although the player does not enter the
fictional world, playing does still entail an act of pretend.
Roger Caillois describes “make-believe” as one of the essential characteristics of
this context, describing play as “a special awareness of a second reality, or of a free
unreality, as against real life.”
355
While the audience operates within the primary
framework of reality, the player instead operates in this “second reality,” a framed
context apart from the first. This is the domain of the play-ground, the “temporary world
within the ordinary world” that Johan Huizinga describes as distinct and bounded from
“ordinary” life by virtue of its arbitrary rules (and frequently by spatial delineation as
well—the arena, the card-table, the tennis court, or, in our case, the screen).
356
It is
demarcated by the “cognitive frame” that Salen and Zimmerman describe, providing a
new context for interpretation and asserting its own unique meaning structure. “As a
player steps in and out of a game,” they tell us, “he or she is crossing that boundary, or
frame, which defines the game in time and space,”
357
echoing Gregory Bateson’s claim
that “play occurs within a delimited psychological frame, a spatial and temporal bounding
of a set of interactive messages.”
358
When a player crosses that boundary, they take on
a new role, a new embodiment; they become part of the make-believe. To become a

355
Caillois, 10.
356
Huizinga, 10.
357
Salen and Zimmerman, Rules of Play, 370.
358
Bateson, 191.

415
player is to enter the space of play and accept the fiction that “this activity matters,” to
accept an alternate reality where scoring points, staying within boundaries, or advancing
to the next level is of consequence. To play is to accept the rules as “real,” counter to
the audience which acknowledges its unreality (or the character that is unaware of the
rules at all). The player is the figure who operates in this second reality, and the
metacommunication of play is the acknowledgment of these two separate realms—in
one context this activity matters, while in another, it does not. As such, our position as a
player is a role common to every game, digital or not.
In many games, however—specifically, those in which the player does not need
an avatar, for they themself are the one performing the activity of the game—we could
end right there. The divide between audience and player—between a person in reality
and their role in this second reality—might be the only boundary of concern. And we
might make a similar case for certain videogames, particularly puzzle games—games in
which the third context is so abstract that one could not conceive of their activity as
participation in any manner. In Tetris, our role as player seems the deepest embodiment
we get, for while the individual blocks might abstractly be referred to as “avatars” in that
they are the player’s identifiable point of interaction—ergodic components that allow the
functionality of gameplay—we’d be hard-pressed to think of ourselves as embodied
within them.
I ought to clarify that the player is the one position for which “embodiment” may
not be the best choice of terms, as, quite to the contrary, the player is rather bodiless.
While we can point to the avatar as a separate body in a different world, our role as
player operates in an artificial context, yet without any tangible presence. If pushed,
we’d have to say that the player is still that person sitting in front of the screen, yet when
operating as a player, we frequently enter some manner of disembodied limbo. The

416
player immersed in Tetris is by no means identifying themself as a block, yet nor are
they paying particular attention to their own real world position. As occurs with many
manners of media, the participant absorbed into the text leaves their body behind and
manifests…well, nowhere. Just as when we talk on our cellphone, plug into our iPod,
open a web browser, turn on our TV set, or settle into a movie theater’s darkened
auditorium, we leave our bodies behind and enter a disembodied elsewhere, with little
sense for where exactly that is. When we play Tetris, we enter into the game, yet
equally, seem to exist nowhere in particular.
In the vast majority of videogames, however, we’re given at least one further
frame beyond the game context—that of the fiction, the diegetic world which we
encounter via an embodiment in the avatar.

This is the “you” embodied in the videogame protagonist, the role-playing you. This
figure has not only entered the second reality of gameplay, but the third reality of the
diegesis. Role-playing occurs beyond the system of rules; embodied in the avatar, we
do not merely watch, nor do we play; rather, we are. While the player interacts with the
game, becoming subject to the artificial constraints of the rules, the avatar is subject only

417
to the internal reality of the diegesis. As an avatar, we simply exist in the videogame
world.
Fine cites this embodiment—which he terms “character”—as that which
distinguishes table-top role-playing games from other manners of game, and in this
regard, we can see the relationship videogames hold to fantasy role-playing (even those
that we would not strictly consider as RPGs).
Finally, this gaming world is keyed in that the players not only manipulate
characters; they are characters. The character identity is separate from the
player identity. In this, fantasy gaming is distinct from other games. It makes no
sense in chess to speak of “black” as being distinct from Karpov the player
(although one can speak of Karpov the player being distinct from Karpov the
man). The pieces in chess (“black”) have no more or less knowledge than their
animator. However, Sir Ralph the Rash, the doughty knight, lacks some
information that his player has (for example, about characteristics of other
characters, and spheres of game knowledge outside his ken such as clerical
miracles) and has some information that his player lacks (about the area where
he was raised, which the referee must supply when necessary). To speak of a
chess knight as having different knowledge from its animator might make for
good fantasy but not for meaningful chess.
359

Though the term was not in wide use at the time of his writing, Fine is describing an
avatar—a figure who is an embodiment for the player in one context while equally
distinct from the player in another. Yet while Fine describes this third frame as unique to
fantasy gaming, videogames see avatars used in genres of all sorts. There are
exceptions; much like chess pieces, Tetris blocks function primarily as units to be
manipulated rather than units to be inhabited, each offering equipment of a sort, parts
which are utilized during the course of the game, and resultantly, it’s difficult to locate
ourselves any deeper than the position of player. Yet in the vast majority of cases, the
player must inhabit an actant of some sort within the diegetic world, and this is not solely
an issue of knowledge, but equally of presence.

359
Fine, 186.

418
Much like Branigan, Fine bases his account primarily on knowledge and
information—on the discursive frames in which we can source particular data, and the
discrepancies between them. This is important to us as well, however knowledge is not
the only marker of embodiment in a videogame; presence is of equal concern, an issue
rare to non-digital games (short of fantasy role-playing) and only relevant to narration in
second-person instances.
In a game of tennis, we can distinguish the tennis pro-as-player from the tennis
pro-as-person, much like Karpov in chess. And of course, there’d be no logic to
distinguishing the player’s racket as a separate figure, entity, or embodiment from the
player— just as chess pieces, the racket is merely equipment, an object that is
manipulated during the course of the game. Yet the paddle in Pong—a fair equivalent to
a tennis racket—is not mere equipment. Instead, it is a point of interaction. Much like
my NBA Live example from chapter two, the space of play is no longer a physical place
that the player can occupy, but is instead a projected space—a situation narrated by the
text—and as a result, the player must have an additional presence beyond their mere
participation in the game; they must also be projected into the game’s space. The player
(in the second reality of the game) cannot perform the action on their own, but must
instead rely on a virtual presence (in the third reality of the gameworld) to do so for them.
While the tennis player’s racket is merely an extension of their playing self, the paddle in
Pong—by existing in a different discursive frame—is a presence in which the player
must locate themself, just as the avatar in NBA Live. The avatar is thus the discursive
agent charged with acting, so that while the player plays, the avatar is the figure who
actually does.
In many cases, we may be able to simply stop here. With these three, we have a
real person, as part of the audience, who is in turn positioned both externally and

419
internally—externally as a player, and internally in the role of the avatar. These three
embodiments provide us a basic model of videogame reception, one that can adequately
describe many instances of videogame play. Yet numerous videogames go further by
adding deeper embodiments—embedded avatars.

An embedded avatar is a “you” that is mediated by a more primary avatar, a you within a
fourth context embedded within the gameworld. This is an avatar for your avatar, and
hypothetically, we could have an endless string of these figures—an avatar for an avatar
for an avatar for an avatar indefinite. Much like embedded narrators, these figures are
simply embodiments projected further into the diegesis.
360

Embedded avatars can occur in several manners. In the most obvious instance,
they are the result of frame narration—a story within the story, a game within the game,
as when San Andreas’ CJ plays Go Go Space Monkey: we control CJ, CJ controls the
pixilated spaceship. Yet embedded narration is not the only context through which we

360
While Fine limits himself to three levels of meaning, and only offers a deliberate labeling for
those first three schemas, he does acknowledge embedded instances in fantasy gaming,
describing them as “a keying from our status as characters, which in turn rest[s] on the keyings of
the gaming framework, further based upon a keying of the primary framework of the players”
(“keying” is a term Fine takes from Erving Goffman, describing a situated alternative reality based
upon a different “key,” or a different reality or context of meaning—in other words, an embedded
fictional context). Fine, 185.

420
get embedded avatars—these also occur anytime our primary avatar takes control of
another person, vehicle, mechanism, or otherwise operable body, framed not by a
separate level of narration, but rather by mediated action: we control CJ, and CJ flies the
RC plane. Or we control CJ, and CJ drives the sedan. The secondary avatar may exist
in the same reality as the primary one, yet our embodiment is still framed by an
embedded context.
Of course, the player’s embodiment in an embedded avatar can often be much
more tenuous than these examples—Civilization’s units are all embedded avatars, yet
they’re framed as merely obeying orders given by Abraham Lincoln or Julius Caesar,
and similarly, the player is never given direct control over the divine pet of Black &
White, relegated instead to molding its behavior from behind the deity avatar. In both
cases, the player remains situated in a first-level avatar while merely giving commands
to some embedded being. Conversely, the embedded avatar may be the only one
foregrounded to the player so that the primary avatar all but disappears. We see this
situation in Okami, where—in somewhat of an inversion of Black & White’s divine pet—
the player controls a sun goddess’s avatar, with only tenuous embodiment in the
goddess herself. Here the player controls the white wolf Amaterasu in order to navigate
the diegetic world, and while this is the only tangible body that the player is given, the
game makes it clear that Amaterasu is an avatar of the sun goddess, who resides on the
celestial plane. However, while the player never navigates the “celestial plane” and is
never provided any sense of the sun goddess as a separate being, they’re frequently
required to pause diegetic time to use the “celestial brush,” a feature that turns the
gameworld into a black-and-white canvas upon which the player can paint, allowing
them to add objects, slash through enemies, and otherwise assist the wolf. The wolf,
then, is an embedded avatar, and the diegetic landscape an embedded context, as

421
these instances are brief removals to the primary avatar’s position—the deity outside the
embedded context, existing on the celestial plane anterior to the wolf—and the celestial
brush is our evidence. Yet few players would conceive of their activity as controlling a
goddess, who in turn controls the wolf. Instead, the sun goddess becomes conflated
with her embedded avatar so that the wolf seems the only figure of concern.
361

For the sake of comparison, we can take our Go Go Space Monkey example and
look at it in much the same manner as the young boy’s story in Baskervilles:

By this model, we see that videogames provide frames of narration in much the same
manner as traditional narratives—they simply privilege the opposite end of the discourse
(in this instance, the outer space context is the content being transmitted from the
designers, so that while the audience feels as though they are simply flying the
spaceship through outer space, this situation is actually mediated by each frame). Note
that while in Baskervilles, the embedded narratee was the only receiver with explicit
presence, second-person narration inverts this situation so that the embedded narrator

361
A comparable, but very different example, is in Burnout: Paradise, a racing game that takes
place in the highly detailed and strenuously photorealistic open-world of Paradise City, a city that,
strikingly, seems to feature no people. One of the defining traits of the Burnout racing franchise is
its use of spectacular crashes as a recurrent component of the series, with players encouraged to
knock other drivers off the road and into oncoming obstacles, yet to avoid the logical violence that
this would entail—that people are dying left and right each time a vehicle careens of cliff or hits a
pylon at 100mph—the game’s designers were very deliberate in choosing not to feature citizens
walking the streets, nor drivers in the cars, in order to elide the consequences of such horrible
collisions. So while logic would tell us that the player must be embodied in a driver behind the
wheel of the car, who in turn is driving the vehicle—making the driver our avatar and the car an
embedded avatar—the game tries to blur the player and driver by suggesting that the player is
simply driving the car, and that no actual people exist within the diegesis. Resultantly, it’s unclear
(deliberately so) whether our avatar is our vehicle or the driver of that vehicle.

422
is the only one clearly evoked. While we can identify each receiving position—a
spaceship, which is an embodiment for CJ, which is an embodiment for our playing self,
which is a role assumed by our real self—the entities that project those embodiments
are relatively invisible, aside from the Go Go Space Monkey arcade cabinet. Yet to
account for the entire situation, we would have to identify a separate narrating presence
for each frame:
• The San Andreas designers project the real you…
• …into a playing self via San Andreas’ system of rules, which projects that
playing self…
• …into Carl Johnson via San Andreas’ diegetic world, which projects CJ…
• …into a spaceship via Go Go Space Monkey.
Alternately, we could identify the second and third frames as the videogame-as-referee
(the figure that enforces the system of rules) and -as-dungeon master (the figure that
projects the diegetic world), respectively. Similar to literature, these distinctions are
clearest in hypertext games, where we can distinguish the narrating voice as a separate
presence from the code of the game-text , so that the designers (frame 1) create the
code (frame 2) which produces the narrating voice (frame 3) which projects any
embedded contexts (frame 4). Yet graphics-based games still operate in the same
manner, with the game-text producing the rules and representations which in turn foster
our sense of the diegetic world. Thus each frame projects the next—the designers
create the system, the system creates the world, the world creates the embedded
context, and so on.
Keep in mind that we experience all of these contexts at once, even if our
attention may be drawn to one over another. As Branigan clarifies, a “crucial principle of
narration” is that at any given time:

423
several levels of narration will be operating simultaneously with varying degrees
of explicitness and compatibility; that is, the spectator may describe the text in
several different ways, all of which may be accurate, each with a particular
context and for a particular purpose.
362

We do not simply move from one discursive locus to another, but rather experience them
all at once, and resultantly understand the game situation in multiple manners (or better,
via all of them). When playing Go Go Space Monkey, we do not move from CJ to the
spaceship, but rather, we are put in four contexts at once—as a spaceship, as CJ, as a
game player, and as an audience member—so that the experience of the game is an
integration of all of our embodiments.
This allows us to once again revise our model of the videogame situation, this
time attending to our three primary embodiments:

Though the “player” is a much less visible figure than the avatar (or, perhaps better, the
player and audience are harder to distinguish from one another than from the avatar),
this gives us a more accurate model of videogame “playership,” acknowledging the
separate contexts by which one engages a videogame. Note that when I addressed
“internal” and “external” positioning in chapter three, I used those phrases in relation to

362
Branigan, 96.

424
the avatar, distinguishing the subject’s position inside and outside the diegetic world, the
“third reality.” If we wished, however, we could apply this model to any frame of
embodiment, as anytime we’re positioned outside of one frame, we’re inside the one
prior, thus while our activity as a player occurs “external” to our embodiment in the
avatar, we can always step further back, outside the game altogether, and recognize our
role as “player” as an artificial position inside the game (just as a chess player
concerned with capturing their opponent’s king can equally step back and recognize that
outside the game, this activity doesn’t really matter). As a player, we’ve already entered
into the game-text, moving from mere audience to a participant in a ludic context, and as
such, the player’s position outside the diegetic space is still contained within the game.
This gives us a telling description of how the videogame player is “immersed” within the
game—or more to the point, the multiple frames through which immersion occurs.

The Borders of the Frame: Immersion, Reflexivity, and the Fourth Wall
As we’ve seen in numerous instances throughout this dissertation, this notion of
“immersion” often seems the Holy Grail of videogames—complete and utter absorption
into the game-text. Popular discourse suggests as much:
To me, the biggest disappointment would be if the game doesn't immerse me into
the setting…
363

– forum user xHappyLunaticx

Commercial discourse suggests as much:
Madden NFL 09 aims to deliver the most immersive gameplay, the richest online
offerings and the most comprehensive feature depth in the history of the
franchise.
364

– Madden NFL 09 product description

363
xHappyLunaticx, comment posted on Insurgency Forums, posted on June 19, 2007,
http://forums.insmod.net/index.php?showtopic=4523&st=0&p=86466&#entry86466.
364
Product description for Madden NFL 09, IGN.com,
http://xbox360.ign.com/objects/142/14229536.html (accessed July 12, 2008).

425
And industrial discourse frequently suggests as much:
EAX technology is awesome! It gives us the tools we need to fully immerse the
player in our gaming worlds.
365

– Steve Pearsal, project director at Looking Glass Studios

This idealization of immersion is perhaps best evinced in Janet Murray’s evocation of the
Holodeck as metaphor, the fantastically seamless VR environment featured on Star
Trek: The Next Generation which Murray claims as “a new stage for the creation of
participatory theater.”
366
Whether we prefer Star Trek’s Holodeck, William Gibson’s
cyberspace, the Wachowski brothers’ Matrix, or merely Lewis Carroll’s Alice tumbling
down the rabbithole as our metaphor of choice, this transportive effect of an alternate,
virtual reality—the sense of being physically here but psychologically there —seems the
ideal that everyone expects from their videogames.
We should not be surprised; immersion has long been considered a vital
component to both fiction and gameplay, from Samuel Taylor Coleridge’s “willing
suspension of disbelief” to Johan Huizinga’s absorption into the “temporary world” of the
play-ground, so one would expect a medium which traffics in both to treat immersion
with such hallowed reverence. Yet if this is the case, how do we explain chapter three’s
example of Metal Gear Solid? As we saw there, Solid combines strident realism,
extensive character development, gripping suspense, and a sophisticated narrative—all
hallmarks of narrative immersion—with striking moments of reflexivity and
acknowledgment of the player, techniques which deliberately rupture immersion
(defeating Psycho Mantis by unplugging the controller, finding a radio code on the CD
case). This chapter’s example of Max Payne—also a narratively immersive game that

365
Creative USA Press Release, "Creative Environmental Audio Technology Now De Facto
Industry Standard for Immersive 3D Gaming," Creative USA (August 3, 1999),
http://us.creative.com/corporate/pressroom/releases/welcome.asp?pid=6214.
366
Murray, 125.

426
relies on believable characters and emotional investment—also includes its own
reflexive moment, when Max finds a letter during one of the flashback dream sequences:
There was something disturbingly familiar about the letter before me. The
handwriting was all pretty curves.

“You are in a computer game, Max.”

The truth was a burning green crack through my brain. Weapons statistics
hanging in the air, glimpsed out of the corner of my eye. Endless repetition of the
act of shooting, time slowing down to show off my moves. The paranoid feel of
someone controlling my every step.

I was in a computer game. Funny as hell, it was the most horrible thing I could
think of.

Despite all of its immersive appeal, Max Payne here flaunts its artificiality, reveling in its
status as text while at the same time enveloping the player within its diegesis. And we
could continue to list countless examples of such reflexivity in other videogames.
In such instances, we see the boundaries of the diegetic context ruptured, and an
outside frame of reference allowed to seep in. This is the breaking of the “fourth wall,” a
transgression of the frame’s self-contained internal reality. Reflexivity has long been
associated with the rupture of immersion; by foregrounding the artificiality of the text, it
violates the fictive illusion and resultantly breaks the immersive spell. Brechtian
narratives often use reflexivity for the specific purpose of acknowledging their own
artificiality, subverting the transparency of classicism and drawing attention to their
constructed nature. Comedic films use reflexivity as a sudden acknowledgment of
absurdity, to confront a fictional situation with reality for the humor of that contrast.
Postmodern texts deploy reflexive tropes to celebrate their style and technique, to revel
in their very textuality. Yet in any of these cases, reflexivity breaks the fictional context
by violating the diegetic boundaries that, for immersive purposes, have traditionally been
treated with reserved sanctity.

427
There is a certain logic to videogames’ embrace of the reflexive. As a second-
person medium, videogames require a significantly different conception of the fourth
wall, as such texts, by their very nature, must address the audience. In order to position
the audience within its bounds, the fourth wall must remain open to some degree so that
the audience outside can participate within. Unlike the interface of the cinema screen or
the literary page, where the audience merely observes and the fourth wall is treated as a
one-way mirror (in that the audience can look in yet characters cannot look out), the
interface of the videogame necessitates a more porous boundary. Videogames require
a two-way exchange across this boundary, demanding a more liberal treatment of the
fourth wall, and the frivolity with which videogames transgress the fourth wall evinces
this.
367

The “illusionism” that we generally consider requisite of immersive fiction has
long been reliant on a diegetic world that is complete, self-contained, and clearly
bounded from both the real world and the discursive context (the second frame, the
frame of narration). The transparency of the classically realist text attests to this
practice, eliding the discursive process—the very projection of the fiction—in order to
maintain the illusion that the projected world simply “is,” to maintain the “poetic faith” of
which Coleridge speaks. For as Marie-Laure Ryan points out, an immersive story-world
relies on a clear sense that the internal world exists independent of the telling.
Immersion is the experience through which a fictional world acquires the
presence of an autonomous, language-independent reality populated with live
human beings. For a text to be immersive, then, it must create a space to which

367
This is exactly the conundrum that Marie-Laure Ryan attacks in Narrative as Virtual Reality by
trying to compromise immersion with interactivity. She deals with the issue in a much larger scale
than ours, well beyond videogames alone, and as a result, her terms are often different and her
conclusion reached by different means, yet ultimately, she is dealing with this very issue—
reconciling the closed boundaries implied by immersion with the porous boundaries requisite of
interactivity. Yet Ryan gives only passing heed to videogames, dealing instead with hypertext,
interactive movies, virtual reality, and other computational media, so while we can use many of
her ideas, we’ll proceed in a significantly different direction.

428
the reader, spectator, or user can relate, and it must populate this space with
individuated objects.
368

Although the fiction is projected by the discourse, it must have the appearance of
existing independent of it, separate from the words or images through which we perceive
it. By maintaining a coherent and cohesive internal reality, we retain the impression that
the diegetic space continues beyond the frame or the page, allowing the story-world to
exist as a self-contained reality, and our willing suspension of disbelief relies on such
self-containment. The integrity of the frame that bounds the diegetic context, then—the
“rigidity of the ontological boundary that separates story-worlds from physical
reality”
369
—is essential to diegetic immersion. It is for this reason that the Brechtian
strategy of “verfremdungseffekt”—the distancing effect, the inverse of immersion—is so
often characterized by a reflexive transgression of the diegetic frame, breaking the fourth
wall and rupturing the internal reality of the diegesis. By allowing the discursive frame to
spill into the diegetic context, the fiction loses its illusion, allowing the real world to come
crashing in with it.
Addressing the nature of play, Johan Huizinga makes similar observations by
stressing the importance of bounding the play context, noting that “inside the play-
ground an absolute and peculiar order reigns, …the least deviation from [which] ‘spoils
the game,’ robs it of its character and makes it worthless.”
370
Much like the sanctity of
the diegetic world’s bounds, the ludic frame is a vital barrier, delimiting the play context
from the world outside and allowing no transgression. Here, it is the rules themselves
that circumscribe the immersive context rather than the fictive world of the diegetic
context, yet this context proves just as fragile, for “as soon as the rules are transgressed,

368
Ryan, Narrative as Virtual Reality, 14-5. Italics are my emphasis.
369
Ibid, 105.
370
Huizinga, 10.

429
the whole play-world collapses. The game is over.”
371
Similarly, Roger Caillois
describes the dangers that the “contagion of reality” pose for play, stating that when “the
universe of play is no longer tightly closed” and resultantly “contaminated by the real
world,” a “perversion” results: “The principle of play has been corrupted.”
372
It is for this
reason, as Huizinga explains, that society has long disavowed the spoil-sport over the
cheat:
The player who trespasses against the rules or ignores them is the “spoil-sport.”
The spoil-sport is not the same as the false player, the cheat; for the latter
pretends to be playing the game and, on the face of it, still acknowledges the
magic circle. It is curious to note how much more lenient society is to the cheat
than to the spoil-sport. This is because the spoil-sport shatters the play-world
itself. By withdrawing from the game he reveals the relativity and fragility of the
play-world in which he had temporarily shut himself with others. He robs play of
its illusion—a pregnant word which means literally “in-play.”
373

Much like Brechtian practice, the spoil-sport reveals the fiction for what it is—shatters the
artificial context and draws our attention back to reality, disallowing our engrossment into
that alternate frame of reality.
374
The cheat may break the rules, yet they still attempt to
maintain play’s illusion, to remain inside the play context; the spoil-sport instead
disregards that context, an even worse transgression.
375
So even in ludic immersion, the
division between the gameworld and the outside world must be maintained—even play
needs that illusion.
In both fiction and gameplay, then, there is a sanctity to the “fourth wall,” a
sacredness to the boundaries between contexts that forbids transgression. Both rely on
their very illusion for their efficacy, on our ability to suspend our disbelief, and thus by
extension, on the integrity of the boundary that divides them from other contexts. So

371
Ibid, 11.
372
Caillois, 45.
373
Huizinga, 11.
374
It is for such reasons that transgressive artists who “rain on the parade” of immersive fiction
are at times described—partly in jest, yet with an ounce of truth—as spoil-sports.
375
Not so coincidentally, filmmakers often refer to minor violations of continuity or diegetic reality
as a “cheat” (such as a “cheat cut,” or “cheating” the staged distance between two objects),
indicating the hope that no one will notice, that this violation will not spoil the fictive illusion.

430
how do we compromise the videogame medium’s apparent fetishization of immersion
with the distanciation effect of reflexivity, the direct address necessary of second-person
narration, and the frivolity with which videogames break the fourth wall?
We need first clarify what exactly we mean by “immersion,” as the term’s
buzzwordiness has led to a wide variance in definition—some too broad, some too
narrow, and as with any good definition, ours should be somewhere in between. In truth,
we often deploy “immersion” to reference several different phenomena that frequently
get conflated, and even in its specific application to videogames, it is used to address at
least two different manners of psychic investment.
In its most common usage, immersion entails being projected into a fictional
space, whether the experience of being consumed by a story or the engagement of a
virtual environment—what we will call diegetic immersion. Some will insist we
distinguish these as separate phenomena—“narrative immersion” as opposed to “spatial
immersion”—yet both unquestionably involve projection into a diegetic space of some
sort, and as such, there are grounds for the shared use of the term. In fact, Marie-Laure
Ryan goes to great lengths to connect these two phenomena, drawing innumerable
connections between the literary notion of being “lost in a book” and the transportive
experience of VR.
376

In a videogame, diegetic immersion is the experience of being wholly occupied
within the diegetic frame. This is probably the most common implication when
discussing “immersive” play: role-playing, assuming the identity of an avatar and feeling
as though we ourselves are navigating the diegetic space, suspending our disbelief and
participating in a fictional context, feeling as though we’re the one trudging across some

376
Ryan, 89-114. This connection is made throughout the book, yet these pages are particularly
direct in that assertion.

431
fictional landscape. As the avatar, we are inside the world of the game, where we
attempt to rescue the princess, defeat the villains, find the treasure, or save the world, as
simply “winning the game” has no meaning here. Diegetic immersion is the willing
acceptance of a fictional context, what some will call the “illusionism” of fiction. When
we traipse through the world of Tamriel in Oblivion, accept the identity of Gordon
Freeman in Half-Life, worry about stopping the terrorists in Metal Gear Solid, or react to
the murder of Max’s family in Max Payne, we are immersed in the diegetic frame.
Diegetic immersion, then, may entail both spatiotemporal positioning (feeling as though
we are there) and emotional or psychological investment (feeling as though fictional
elements are of real concern).
377

However, diegetic engrossment is not the only manner by which we might be
immersed. Other manners of psychic investment, if not “immersive,” certainly seem very
close to it: when we become so engaged in an activity (work, sport, craft, artistry) that we
“lose track” of the world around us; when we become so consumed by our thoughts that
we “zone out,” stare off into space, and somehow fail to see things right in front of our
eyes; or when we become so invested in someone’s company that we stop and wonder
where the time went. Much like being “lost in a book” or navigating a virtual
environment, these equally involve a loss of conscious attention to the world around us,
yet do not involve absorption into an alternate reality. And in the case of videogames, if
nothing else, we’d need account for Tetris. Any player who has found themself playing
Tetris—or Hexic, or Bejeweled, or any other such puzzle game—can attest that such
games can be just as consuming as Oblivion, Half-Life, Metal Gear Solid, or Max Payne,

377
Recalling the two functions of the avatar, we could phrase these as being psychologically
immersed via a fictional character or spatiotemporally immersed via an ergodic vehicle.

432
yet we’d be hard-pressed to say that such players are being “transported” to a fictional
world of falling blocks.
The player of Tetris is indeed immersed in the videogame, yet not in its diegetic
frame. Rather, they are immersed in its systemic frame—in the game’s ludic context,
what we’ll term ludic immersion. Ludic immersion is the experience of engrossing
gameplay—when we become consumed by the very task of the game itself. Salen and
Zimmerman note: "When we play a game, we feel engaged and engrossed, and play
seems to take on its own ‘reality’…[b]ut it is an engagement that occurs through play
itself.”
378
Rather than a diegetic reality (which they actually dismiss altogether), they see
the player of a game engrossed in the “second reality” of which Roger Caillois speaks—
the world recast by the rules of the game—concluding this as the context which
characterizes ludic investment:
It is possible to say that the players of a game are “immersed”—immersed in
meaning. To play a game is to take part in a complex interplay of meaning. But
this kind of immersion is quite different from the sensory transport promised by
[diegetic immersion].
379

Here, winning the game does have meaning—it may even be the only meaning. When
we look at puzzle games like Tetris or Bejeweled, it would be difficult to explain the
situation in any context other than the playing of a game; unlike narrative games where
we can distinguish systemic components (power-ups, save points, avatars) from their
diegetic implementation (first-aid kits, telephones, characters), we cannot translate Tetris
to mean anything beyond the play itself.

378
Salen and Zimmerman, Rules of Play, 451.
379
Ibid, 452. It should be noted that Salen and Zimmerman actually dismiss diegetic immersion
as what they call “the immersive fallacy”—as they describe it, the false belief that the immersion
of gameplay relies on a suspension of disbelief and the illusion of a simulated reality. However,
this dismissal seems largely the result of their focus on the entirety of all games, both traditional
and digital. The vast majority of non-digital games (barring fantasy role-playing games) deal only
in the ludic frame, and as a result, anyone looking to games as a super-category would be
inclined to dismiss diegetic immersion. Although I disagree that we should dismiss diegetic
immersion altogether, their argument is a valid one, so long as we frame is as addressing ludic
immersion in particular.

433
This is not to say that ludic immersion does not involve the diegetic frame,
however; the diegetic world can certainly remain relevant to gameplay without
necessarily being the context in which we are immersed. The science fiction backdrop
of Space Invaders is vital to understanding the game, and certainly a part of its pleasure,
yet we’d be hard-pressed to argue that the player is being positioned in a world being
overrun by aliens. Instead, the sci-fi context helps situate the player’s activity while they
remain engrossed by the play itself, much as the medieval context of chess remains
relevant to the game, though no one becomes engrossed in that context when playing.
Whereas “internal” games absorb us into their avatars, providing a diegetic context in
which the player is made to feel a part, “external” games—those in which the player
remains outside the diegetic frame—instead situate us within a ludic context, the literal
playing of the game. This is the experience of playing endless hours of Tetris then
suddenly wondering “where did the time go?”; of playing Pac-Man not because we
identify with the titular avatar, but because we are compelled toward the very activity of
controlling him; of revisiting Grand Theft Auto: San Andreas not to advance CJ’s story,
but rather to unearth the multitude of hidden discoverables scattered about the
landscape. In each instance, it is not role-playing, fictional investment, or diegetic
absorption that drives us, but rather the very act of playing.
Such ludic immersion is perhaps nowhere better described than in David
Sudnow’s Pilgrim in the Microworld, a lengthy account of his growing obsession with the
Atari classic Breakout in which he describes the “cathexis” of videogame play:
Down to three bricks, the closest I’d come, a serve yet to go… Atari had me
hooked. I’d said that before, but this was a whole different business, nothing like
I’d known… Thirty seconds of play, for three bricks, and I’m on a whole new
plane of being, all synapses wailing as I’m poised there with the paddle, ball, a
few remaining lights on screen, and a history that made this my first last brick. …

Was I hooked? “I’ve been trying to reach you all day, were you out?” they’ll be
asking. No way. Not me. Not for that kind of thrill. …Meanwhile, next morning I

434
was back at it. As soon as I got out of bed I glanced at the darkened TV set and
flashed on an image of those last three bricks. …I woke up not eight hours later
and I wanted a fix, so I plugged myself in with the first cup of coffee, stuffed to
the gills with electric anticipation.
380

Much like the Tetris player’s relationship to their falling blocks, we’d have a hard
time saying that Sudnow is immersed in “being the paddle,” yet the fixation and
obsession that he details certainly suggests a similar engrossment. His account
becomes even more telling as he describes the pleasures of mastery and the
investment he develops beyond a mere interest in winning:
[T]he object was mine for a while, I was doing the playing for a change, precisely
managing a spate of action in that way that characterizes all competent conduct:
knowing just where you’re going to go and then going there. It’s becoming an
instrument. Instantaneously punctuated picture music. Supercerebral crystal
clear Silicon Valley eye jazz.

Again a big change in the significance of “playing the game.” I’m directing the
ball just here and just there, playing the circuit breaker as I’ve wanted from the
start, not that I knew it was that sort of tool. And it’s improvised. …

Do I really care about winning by this point, about those points, the clock, what a
“good player” can or can’t do? …You sit and play a home cartridge by yourself.
Get where you can manage the affairs well enough to pull off a tolerably decent
game in terms of objectives in the instruction book. …Then stray from the official
objectives and get a little into creating patterns as an end in its own right apart
from “winning.”
381

Sudnow’s entire book is telling of ludic immersion, yet these sample passages provide a
fair summation of the psychic investment Sudnow places in Breakout, along with the
liminal pleasures that emerge even from a game this simplistic and abstract. His
account seems the perfect example of the “seriousness” of play, a phenomenon
described by Johan Huizinga:
The consciousness of play being “only a pretend” does not by any means
prevent it from proceeding with the utmost seriousness, with an absorption, a
devotion that passes into rapture and, temporarily at least, completely abolishes
that troublesome “only” feeling. A game can at any time wholly run away with the
players.
382

380
David Sudnow, Pilgrim in the Microworld (New York: Warner Books, 1983) 56-60.
381
Ibid, 191-5.
382
Huizinga, 8.

435
Such is the nature of ludic immersion.
To bring these two manners of investment together, then (along with the several
other phenomena which may or may not be immersion), we will consider immersion as
being consumed or engrossed within a context. More specifically, immersion is the
phenomenon of being engrossed in a context to the point where more immediate
contexts fade from attention. We ought to remember the origins of this word, before its
metaphorical turn: to be submerged, as in water. In a videogame, immersion is
movement to a deeper embodiment—being positioned in an interior context such that we
willingly “forget” the narrative act (the artificial projection of that context) and temporarily
accept one of these interior realities as our own—whether the second reality of play, the
third reality of the diegesis, or a further reality of an embedded context. Taking its literal
meaning into account, Marie-Laure Ryan notes that “for immersion to take place, the text
must offer an expanse to be immersed within.”
383
We might consider such “expanses”
as fiction (a narrative context), thought (a mental context), conversation (a social
context), work (an active context), play (a ludic context), or another reality (a spatio-
temporal context), yet all immersion occurs within a context, such that our conscious
being is submerged in one particular frame of reference at the expense of a more
primary one (at the very least, the immediate here and now).
In a videogame, then, immersion is directly tied to our frames of narration—the
means by which the game-text projects us into a context separate from the here and
now, whether a fantastic landscape in a faraway world (the diegetic frame) or the mere
activity of being subject to a system of rules (the ludic frame). When we discuss
immersion in videogames, we’re addressing our engrossment in one of the game-text’s
interior contexts.

383
Ryan, Narrative as Virtual Reality, 90.

436

When positioned internally, we are immersed in the diegetic fiction; when positioned
externally, we are immersed in the play of the game. So for our purposes, we’ll use
these two fundamental frames to distinguish the primary manners of immersion with
which we need be concerned, yet keep in mind that these manners of immersion can
apply to embedded frames as well.
384

With this model in mind, we can begin to see why reflexivity, while rupturing to a
point, does not necessarily violate the entirety of videogame play.
Before we proceed, however, I should clarify that not all videogames invoke such
deliberate transgressions of the fourth wall; while reflexivity does seem a rampant
practice, there are a large number of videogames which overcome the diegetic rupture
of second-person address by veiling their second-person nature, thus eliding the
discursive situation and maintaining diegetic immersion. Such games do so by
appropriating the self-contained transparency of classical third-person narration and
subtly conflating this with second-person address.

384
The embedded context of Okami is still a diegetic one, as players invested in the wolf avatar
(rather than the celestial deity) navigate a diegetic space much like any other, while GTA: San
Andreas’ embedded context of Go Go Space Monkey—the abstract game within the game—
relies primarily on ludic investment, as the player (much like that of Space Invaders) likely has
little concern for the goings-on of the outer space context, instead compelled to the very act of
“playing the game.”

437
Fictional narratives commonly reference elements of the “real world” outside their
boundaries, however they do so not by pointing outward to the actual thing in question,
but rather bringing those components inside their own diegetic space. When we watch
Casablanca, the film is able to reference such real world elements as the Nazi party and
the Second World War—elements which exist outside the film—without damaging the
fragile membrane of its fictional world because the film brings those elements into their
own world. When Rick sees a German officer at the door, he’s looking not at a real
German outside the film, but a fictional mock-up meant to evoke the real thing, one
safely contained within the diegesis. Equally, then, while the audience does not exist
within the world of a videogame, the game can reference the player by pointing not to
the real player outside the game, but rather a fictional mock-up meant to evoke that
person—the avatar. Avatars allow videogames to address the player without breaking
the fourth wall because the avatar becomes a fictional proxy, allowing the game to
implicitly address the audience while explicitly maintaining the veil of transparent third-
person narration. Tomb Raider is able to suggest “this is happening to you” by explicitly
saying “Here is Lara Croft, and this is happening to her” while carrying the unspoken
implication “…and you are Lara Croft.” In this manner, the narration merely presents
“someone else,” someone inside the diegetic space, with only extra-diegetic cues
making explicit reference to the player outside, thus any instance of second-person
address is simply directed to the fictional avatar, safely contained behind the fourth wall,
requiring no explicit rupture of the diegetic boundary. Rather than allow discourse
sourced within the diegesis to address the audience outside the game, this mode
maintains the effacement of the audience, keeping any second-person address
contained by the illusion of being a diegetic participant. And not surprisingly, the most

438
internally immersive of videogames—those like Half-Life, Bioshock, and Oblivion—tend
to avoid reflexive tropes and maintain their diegetic immersion.
Our concern here, however, is with those videogames that deploy reflexivity,
disregard diegetic reality, and otherwise neglect the barrier between the diegetic and
ludic contexts with such casual regularity, and yet still remain “immersive” experiences.
And with our model of embodiments in mind, we can see why such games are able to do
so: because the rupturing of the diegetic frame can still be contained within the ludic
frame.
In “breaking the fourth wall,” we often assume that the frame we are breaking is
the only one separating the audience from the fiction; but this is not the case. As we
saw in chapter two, narration entails a “double framing,” what the Russian formalists
terms the fabula and syuzhet, the story and the narration. Prior to the fiction is the
context of telling—we agree to the contract of being told a story and enter the discourse
(the systemic frame), and within that context, the diegetic world is further framed. The
systemic frame is the very activity itself, the experience of engaging a text or of playing a
game, and the mediating context between the real world and the diegetic one. This is
the magic circle—not the fiction itself, but the context before it, the frame of agreeing to
the fiction in the first place.
In classical narration, the diegetic frame is held as the most sacred; the telling is
but an added detail, one deliberately made transparent so that the projection of the
fiction can go unnoticed. Here, reflexivity is particularly rupturing because the fiction is
the whole point of the text, and merely drawing attention to the discursive context is
usually enough to rupture the fictive experience and draw attention to its artificiality. This
is to be reminded of the fiction’s constructed nature, thus breaking the spell of the
diegetic illusion, and we might apply that same logic to diegetically immersive games.

439
Yet in many videogames—as in reflexive texts—the systemic frame is privileged, and in
these instances, where the very act of engaging the text is made significant, we may still
remain compelled by the textual experience, despite ejection from the diegetic context.
Although Brechtian reflexivity is often associated with alienation and a
confrontational style, such self-conscious viewing or reading is not entirely alien to
pleasurable enjoyment of a narrative text. Yet rather than look to Brecht, modernism, or
the avant-garde, where aesthetes can appreciate the experimental transgressions of the
self-conscious text, we need look no further than The Rocky Horror Picture Show, trash
cinema, midnight movies, and the pleasurably “bad” film, all telling examples in which
the audience revels not in the fiction itself, but rather in the experience of the text. The
pleasures of such films are in the telling, not the believing; instead of suspending their
disbelief, such audiences celebrate it, remaining outside the fiction and entertained by
the ridiculousness of the thing itself. The genres of horror, musicals, and slapstick
comedy all utilize a similar effect, as the fictional content of these genres would be
exceedingly disturbing or nonsensical were we not to distance ourselves from the fiction
to some degree (hence the long history of reflexivity in comedies and musicals, the
excessive tastelessness common to slasher films, and the sense of spectacle common
to all three).
385
All these instances entail a metatextual pleasure—a removal from the
diegetic situation to the textual experience itself.
386
The celebration of form and style in
postmodern narration has equally led to a loosening of diegetic immersion and the pure
sense of willing disbelief (as in the films of Quentin Tarantino, Baz Luhrman, or Tim

385
For example, Kendall Walton addresses this distancing in horror in order to argue against the
willing suspension of disbelief; though we can dispute her dismissal of suspended disbelief, her
position on the reception of horror is still telling. Kendall Walton, “Fearing Fictions,” Journal of
Philosophy, vol. 75, no. 1 (January 1978), 5-27.
386
In fact, many people who “don’t get” musicals, horror, cult films, and Brechtian texts are
confounded by this very quality, assuming diegetic immersion to be the only point of a narrative
and unaware of this distanced, systemic appeal.

440
Burton), and even many of the deliberately self-conscious narratives of the modernist
school do so for the very pleasure of the telling just as much as the transgression of the
diegesis, from Henry Fielding’s Tom Jones to Woody Allen’s Annie Hall. In any such
case, the point is not only in our acceptance of the fiction, but also in the discursive
experience of receiving it.
As Robert Stam describes, “reflexive texts inscribe the reader/spectator within
their own rhetorical space”
387
—rather than eject them from the text altogether, they
position the audience within the discourse, within the context of the telling, and in doing
so form an alternate context that runs parallel to the diegetic content.
Reflexive artists often foreground the narrating instance, and in doing so call
attention to the reader or spectator. They favor the I-You of “discourse” to the
He-She-It of “histoire.” Rather than narratorless fiction, where no one speaks
and “events seem to tell themselves,” reflexive fiction models its discourse on
human conversation, from which it borrows its manner of expression. …The
presence of the reader or spectator is inscribed and signaled within the text,
shifting the interest from the diegesis to what Geoffrey Nowell-Smith calls the
“intersubjective textual relation,” which comes to form a kind of parallel plot.
388

This “intersubjective textual relation” is the relationship between the spectator and the
text, the contract of the discursive exchange. And in a game, this is the context of the
player.
Much like reflexive reception, traditional gameplay proceeds at the systemic
level. Still within the magic circle of play, the enjoyment of baseball, chess, Monopoly, or
poker all results from the ludic adherence to rules for their own sake, for the pleasure of
gameplay, rather than any sense of diegetic immersion.
389
After all, no one believes the
player of chess to be consumed in a medieval context, or the player of Monopoly to be
engrossed in the role of real estate baron, and similarly, while games like Half-Life,

387
Robert Stam, Reflexivity in Film and Literature: From Don Quixote to Jean-Luc Godard, 2
nd
ed.
(New York: Columbia University Press, 1992), 153.
388
Ibid, 149.
389
Hence Salen and Zimmerman’s argument on “the immersive fallacy.”

Salen and Zimmerman,
Rules of Play, 450-5.

441
Resident Evil, and Final Fantasy all privilege diegetic immersion, others ranging from
Tetris to The Sims to Metal Gear Solid address a ludic player just as much as—if not
more than—a diegetic avatar. When a videogame practices self-conscious tropes, it
does not eject the audience from the game altogether, but rather—much like Stam’s
description—positions them in the systemic context: as a discursive receiver of the
fiction, as a player playing a game.
It is for this reason that videogames enjoy such rampant use of reflexive tropes,
liberal deployment of metatextual elements, and casual disregard for diegetic logic.
While all such practices readily break the diegetic barrier, they still remain contained by
the systemic wall—the ludic frame of play. Zork, Space Quest, Monkey Island, Metal
Gear Solid, Max Payne, and Grand Theft Auto all break the fourth wall in one way or
another, yet they still maintain the magic circle, for while the barrier between the avatar
and player may be broken, we need remember that the player is still a projected
embodiment: the real audience member recast by the game’s system of rules. The
activity of playing the game is still bound by the magic circle—the frame between play
and the real world—and so long as that barrier is not ruptured, play can still remain an
immersive experience.
Gameplay has always proceeded with an awareness of the game; before the age
of videogames, players’ needed to be aware of the systemic frame if, for nothing else, to
know the rules which they needed to obey. In fact, Bernard Suits sees this as a defining
characteristic of games: that players voluntarily accept inefficient means for attaining a
goal—that they consciously obey arbitrary rules, simply for the sake of the activity they
make possible.
390
While spectators and readers have become accustomed to a
transparency of form, players have only recently been able to enjoy this luxury, as in the

390
Suits, 34-41.

442
past, a transparency of rules would result in ludic chaos—without deliberate and visible
rules, no one would know proper game behavior.
So in a videogame, reflexivity takes us out of one context merely to immerse us
in another. In describing fantasy role-playing games, Gary Alan Fine observes the
“oscillating character of engrossment” in that players slip in and out of contexts, sliding
amongst frames, and this happily matches Branigan’s description of the different roles
an audience plays—not merely positioned as a narratee, but continuously alternating
between contexts. Even though we oscillate amongst frames, we still remain within the
magic circle; reflexive tropes merely draw attention to the fact. Diegetic transparency
and reflexivity, then, are techniques for moving the audience into two different contexts,
yet both potentially immersive contexts. We may be immersed as a player (in the
game), as an avatar (in the diegesis), or even as an embedded avatar (in an embedded
context), but never merely as an audience—never outside altogether. The ludic context
is always framing the game, even when the diegetic context breaks down, so even when
we reflexively fracture the game’s fictive nature, we’re still playing the game, still in one
of the videogame’s interior contexts. It is only when the ludic context breaks down that
the game is threatened—when frustration with a poor interface sets in, when the
difficulty becomes too great, when a fellow player “ruins the game,” or when some
component of the text deeply offends us. It is then that we may break from the ludic
context altogether, allowing the real world to come crashing in and the fragile boundary
of which Huizinga speaks fails us. At this point, in Huizinga’s words, “the whole play-
world collapses. The game is over.”

443

Conclusion

As my introduction discussed, many “serious adults” in our society still harbor
assumptions much like Roger Ebert’s, that “the nature of the [videogame] medium
prevents it from moving beyond craftsmanship to the stature of art…worthy of
comparison with the great dramatists, poets, filmmakers, novelists and composers”
because videogames are inherently different from such established artforms. The
assumption seems to be that great art is what we already know to be art (or perhaps,
even worse, what textbooks and critics tell us to be art), and so to join the ranks of the
elite, videogames must challenge the reigning champions on their home turf and beat
them at their own game—they must match cinema at being cinematic, literature at being
literary, poetry at being poetic, theater at theatricality, painting at being painterly, and
sculpture at sitting on a pedestal.
391

Yet this is a foolhardy assumption.

391
I should clarify that when such critics use the term “art,” what they mean to reference is “High
Art”—those works that someone or other deems as having value or aesthetic merit over other,
lesser works. These discussions center on subjective evaluations of worth or quality. When I use
the term, however, I use it in a much more objective fashion, without the baggage of aesthetic
worth—to discuss “videogames as art” is merely to discuss its potential as a medium for
expression, as opposed to media that merely transmit information (such as email, letter-writing, or
the telephone).

444
Centuries of aesthetic and narratological criticism have taught us the value of
medium-specificity—of attending to the formal qualities specific to a given artform rather
than expecting all arts to operate equally. Media critics have long recognized the
importance of medium-specificity when examining discourse, comparing media not only
by their similarities, but also in terms of how one medium conveys information differently
than another. Judging new forms by the standards of the old is a rigged contest—no
text can stand up to such a loaded comparison. In the previous century, cinema was
probably the best example of these fixed matches. In its infancy, the cinema found itself
compared to numerous media—especially theater and literature—and judged by the
existing standards of those arts; not surprisingly, cinema was found inferior each time.
Whether judged by theatrical standards (mise-en-scene, live performance, audience
address) or by literary ones (introspection, abstract thought, authorial voice), the cinema
was considered vastly inferior to the legitimate arts of the day.
392

Critics were not the only parties guilty in the fix, mind you. Early filmmakers (as
well as their audiences) often looked at the cinema as an extension of previous media,
particularly theater. The theatricality of early films—each scene photographed in a
consistent wide-shot, taking in the entire “stage”—testifies to this assumption. Yet
filmmakers and critics alike soon realized that cinema had its own medium-specific
qualities: cinematography and editing. Using these tools, filmmakers broke out of the
theatrical mold and began to move the camera, began to cut between shots to break up
a scene, and critics and audiences equally began to expect a different mode of

392
In fact, we call them “movies” for just this reason—the term was originally coined as a
denigrating one, used to infantilize an inferior medium. “Movies” were those childish little
amusements comprised of moving pictures; the term “cinema,” on the other hand, held much
more esteem. After all, would literature seem nearly as distinguished if we called them
“wordies?” The term “videogame,” and the association of games with children, has created
similar problems, precisely the reason that the videogame industry refers to “interactive
entertainment” and academia refers to “new media.” Sounds much more prestigious than
“videogames,” doesn’t it?

445
discourse than the theater. When the cinema was reevaluated under cinematic
standards (editing, camerawork, photographic realism), the medium was suddenly able
to hold its own against other media—as it ended up, the cinema was art after all!
Certainly, the cinema utilizes many components that do exist in other manners of
discourse—set design and performance mimics the theater, voiceover narration is
largely lifted from literature’s narrating voice, music is itself another medium—yet if we
were to only focus on the qualities that it shares with other media, we would severely
limit our discussion. To ignore the discursive properties specific to the medium would be
to overlook the very things that make the medium unique—how could one discuss the
way in which cinema conveys information without addressing the closeups and wide-
shots of the camera, or the shot/reverse shots and cross-cutting of editing?
Similar battles played out decades later after television arrived, but by this time,
cinema was a reigning champ, so the fix went in its favor. When TV was judged by the
standards of cinema…well, you can guess the outcome. And history seems destined to
repeat itself, as videogames have entered the public consciousness in much the same
manner, judged by the standards of its predecessors.
In looking at videogame discourse, we’ve all-too-often been seduced by its
similarities to previous media, especially cinema. Just as the artists, audiences, and
critics of the young cinema could not look past its theatrical properties, leaving them
blind to its medium-specific potential, so too videogames have been overshadowed by
its similarities to past forms. As a medium invested in audio-visual representation
delivered via a rectangular screen, many of us seem unwilling or unable to look past
cinematic values and expect videogames to be anything but “interactive movies,” and as
a result, we insist on discussing them as such. Those who try to discuss videogames
otherwise typically do so merely by applying a different archetype—comparing their

446
visual design to painting, their stories to literature, or their gameplay to previous
manners of game. Note the vast range of videogame criticism that has given in to this
centuries-old oversight by focusing on visual and auditory style, discursive subtext,
authorial marks, narrative complexity, and other issues that have direct correlations to
previous forms. Neither side of the “videogames as art” debate are less guilty than the
other—the medium’s proponents have done it (“Call of Duty is so cinematic!” “Shadow
of the Colossus is so painterly!” “Final Fantasy VII offers such a sophisticated
narrative!”), and the medium’s detractors have done it (“Videogames lack the aesthetic
sophistication, narrative complexity, emotional affect, and/or authorial control of my
favorite artform!”). The problem is the enchanting ease of such comparisons—we
already have the theoretical and conceptual models for such discussions leftover from
the artists and critics of the past. Further, the expectations that audiences bring to new
media have already been shaped by their experience of previous media, building an
assumption that new forms should do similar things. It is for this reason that new media
throughout history have generally been taken up first by young audiences—unlike older
generations whose expectations have been firmly shaped by the standards of past
forms, young audiences lack such longstanding preconceptions, and as a result, are
more likely to accept new media on their own terms.
393

Past art has often moved in new, creative directions as a result of discovering (or
rediscovering) its medium-specific potential in the face of some competing artform.
Painting moved into impressionism and expressionism partly as a result of
photography’s rise—the medium could not compete with the photo-realism of
photography, so artists moved toward manners of expression that photography could not

393
This phenomenon has also lead to the recurrent fear of new media that arises century after
century, as young people indulge in some new artform that their parents’ generation sees as
divergent from the respectable or legitimate arts with which they are familiar.

447
achieve, focusing on the traits specific to their medium. Similarly, cinema quickly gave
up on theatrical staging (camera “fifth-row center,” capturing the entire scene at one time
from one side of the action) and moved toward editing and camera-work, characteristics
unavailable to the theater. Even cinema and television, two exceedingly similar media,
learned to distinguish themselves early on, with cinema privileging the vistas and wide-
shots allowed by their enormous screens and television instead privileging the close-up
intimacy offered by small screens much closer to the viewer (although the emergence of
the home-viewing secondary market, as well as wide-screen HDTVs, have lessened
these distinctions). And just as all these choices have allowed for a means of
expression in their respective media unique from any other, so too videogames have
their own unique means of signification.
It’s easy to forget that the videogame medium is still in its adolescent stages,
relatively speaking. If we take the November 1971 arrival of Computer Space—the first
commercially sold, coin-operated videogame—as the dawn of the medium, videogames
are still less than forty years old, a mere child in the grand scheme of things. At similar
ages, cinema was still adjusting to the recent coming of sound, radio was only beginning
to be used for public broadcasting, and the novel was still a disparate and ambiguous
form. As a technology and an industry, videogames have grown at an exponential rate
in recent years, outpacing most computer simulation R&D and rivaling Hollywood
filmmaking as a worldwide money-maker. Yet as an artform, videogames are still quite
green, a fact evinced by the frequency by which they still feel compelled to imitate their
older siblings: cinema, television, and literature; sports, casinos, and board games.
In fact, it seems a large portion of videogames today are still deliberately
imitating some previous manner of narrative or some traditional manner of game. Sports
games, racing games, and fighting games are all imitations of some real-world sport,

448
and videogames have seen a long history of board games, card games, game shows,
and such equally being adapted to videogames. Beyond these, many videogames
which cannot be traced to a specific game source still use a traditional competition
model as their core (such as multiplayer shooters). Meanwhile, movies, television
programs, comic books, and other existing narrative texts are commonly adapted into
videogames. Beyond literal adaptations, a plethora of videogame genres (roleplaying
games, adventure games, survival-horror, and shooters to name a few) stridently imitate
the conventions of cinema, using continuity editing, voiceovers, character development,
linear three-act plots, cinematic genres, and other devices lifted straight from the
movies—not to mention cut-scenes, the most flagrant “borrowing” imaginable, in which
gameplay is intercut with actual passages of digital cinema.
And yet, much as cinema began with imitation but gradually evolved into its own
medium, we’re already seeing videogames’ ability to move beyond mere appropriation
by evolving into medium-specific experiences. Halo and Resident Evil both take their
premise from past films (they are, essentially, reworkings of Aliens and Night of the
Living Dead, respectively) and deploy highly cinematic techniques throughout (cut-
scenes, character development and story arc, lighting and atmosphere, score, etc.),
while the Madden NFL franchise provides a deliberate imitation of a real-world game.
Yet despite their derivative foundation, these games have found innovative ways to
assimilate their borrowed materials into the videogame medium: the fixed angle
perspective of Resident Evil, imitating the composition and editing of cinema, helps
create the game’s terrifying and claustrophobic affect by strictly controlling what the
player can and cannot see, while Madden smartly integrates the roles of player, coach,
owner, and fan into a unique play experience quite unlike the actual playing of football.
Which is to say: the designers didn’t just settle on imitating the way that other texts or

449
other games do things; instead, they adapted to the strengths of their medium and
crafted a distinctly “videogamic” experience.
Equally, instances of “original” videogame-play frequently combine attributes of
both games and traditional narratives to some degree. Katamari Damacy was lauded for
an exceedingly clever new game mechanic—rolling around a ball that picks up any
object small enough to stick to it, creating a snowball of objects that begins at the size of
a thumbtack and grows big enough to collect buildings, islands, continents, clouds, and
anything else on the planet. It is a highly unique experience, one that would be difficult
to imagine being translated to any other type of text or game, and yet, the discursive
elements involved are familiar: Katamari utilizes narrative passages, characters, an
explicable diegetic world, and a linear story (though a rather bizarre one) that explains
why you are performing this activity, and combines these with timed rounds, specific
objectives, and point totals (the size of your katamari), along with opportunities to go
back and beat your high score. In such an instance, we can clearly see the medium’s
roots, yet can hardly reduce the medium to those roots alone.
As such, I think it is an exciting time to be witness to the rise of videogames, both
as a player and a scholar. The medium is just beginning to come into its own, emerging
as a prominent member of the cultural landscape, and the growing field of “serious
games” perhaps best evinces this: from educational and informative videogaming to
abstractly expressive art-house games to politically-invested game design (sometimes
called “persuasive games,” “games for change,” or “social impact games”). For anyone
still skeptical of videogames place in the “legitimate” media environment: just keep yours
eyes on them, because they’re only getting better.
As I’ve (hopefully) argued herein, videogames are a complex and dynamic
medium, one that defies many conventional models of discourse but is equally not

450
entirely unfamiliar. As a thing that is both “read” (a text) and “done” (a game), they
combine a number of disparate impulses. They provide a tangible, signifying object, but
equally defy possession via the “liveness” of their iterative discourse. They offer finite,
rule-bound spaces in which to play, but equally the infinite vistas of a diegetic world.
They position their players inside their worlds, but equally outside as a disembodied
spectator. They provide an avatar that is the embodiment of the player, but equally
autonomous and sentient, and as a result, combine innovative techniques of present-
tense, second-person narration with the conventional tropes of a past-tense, third-
person address. With such a vast medium, no one theoretical approach will dissolve all
others, yet as place to begin thinking of the unique and medium-specific qualities of this
contemporary art, I hope that the conceptual models that I’ve presented might provide a
basis for more fruitful examination of this revolutionary (or at the very least, playfully
rebellious) young medium. Whether as designers, as players, or merely as members of
contemporary society, understanding the nature of videogame discourse is a worthwhile
endeavor, and to appreciate the medium, we need recognize and accept the videogame
situation for what it is: a player, a system, a world, and an avatar.

451

Bibliography

Aarseth, Epsen J. “Aporia and Epiphany in Doom and The Speaking Clock: The
Temporality of Ergodic Art.” In Cyberspace Textuality: Computer Technology and
Literary Theory, edited by Marie-Laure Ryan. Bloomington: Indiana University
Press, 1999.

----. Cybertext: Perspectives on Ergodic Literature. Baltimore: The Johns Hopkins Press,
1997.

Abt, Clark C. Serious Games. New York: Viking Press, 1970.

Altman, Rick. “A Semantic/Syntactic Approach to Genre.” In Film Genre Reader, edited
by Barry Keith Grant, 26-40. Austin, TX: University of Texas Press, 1986.

American Heritage Dictionaries. Dictionary of Computer and Internet Words: An A to Z
Guide to Hardware, Software, and Cyberspace. Boston: Houghton Mifflin Co.,
2001.

The American Heritage Dictionary of the English Language, Fourth Edition. Boston:
Houghton Mifflin Company, 2004

Aristotle. Poetics. Translated by Malcolm Heath. New York: Penguin Classics, 1996.

Avedon, Elliot, and Brian Sutton-Smith. The Study of Games. New York: John Wiley &
Sons, Inc., 1971.

Bal, Mieke. Narratology: Introduction to the Theory of Narrative. Toronto: University of
Toronto Press, 1997.

Barthes, Roland. Camera Lucida. Translated by Richard Howard. New York: Hill and
Wang, 1981.

----. Mythologies. Translated by Annette Lavers. New York: Hill and Wang, 1972.

452
----. S/Z. Translated by Richard Miller. New York: Hill and Wang, 1974.

Bateson, Gregory. “A Theory of Play and Fantasy.” In Steps to an Ecology of Mind:
Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology, 177-
193. Chicago: University of Chicago Press, 2000.

Bolter, Jay David, and Richard Grusin. Remediation: Understanding New Media.
Cambridge: MIT Press, 2001.

Bordwell, David. Narration in the Fiction Film. Madison, WI: The University of Wisconsin
Press, 1985.

Bordwell, David, Janet Staiger, and Kristen Thompson. The Classical Hollywood
Cinema: Film Style & Mode of Production to 1960. New York: Columbia
University Press, 1985.

Branigan, Edward. Narrative Comprehension and Film. London: Routledge, 1992.

Brewster, Ben. “A Scene at the Movies.” Screen, vol. 23, no. 2 (July/August 1982).

Brummett, Barry. Rhetoric in Popular Culture, 2
nd
ed. Thousand Oaks, CA: Sage
Publications, Inc., 2006.

Caillois, Roger. Man, Play, and Games. Translated by Meyer Barash. Champaign:
University of Illinois Press, 2001.

Cardwell, Mike, and Cara Flanigan. Psychology A2: The Complete Companion.
Cheltenham, UK: Nelson Thornes: 2003.

Carlyle, Thomas. Latter-day Pamphlets. n.p.: Chapman and Hall, 1850.

Carroll, Noel. The Philosophy of Horror. New York: Routledge, 1990.

Chatman, Seymour. Coming to Terms: The Rhetoric of Narrative in Fiction and Film.
Ithaca, NY: Cornell University Press, 1990.

----. Story and Discourse: Narrative Structure in Fiction and Film. Ithaca, NY: Cornell
University Press, 1978.

Civilization III packaging, European release. Lyon, France: Infogrames Interactive, Inc.,
2002.

Greg Costikyan. “I Have No Words and I Must Design.” Greg Costikyan’s homepage.
http://www.costik.com/nowords.html (accessed July 13, 2007).

Covino, William A., and David A. Jolliffe, Rhetoric: Concepts, Definitions, Boundaries.
Boston: Allyn and Bacon, 1995.

453
Corning, Peter A. “The Re-Emergence of ‘Emergence’: A Venerable Concept in Search
of a Theory.” Complexity 7:6 (2002): 18-30.
http://www.complexsystems.org/publications/pdf/emergence3.pdf.

Crawford, Chris. The Art of Computer Game Design. Washington State University.
http://www.vancouver.wsu.edu/fac/peabody/game-book/Chapter1.html (accessed
July 13, 2007).

Creative USA Press Release. "Creative Environmental Audio Technology Now De Facto
Industry Standard for Immersive 3D Gaming." Creative USA (August 3, 1999).
http://us.creative.com/corporate/pressroom/releases/welcome.asp?pid=6214.

de Certeau, Michel. The Practice of Everyday Life. Translated by Steven Rendall.
Berkeley, CA: University of California Press, 1984.

Dictionary.com Unabridged (v 1.1), Random House, Inc, http://dictionary.reference.com.

Dowden, Edward. The Life of Percy Bysshe Shelley. n.p.: Lippincott, 1887.

Ebert, Roger. “RogerEbert.com: Answer Man.” Digital Chicago, Inc.
http://rogerebert.suntimes.com/apps/pbcs.dll/section?category=ANSWERMAN&d
ate= 20051127 (accessed 7/26/2007).

Eco, Umberto. A Theory of Semiotics. Bloomington, IN: Indiana University Press, 1976.

Eggins, Suzanne. An Introduction to Systemic Functional Linguistics. New York:
Continuum, 2005.

Fine, Gary Alan. Shared Fantasy: Role-Playing Games as Social Worlds. Chicago: The
University of Chicago Press, 1983.

Frasca, Gonzolo. “Ludologists love stories, too: notes from a debate that never took
place.” Ludology.org. http://www.ludology.org/articles/Frasca_LevelUp2003.pdf
(accessed 7/18/2008).

----. "Ludology Meets Narratology: Similitude and differences between (video)games and
narrative." Ludolgy.org. http://www.ludology.org/articles/ludology.htm (accessed
7/17/2008).

----. “Simulation vs. Narrative: Introduction to Ludology.” In The Video Game Theory
Reader, edited by Mark J.P. Wolf and Bernard Perron, 221-35. New York:
Routledge, 2003.

Frith, Christopher D. Making Up the Mind: How the Brain Creates Our Mental World.
Boston: Blackwell Publishing: 2007.

Fuller, Mary, and Henry Jenkins. “Nintendo and New World Travel Writing: A Dialogue.”
In Cybersociety: Computer-Mediated Communication and Community, edited by
Steven G. Jones, 57-72. Thousand Oaks: Sage Publications, 1995.

454

Gee, James Paul. What Video Games Have to Teach Us About Learning and Literacy.
New York: Palgrave/Macmillan, 2003.

----. Why Video Games are Good for Your Soul: Pleasure and Learning. Urbana, IL:
Common Ground: 2005.

Genette, Gerard. Narrative Discourse: An Essay in Method. Translated by Jane E.
Lewin. Ithaca: Cornell University Press, 1980.

Gerrig, Richard. Experiencing Narrative Worlds: On the Psychological Activities of
Reading. New Haven, CT: Yale University Press, 1993.

Gibson, William. Neuromancer. New York: Ace Books, 1984.

Graddol, David. “What is a Text?” In Media Texts: Authors and Readers, edited by David
Graddol and Oliver Boyd-Barrett, 40-50. Clevedon, UK: Multilingual Matters, Ltd.,
1994.

Guérard, Albert Léon. French Prophets of Yesterday. n.p.: T.F. Unwin, 1913.

Half-Life Platinum Collection Instruction Manual. Los Angeles: Sierra Entertainment, Inc.
and Valve, L.L.C., 2002.

Halliday, Michael. “Text as semantic choice in social contexts.” In Linguistic Studies of
Text and Discourse (Collected Works of M.A.K. Halliday), edited by Jonathan J.
Webster, 23-84. New York: Continuum, 2002

Hjelmslev, Louis. Prolegomena to a Theory of Language. Translated by Francis J
Whitfield. Madison: University of Wisconsin Press, 1961.

Herz, J.C. Joystick Nation. Boston: Little, Brown and Company: 1997.

Holland, John. Emergence: From Chaos To Order. New York: Basic Books, 1998.

Huizinga, Johan. Homo Ludens: A Study of the Play Element in Culture. Boston: Beacon
Press, 1955.

Insurgency Forums, http://forums.insmod.net/index.php

Jansen, Erin. Netlingo: The Internet Dictionary. Ojai, CA: NetLingo Inc., 2002.

Jenkins, Henry. “’Complete Freedom of Movement’: Video Games as Gendered Play
Spaces.” In From Barbie to Mortal Kombat: Gender and Computer Games,
edited by Justine Cassell and Henry Jenkins, 262-297. Cambridge: The MIT
Press, 1998.

455
----. Foreword to a reprinting of “Game Design as Narrative Architecture.” In The Game
Design Reader, edited by Katie Salen and Eric Zimmerman, 670. Cambridge:
The MIT Press, 2006.

----. “Game Design as Narrative Architecture.” In First-Person: New Media as Story,
Performance, and Game, edited by Noah Wardrip-Fruin and Pat Harrigan, 118-
130. Cambridge: The MIT Press, 2004.

Johnson, Steven. Emergence: The Connected Lives of Ants, Brains, Cities, and
Software. New York: Scribner, 2001.

Juul, Jesper. “Games Telling Stories?: A brief note on games and narratives.” Game
Studies, vol. 1, issue 1 (July 2001), http://www.gamestudies.org/0101/juul-gts/.

----. Half-Real: Video Games Between Real Rules and Fictional Worlds. Cambridge: The
MIT Press, 2005.

Kinder, Marsha. “Narrative Equivocations Between Movies and Games.” In The New
Media Book, edited by Dan Harries, 119-132. London: The British Film Institute,
2002.

----. Playing with Power. Berkeley, CA: University of California Press, 1991.

King, Brad, and John Borland. Dungeons and Dreamers: The Rise of Computer Game
Culture. New York: McGraw-Hill/Osborne, 2003.

Kinneavy, James. A Theory of Discourse. Englewood Cliffs, NJ: Prentice-Hall, 1971.

Kushner, David. Masters of Doom: How Two Guys Created an Empire and Transformed
Pop Culture. New York: Random House, Inc., 2004.

Lanser, Susan Sniader. The Narrative Act: Point of View in Prose Fiction. Princeton, NJ:
Princeton University Press, 1981.

Laurel, Brenda. Computers as Theater. Reading, MA: Addison-Wesley Publishing
Company, 1993.

MacCabe, Colin. Tracking the Signifier: Theoretical Essays: Film, Linguistics, Literature.
Manchester: Manchester University Press, 1985.

Macdonell, Arthur Anthony. “Ramayana.” In Encyclopedia of Religion and Ethics Part 20,
edited by James Hastings, 574-8. Whitefish, MT: Kessinger Publishing, 2003.

MacGraw Hill Book Company. Dictionary of Computing & Communications. New York:
McGraw-Hill Publishing, 2003.

MajorMike. Review of Metal Gear Solid. GamePro.com,
http://www.gamepro.com/sony/psx/games/reviews/236.shtml (posting date
undetermined, accessed June 29, 2006).

456

Maltby, Richard. Hollywood Cinema, 2nd ed. Cambridge: Blackwell Publishing: 2003.

Manovich, Lev. The Language of New Media, 2
nd
edition. Cambridge: The MIT Press,
2001.

Mayne, Judith. Cinema and Spectatorship. London: Routledge, 1993.

McCloud, Scott. Understanding Comics. New York: HarperCollins Publishers, 1993.

McKechnie, Jean L. ed. Webster’s New Universal Unabridged Dictionary, Deluxe
Second Edition. New York: Simon & Schuster, 1983.

McLuhan, Marshall. Understanding Media: The Extensions of Man. New York:
Routledge, 2001.

Meadows, Mark S. Pause and Effect: The Art of Interactive Narrative. Berkeley: New
Riders Press, 2002.

Metal Gear Solid Instruction Manual. Redwood City, CA: Konami of America, Inc., 1998.

Metz, Christian. Film Language: A Semiotics of the Cinema. Translated by Michael
Taylor. New York: Oxford University Press, 1974.

----. The Imaginary Signifier: Psychoanalysis and the Cinema. Translated by Celia
Britton, Annwyl Williams, Ben Brewster, and Alfred Guzzetti. Bloomington:
Indiana University Press, 1982.

Morabito, Margaret. “Enter the On-Line World of Lucasfilm.” RUN, August 1986.

Mori, Masahiro. “The Uncanny Valley.” Translated by K. F. MacDorman & T. Minato.
Energy, 7:4 (1970).

Morningstar, Chip and F. Randy Farmer. “The Lessons of LucasFilm’s Habitat.” In
Cyberspace: First Steps, edited by Michael Benedikt. Cambridge: The MIT
Press, 1991.

Murray, Janet. Hamlet on the Holodeck. New York: The Free Press, 1997.

Myst User’s Manual. Novato, Calif.: Broderbund Software Inc. and Cyan Inc., 1993.

Nash, Mark. "Vampyr and the Fantastic." Screen 17 (1976).

National Basketball Association Basketball Operations Dept. Official Rules of the
National Basketball Association 2007-2008. National Basketball Association.
http://www.nba.com/media/rule_book_2007-08.pdf (accessed December 12,
2007).

457
Nelson, Randy. Review of Metal Gear Solid. IGN, posted on October 21, 1998,
http://psx.ign.com/articles/150/150569p1.html.

Parlett, David. The Oxford History of Board Games. New York: Oxford University Press,
1999.

Park, Stephen. “Abstraction.” The University of Chicago Theories of Media Keywords
Glossary. http://csmt.uchicago.edu/glossary2004/abstraction.htm (accessed
November 2007).

Peirce, Charles Sanders. Collected Papers of Charles Sanders Peirce, vol. 1-6. Edited
by Charles Hartshorne and Paul Weiss. Cambridge: Harvard University Press,
1935.

Perry, Douglass C. Review of Jak II. IGN, Posted on October 9, 2003,
http://ps2.ign.com/articles/454/454132p1.html.

----. Review of Metal Gear Solid. IGN, Posted on October 21, 1998,
http://psx.ign.com/articles/150/150569p1.html.

Petersen, Julie K. The Telecommunications Illustrated Dictionary, 2
nd
Edition. Boca
Raton, FL: CRC Press, Inc., 2002.

Product description for Madden NFL 09. IGN.com.
http://xbox360.ign.com/objects/142/14229536.html (accessed July 12, 2008).

Ricoeur, Paul. Hermeneutics and the Human Sciences: Essays on Language, Action,
and Interpretation. Translated by John B. Thompson. Cambridge University
Press, 1981.

Ryan, Marie-Laure. “Beyond Myth and Metaphor: The Case of Narrative in Digital
Media.” Game Studies vol. 1, issue 1 (July 2001),
http://gamestudies.org/0101/ryan/ (accessed April, 2003).

----. Narrative as Virtual Reality: Immersion and Interactivity in Literature and Electronic
Media. Baltimore: The Johns Hopkins University Press, 2001.

Salen, Katie, and Eric Zimmerman. The Game Design Reader. Cambridge: The MIT
Press, 2006.

----. Rules of Play: Game Design Fundamentals. Cambridge: MIT Press, 2004.

Saussure, Ferdinand de. Course in General Linguistics. Edited by Charles Bally and
Albert Sechehaye with Albert Riedlinger. Translated by Roy Harris. Chicago:
Open Court, 1986.

Scrabble Crossword Game Game Rules. Pawtucket, RI: Hasbro, 2003.

458
Searle, John. Expression and Meaning: Studies in the Theory of Speech Acts.
Cambridge: Cambridge University Press, 1979.

Singer, Dorothy G., Roberta Michnic Golinkoff, and Kathy Hirsh-Pasek. Play = Learning:
How Play Motivates and Enhances Children's Cognitive and Social-Emotional
Growth. Oxford University Press: 2006.

Stam, Robert. Reflexivity in Film and Literature: From Don Quixote to Jean-Luc Godard,
2
nd
ed. New York: Columbia University Press, 1992.

Stam, Robert, Robert Burgoyne, and Sandy Flitterman-Lewis. New Vocabularies in Film
Semiotics. London: Routledge, 1992

Steele, Guy L. The New Hacker’s Dictionary. Cambridge: The MIT Press, 1996.

Stephenson, Neal. Snow Crash. New York: Bantam Books, 1992.

Sudnow, David. Pilgrim in the Microworld. New York: Warner Books, 1983.

Suits, Bernard. Grasshopper: Games, Life, and Utopia. Boston: David R. Godine, 1990.

Todorov, Tzvetan. Genres in Discourse, trans. Catherine Porter. Cambridge: Cambridge
University Press, 1990.

“The Top Ten Scariest Moments in Gaming.” Game Informer 174, October 2007.

Tzvetan Todorov. The Poetics of Prose. Translated by Richard Howard. Ithaca, NY:
Cornell University Press, 1977.

Truffaut, François. Hitchcock. London: Paladin-Granada Publishing, 1978.

Valdes, Robert. “The Artificial Intelligence of Halo 2.” How Stuff Works.
http://electronics.howstuffworks.com/halo2-ai.htm (accessed February 12, 2008).

Venkatesananda, Swami. The Concise Ramayana of Valmiki. Albany: State University of
New York Press, 1988.

Victor (last name unstated). Review of Metal Gear Solid. The Electric Playground,
http://www.elecplay.com/reviews_article.php?article=485 (posting date
undetermined, accessed June 29, 2006).

Walton, Kendall. “Fearing Fictions.” Journal of Philosophy, vol. 75, no. 1 (January 1978).

Weiss, D.B. Lucky Wander Boy. New York: Plume, 2003.

Winn, Philip. Dictionary of Biological Psychology. London: Taylor & Francis: 2001.

459
Wolf, Mark J.P. “Abstraction in the Videogame.” In The Video Game Theory Reader,
edited by Mark J.P. Wolf and Bernard Perron, 47-65. New York: Routledge,
2003.

----. The Medium of the Video Game. Austin, TX: University of Texas Press, 2001.

460

Appendix A: Videogames Referenced

007: Nightfire (PlayStation 2). Eurocom Entertainment Software and Savage
Entertainment, LLC. Electronic Arts, Inc., 2002.

Adventure (Apple II). Softwin Associates. Microsoft, 1979.

Adventure (Atari 2600). Atari, Inc. Sears, Roebuck and Co., 1980

America’s Army (Windows). U.S. Army. U.S. Army, 2002.

American McGee presents Bad Day L.A. (Windows). Enlight Software Ltd. and The
Mauretania Import Export Company, Inc. Aspyr Media, Inc., 2006.

American McGee’s Alice (Windows). Rogue Entertainment. Electronic Arts, Inc., 2000.

Amplitude (PlayStation 2). Harmonix Music Systems, Inc. Sony Computer
Entertainment America, Inc., 2003.

Asteroids (arcade). Atari, Inc. Atari, Inc., 1979.

Baldur’s Gate (Windows). Bioware Corp. Interplay Productions, Inc., 1998.

Battlefield 1942 (Windows). Digital Illusions CE AB. Electronic Arts, Inc., 2002.

Bejeweled Deluxe (Windows). PopCap Games, Inc. PopCap Games, Inc., 2000.

Bioshock (Xbox 360). 2K Australia and 2K Boston. 2K Games, 2007.

Black & White (Windows). Lionhead Studios Ltd. Electronic Arts, Inc., 2001.

Blue Dragon (Xbox 360). Artoon Co., Ltd. and Mistwalker. Microsoft Game Studios,
2007.

Breakout (arcade). Atari, Inc. Atari, Inc., 1976.

461
Brimstone (Commodore 64). Synapse Software Corporation. Brøderbund Software,
Inc., 1985.

Burnout Paradise (Xbox 360). Criterion Studios. Electronic Arts, Inc., 2008.

Call of Duty (Windows). Infinity Ward, Inc. Activision Publishing, Inc., 2003.

Castlevania (Nintendi Entertainment System). Konami Industry Co. Ltd. Konami, Inc.,
1987.

Colossal Cave Adventure (PDP-10). Crowther, Will and Don Woods. CRL, 1977.

Contra (Nintendo Entertainment System). Konami Industry Co. Ltd. Konami, Inc., 1988.

Dance Dance Revolution (PlayStation). KCE Tokyo. Konami Corporation, 2001.

Dead Rising (Xbox 360). Capcom Co., Ltd. Capcom Entertainment, Inc., 2006.

Defender (arcade). Williams Electronics. Williams Electronics, 1980.

Deus Ex (Windows). Ion Storm Inc. Eidos Interactive, Inc., 2000.

Doom (DOS). id Software, Inc. id Software, Inc., 1993.

Doom
3
(Windows). id Software, Inc. Activision Publishing, Inc., 2004.

EarthBound (Super Nintendo). Ape and HAL Laboratory, Inc. Nintendo of America Inc.,
1995.

The Elder Scrolls IV: Oblivion (Xbox 360). Bethesda Game Studios. 2K Games and
Bethesda Softworks LLC, 2006.

Eternal Darkness: Sanity’s Requiem (GameCube). Silicon Knights, Inc. Nintendo of
America Inc., 2002.

Fallout (Windows). Interplay Productions, Inc. Interplay Production, Inc., 1997.

F.E.A.R.: First Encounter Assault Recon (Windows). Monolith Productions, Inc. Vivendi
Universal Games, Inc., 2005.

Final Fantasy (Nintendo Entertainment System). Square Co., Ltd. Nintendo of America
Inc., 1990.

Final Fantasy VII (PlayStation). Square Co., Ltd. SCEA, 1997.

Final Fantasy VIII (PlayStation). Square Co., Ltd. Square Electronic Arts L.L.C., 1999.

flOw (PlayStation 3). thatgamecompany. Sony Computer Entertainment America, Inc.,
2007.

462

Football (Atari 2600). Atari, Inc. Atari, Inc., 1978.

Freedom Force (Windows). Irrational Games. Crave Entertainment, Inc. and Electronic
Arts, Inc., 2002.

Frogger (Atari 8-Bit). Konami Industry Co. Ltd. Sierra On-Line, Inc., 1982.

Gabriel Knights: Sins of the Fathers (Windows). Sierra On-Line, Inc. Sierra On-Line,
Inc., 1993.

Gears of War (Xbox 360). Epic Games, Inc. Microsoft Game Studios, 2006.

Ghosts ‘N Golbins (Commodore 64). Capcom Co., Ltd. Elite Systems Ltd., 1986.

God of War (PlayStation 2). SCE Studio Santa Monica. Sony Computer Entertainment
America, Inc., 2005.

GoldenEye 007 (Nintendo 64). Rare, Ltd. Nintendo of America Inc., 1997.

Gradius (Nintendo Entertainment System). Konami Industry Co. Ltd. Konami, Inc.,
1986.

Gran Turismo (PlayStation). Polyphony Digital Inc. SCEA, 1998.

Grand Theft Auto III (PlayStation 2). DMA Design Limited. Rockstar Games, Inc., 2001.

Grand Theft Auto: San Andreas (PlayStation 2). Rockstar North Ltd. Rockstar Games,
Inc., 2004.

Grand Theft Auto: Vice City (PlayStation 2). Rockstar North Ltd. Rockstar Games, Inc.,
2002.

Guitar Hero (PlayStation 2). Harmonix Music Systems, Inc. RedOctane, Inc., 2005

Habitat (Commodore 64). Lucasfilm Games. Lucasfilm Games, 1987.

Half-Life (Windows). Valve L.L.C. Sierra On-Line, Inc., 1998.

Half-Life: Counter-Strike (Windows). Valve Corporation. Sierra On-Line, Inc., 2000.

Half-Life 2 (Windows). Valve Corporation. Sierra Entertainment, Inc., 2004.

Halo: Combat Evolved (Xbox). Bungie Studios. Microsoft Game Studios, 2001.

Halo 2 (Xbox). Bungie Studios. Microsoft Game Studios, 2004.

Halo 3 (Xbox 360). Bungie Studios. Microsoft Game Studios, 2007.

463
Hexic (Xbox 360). Carbonated Games. Microsoft Game Studios, 2005.

The Hitchhiker’s Guide to the Galaxy (Apple II). Infocom, Inc. Infocom, Inc., 1984.

Hitman: Codename 47 (Windows). IO Interactive A/S. Eidos Interactive, Inc., 2000.

Indiana Jones and the Emperor’s Tomb (Xbox). The Collective. LucasArts, 2003.

Jak and Daxter: The Precursor Legacy (PlayStation 2). Naughty Dog, Inc. SCEA, 2001.

Jak II (PlayStation 2). Naughty Dog, Inc. SCEA, 2003.

John Madden Football (Apple II). Electronic Arts, Inc. Electronic Arts, Inc., 1988.

Kameo: Elements of Power (Xbox 360). Rare, Ltd. Microsoft Game Studios, 2005.

Katamari Damacy (PlayStation 2). Namco Limited. Namco Hometek Inc., 2004.

King’s Quest (DOS). Sierra On-Line, Inc. Sierra On-Line, Inc., 1984.

The Legend of Zelda (Nintendo Entertainment System). Nintendo Co., Ltd. Nintendo of
America Inc., 1987.

The Legend of Zelda: Link’s Awakening DX (Game Boy Color). Nintendo EAD.
Nintendo of America Inc., 1998.

The Legend of Zelda: Ocarina of Time (Nintendo 64). Nintendo EAD. Nintendo Co.,
Ltd., 1998.

The Legend of Zelda: The Wind Waker (GameCube). Nintendo EAD. Nintendo of
America Inc., 2003.

Lego Star Wars: The Video Game (PlayStation 2). Traveler’s Tales Ltd. Eidos
Interactive, Inc. and Giant Interactive Entertainment, Ltd., 2005.

Lemmings (Amiga). DMA Design Limited. Psygnosis Limited, 1991.

The Longest Journey (Windows). Funcom Oslo A/S. Funcom Oslo A/S, 2000.

Lumines Live! (Xbox 360). Q ENTERTAINMENT, Inc. Q ENTERTAINMENT, Inc., 2006.

Madden NFL 09 (Xbox 360). EA Tiburon. Electronic Arts, Inc., 2008.

Maniac Mansion (DOS). Lucasfilm Games. Lucasfilm Games, 1987.

Marathon (Macintosh). Bungie Software Products Corporation. Bungie Software
Products Corporation, 1994.

Mario Kart 64 (Nintendo 64). Nintendo EAD. Nintendo Co., Ltd., 1996.

464

Mass Effect (Xbox 360). Bioware Corp. Microsoft Game Studios, 2007.

Max Payne (Windows). Remedy Entertainment Ltd. Gathering, 2001.

Medal of Honor (Play Station). Dreamworks Interactive L.L.C. Electronic Arts, Inc.,
1999.

Metal Gear Solid (PlayStation). Konami Computer Entertainment Japan, Inc. Konami of
America, Inc., 1998.

Metal Gear Solid 2: Sons of Liberty (PlayStation 2). Konami Computer Entertainment
Japan, Inc. Konami of America, Inc., 2001.

Metroid (Nintendo Entertainment System). Nintendo R&D1. Nintendo of America Inc.,
1986.

Missile Command (arcade). Atari, Inc. Atari, Inc., 1980.

Mortal Kombat II (Sega Genesis). Midway Games, Inc. Acclaim Entertainment, Inc.,
1994.

Mortal Kombat 3 (Sega Genesis). Midway and General Computer Corporation.
Williams Entertainment Inc., 1995.

Ms. Pac-Man (arcade). Midway and General Computer Corporation. Midway and
Namco, 1981.

Myst (Windows). Cyan Worlds, Inc. Brøderbund Software, Inc., 1993.

NBA Live 95 (Super Nintendo). Hitmen Productions. Electronic Arts, Inc., 1994.

NBA Street (PlayStation 2). Electronic Arts Canada and NuFX, Inc. Electronic Arts,
Inc., 2001.

NFL Street (PlayStation 2). EA Tiburon. Electronic Arts, Inc., 2004.

Nintendogs (Nintendo DS). Nintendo EAD. Nintendo of America Inc., 2005.

Ōkami (PlayStation 2). Clover Studio Co., Ltd. Capcom Entertainment, Inc., 2006.

The Operative: No One Lives Forever (Windows). Monolith Productions, Inc. Fox
Interactive, Inc., 2000.

Pac-Man (arcade). Namco and Midway. Namco, 1980.

Paper Mario (Nintendo 64). Intelligent Systems Co., Ltd. Nintendo of America Inc.,
2002.

465
Parasite Eve (PlayStation). Square Soft, Inc. Square Electronic Arts L.L.C., 1998.

Planetfall (DOS). Infocom, Inc. Infocom, Inc., 1983.

Pong (arcade). Atari, Inc. Atari, Inc., 1972.

Prince of Persia: The Sands of Time (Xbox). Ubisoft Divertissements Inc. Ubisoft
Entertainment SA, 2003.

Puzzle Quest: Challenge of the Warlords (Nintendo DS). 1
st
Playable Production, LLC,
Engine Software B.V., and Infinite Interactive Pty. Ltd. D3Publisher of America,
Inc., 2007.

Quake (Windows). id Software, Inc. id Software, Inc., 1996.

Qix (Apple II). Taito America Corporation. Taito Corporation, 1989.

Ratchet & Clank (PlayStation). Insomniac Games, Inc. SCEA, 2002.

Resident Evil (PlayStation). Capcom Co., Ltd. Capcom Co., Ltd., 1996.

Resident Evil 2 (PlayStation). Capcom Co., Ltd. Capcom Entertainment, Inc., 1998.

Resident Evil 4 (GameCube). Capcom Production Studio 4. Capcom Entertainment,
Inc., 2005.

Rogue: The Adventure Game (DOS). Artificial Intelligence Design. Epyx, Inc., 1983.

The Secret of Monkey Island (DOS). Lucasfilm Games. Lucasfilm Games, 1990.

Shadow of the Colossus (PlayStation 2). Team Ico. SCEA, 2005.

Sid Meier’s Civilization (Windows). MPS Labs. MicroProse Software, Inc., 1991.

Sid Meier’s Civilization II (Windows). Microprose Software, Inc. Microprose Software,
Inc., 1996.

Sid Meier’s Civilization III (Windows). Firaxis Games. Infogrames, Inc., 2001.

Sid Meier’s Civilization III (Macintosh). Firaxis Games. MacSoft, 2002.

Sid Meier’s Civilization III: Play the World (Windows). Firaxis Games. Infogrames, Inc.,
2002.

Sid Meier’s Civilization IV (Windows). Firaxis Games. 2K Games, 2005.

SimCity (DOS). Maxis Software Inc. Brøderbund Software, Inc., 1989.

SimCity 4 (Windows). Maxis Software Inc. Electronic Arts, Inc., 2003.

466

The Sims (Windows). Maxis Software Inc. Electronic Arts, Inc., 2000.

The Sims 2 (Windows). Maxis Software Inc. Electronic Arts, Inc., 2004.

Skate (Xbox 360). Electronic Arts Black Box. Electronic Arts, Inc., 2007.

Sly Cooper and the Thievius Raccoonus (PlayStation 2). Sucker Punch Productions
LLC. SCEA, 2002.

Space Invaders (arcade). Taito Corporation. Midway, 1978.

Space Quest (DOS). Sierra On-Line, Inc. Sierra On-Line, Inc., 1986.

Spore (Windows). Maxis Software Inc. Electronic Arts, Inc., 2008.

Star Wars (arcade). Atari, Inc. Atari, Inc., 1983.

Super Breakout (Atari 2600). Atari, Inc. Atari, Inc., 1981.

Super Mario Bros (Nintendo Entertainment System). Nintendo Co., Ltd. Nintendo of
America Inc., 1985.

Super Mario Bros. 2 (Nintendo Entertainment System). Nintendo Co., Ltd. Nintendo of
America Inc., 1988.

Super Mario World (Super Nintendo). Nintendo Co., Ltd. Nintendo of America Inc.,
1991.

Super Metroid (Super Nintendo). Intelligent Systems Co., Ltd. and Nintendo R&D1.
Nintendo of America Inc., 1994.

Star Wars: Battlefront (Xbox). Pandemic Studios, LLC. LucasArts, 2004.

Star Wars Galaxies: An Empire Divided (Windows). Sony Online Entertainment.
Lucasarts, 2003.

Star Wars: The Empire Strikes Back (Atari 2600). Parker Brothers. Parker Brothers,
1982.

System Shock 2 (Windows). Irrational Games and Looking Glass Studios, Inc.
Electronic Arts, Inc., 1999.

Team Fortress Classic (Windows). Valve L.L.C. Sierra On-Line, Inc., 1999.

Tennis (Atari 2600). Activision, Inc. Activision, Inc., 1981.

Tetris (Amiga). Elorg. Mirrorsoft Ltd., 1987.

467
Texas Hold ‘em (Xbox 360). TikGames, LLC. TikGames, LLC, 2006.

Tom Clancy’s Rainbow Six (Windows). Red Storm Entertainment, Inc. Red Storm
Entertainment, Inc., 1998.

Tom Clancy’s Splinter Cell (Xbox). Ubisoft Divertissements Inc. Ubisoft Entertainment
Software, 2002.

Tomb Raider (PlayStation). Core Design Ltd. Eidos, Inc., 1996.

Tomb Raider: Legend (PlayStation 2). Crystal Dynamics, Inc. Eidos, Inc., 2006.

True Crime: Streets of LA (Xbox). Luxoflux Corp. Activision Publishing, Inc., 2003.

Ultima (Apple II). Garriott, Richard. California Pacific Computer Co., 1980.

Ultima II: Revenge of the Enchantress (Apple II). Garriott, Richard. Sierra On-Line, Inc.,
1982.

Ultima III: Exodus (Apple II). ORIGIN Systems, Inc. ORIGIN Systems, Inc., 1983.

Ultima IV: Quest of the Avatar (Apple II). ORIGIN Systems, Inc. ORIGIN Systems, Inc.,
1985.

Unreal (Windows). Digital Extremes and Epic MegaGames, Inc. GT Interactive
Software Corp., 1998.

Wasteland (DOS). Interplay Productions, Inc. Electronic Arts, Inc., 1988.

Wii Sports (Wii). Nintendo Co., Ltd. Nintendo of America Inc., 2006.

Wolfenstein 3D (DOS). id Software, Inc. Apogee Software, Ltd., 1992.

Zork (Apple II). Infocom, Inc. Infocom, Inc., 1980.

Abstract (if available)

Abstract A study of the medium-specific properties of videogame discourse, this dissertation considers the videogame situation as the interaction of four component parts: a PLAYER, participating in a formal SYSTEM that projects a fictional WORLD, via the mediation of an AVATAR, and these four components form the basis for the four chapters of this project.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The moonlighters: a narrative listening approach to videogame storytelling

PDF

Juliette: revealing character from choice to world

PDF

Spectre: exploring the relationship between players and narratives in digital games

PDF

When a story calls: the narrative potential of mobile media

PDF

Lesbianscape of Taiwan: media history of Taiwan's lesbians

PDF

The witch: Identity construction and the fairy tale in interactive narrative

PDF

Exporting tears and fantasies of (under) development: popular television genres, globalization and nationalism in Mexico after World War II

PDF

Runesinger: a demonstration of coupling Korean phonics and spelling to a videogame

PDF

Rethinking feminist cinema: Agnès Varda and filmmaking in the feminine

PDF

Riddles of representation in fantastic media

PDF

"Shaken out of the ruts of ordinary perception": vision, culture and technology in the psychedelic sixties

PDF

Eyez: Spatial perception in videogames

PDF

Leafcutters: life simulation gameplay designed to evoke engagement with real-world subject matter

PDF

Dead zones: human mobility and the making of media nationalism

PDF

Sacred vocabularies for world cinema: transfiguring ancient aural and visual modalities to express sacredness for the contemporary age

PDF

The Hollywood left: cinematic art and activism in the 1930s

PDF

The exquisite corpse: disarticulations of the artificial female

PDF

Emetic Labs: simulator sickness in videogame players

PDF

Caution children crossing: home, integration narratives, and the gentle warrior, 1950-1965

PDF

Embodied memory: the formation of archived audiovisual holocaust testimony in the United States

Asset Metadata

Creator Buerkle, Robert (author)

Core Title Of worlds and avatars: a playercentric approach to videogames

School School of Cinematic Arts

Degree Doctor of Philosophy

Degree Program Cinema-Television (Critical Studies)

Publication Date 10/10/2008

Defense Date 08/20/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag avatars,discourse,Fiction,Games,narrative,new media,OAI-PMH Harvest,Play,Players,rules,videogames

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Friedberg, Anne (committee chair), Fullerton, Tracy (committee chair), Thomas, Douglas (committee member)

Creator Email robertbuerkle@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1653

Unique identifier UC1137210

Identifier etd-Buerkle-2352 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-125995 (legacy record id),usctheses-m1653 (legacy record id)

Legacy Identifier etd-Buerkle-2352.pdf

Dmrecord 125995

Document Type Dissertation

Rights Buerkle, Robert

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu