Content uploaded by Gareth Schott
Author content
All content in this area was uploaded by Gareth Schott on Apr 03, 2017
Content may be subject to copyright.
Heavy Hero or Digital Dummy:
multimodal player-avatar relations in
FINAL FANTASY 7
Andrew Burn
Gareth Schott
University of London Institute of Education
Abstract
This article analyses the player-avatar relation in Final Fantasy 7, drawing on
multimodality theory to analyse textual structures both in the game and in the
discourse of player-interviews and fan writing. It argues that the avatar is a two-part
structure, partly designed in conventional narrative terms as a protagonist of popular
narrative, and partly as a vehicle for interactive game-play. The former structure is
replete with the traditions and designs of Japanese popular narrative, oral formulaic
narrative and contemporary superhero narratives; and is presented to the player as an
offer act – a declarative narrative statement. The latter is a construct of evolving
attributes and economies characteristic of roleplaying games; and is presented to the
player as a demand act – a rule-based command. Though these two functions separate
out in the grammar of player and fan discourse, it is their integration which provides
the pleasure of gameplay and narrative engagement.
Squaresoft’s Final Fantasy 7 is a hugely-successful Japanese role-playing game,
which sold to virtually all Japan’s Playstation owners within the first 48 hours of its
release, and was no less popular in the US on its release there later in 1997 (UK
Playstation Magazine, November 1997). Cloud Strife is the protagonist-avatar - a
mysterious mercenary, in leather and big boots, wielding a sword as big as himself;
but an oddly childish face, whimsically delineated in the ‘deformed aesthetic’ of
manga, with enormous, glowing blue eyes, framed in cyberpunk blond spikes (Fig. 1).
We will explore the player’s engagement with the avatar through a social semiotic
analysis of the design of the character, of two interviews with players, and of fan
writings.
[INSERT FIGURE 1 ABOUT HERE]
This is how Rachel, a 17-year-old English player of Final Fantasy 7, describes Cloud
in one of two research interviews we will refer to:
It’s just basically you play this character who’s in this like really cool like
cityscape and you have to, er, and he finds out … and, er, he escapes because
he finds out that, um, he’s, because he starts having these flashbacks, and he
escapes from this city because he’s being pursued I think, and, um, he has to
defeat this big corporation and try and – oh yeah, Sephiroth, he’s this big
military commander, and you have to go and try to stop him, cos he’s trying to
raise up all the beasts, and you do this by collecting materia, which you can
use for magic and stuff, and you use your own weapons, and –
The apparently innocent clause ‘You play this character’ in Rachel’s account conceals
the central, powerful structures of RPG play. ‘This character’ evokes the conventional
fictional character operating as protagonist in a narrative. However, as well as
protagonist in the conventional sense, Cloud is the player’s embodiment in the game,
the avatar (the word, from the Sanskrit for “descent”, refers to the embodiment of a
god on earth). This article will explore how this dual function is constructed, how it is
experienced by the player, and why the words “You play” indicate very precisely the
grammatical relation of player and avatar. At the same time, it will explore how this
dual function relates to the two fundamental elements of the game: what Linderoth
(2002) refers to as system and guise, the former being the rule-based system of the
game, in computer-games produced by the procedural work of the game engine; the
latter being the visible gameworld, narrative and characters overlaid on the system.
The interaction of player and avatar is played out in two fundamental functions of the
text: how it represents aspects of the world (in this case fantasy narratives); and what
it offers to do to, for or with its audience. The multimodality theory on which we draw
in this article (Kress & van Leeuwen, 2002; Lemke, 2002) would call the first
function ideational; and the second interpersonal , adapting these terms from
functional linguistics (Halliday 1985). These are very large categories; and have been
re-thought and re-named by successive theorists in linguistics and semiotics. Kress
and van Leeuwen (1996), for instance, call the first function representational and the
second interactive; while Lemke (2002) calls them presentational and orientational.
Central questions that multimodal semiotics can help to address, in the context of a
game, are: how do such functions interrelate (how is the game-narrative entangled
with the interactive experience it provides?); and how are they realised by
combinations of communicative modes: animation, visual design, music, text, sound?
Another way to think of the ideational, or representational function, at least in
narrative, is to think of what in language is the transitivity system: how Actors
perform Actions upon Goals, or, simply put, who does what to whom. This is the
basic idea on which the French narratologist Gerard Genette builds his theory,
proposing that narrative is an expansion of the grammatical category of verb: it is
about action (1980). The grammatical structure of Rachel’s account suggests that the
element of Actor in the transitivity system of the game is divided. In parts of her
account the Actor is, conventionally, Cloud, rendered in the third person (“he
escapes”). Elsewhere, the pronoun representing the Actor changes to indicate the
player (“you have to go …”).
As for the interpersonal or interactive function, the question here is how a game
establishes a relation between itself and the player. Genette suggests that, in relation
to the mood of the verb, narratives are by definition indicative – they make
statements. Rachel’s account of Cloud, in its grammatical rendering of the avatar,
makes it clear that, while in some parts of the narrative the hero is going about his
business wrapped in the familiar indicative mood (“he escapes from this city because
he’s being pursued”), in other parts, the player has become the protagonist, and the
game is clearly in the imperative (“you have to go and try to stop him, cos he’s trying
to raise up all the beasts”). In terms of its interactive function, then, the game is not
only offering a narrative statement but telling the player to do something – in effect,
telling the player to insert herself into the transitivity system of the game.
Roleplaying games do offer narrative statements, as in Genette’s classic model. In this
respect, we can and should ask the usual questions about how the protagonist is
constructed: what is his cultural provenance, what kinds of narrative function will he
perform, how might audiences engage with these? On the other hand, RPGs, like all
games, ask you questions and tell you to do things. If narrative requires a willing
suspension of disbelief, games require a willing submission to rule-based systems. In
this latter respect, we can and should go back to the question of transitivity, to ask
how the player is involved in the actions the character performs; and back to the
question of mood, to ask how text-player relations are invited and constructed.
Cloud – Heavy Hero?
Rachel’s word “character” suggests Cloud’s function as part of the guise of the game.
In the ideational system of the game, if the narrative is Genette’s verb-writ-large,
Cloud is the Actor who performs that verb. His narrative function, as a hero-
mercenary who defies the ruthless Shinra corporation and his former hero, now his
nemesis and enemy, Sephiroth, is typical of hero-roles in popular narrative, and, in
many respects, of the formulaic character-types of Propp’s well-known analysis of
folktale narratives (Propp, 1970).
As the folktale analogy suggests, a character like Cloud does not spring out of
nowhere to fulfil the needs of commodified mass entertainment in the 20th century,
but draws deeply on popular forms in folk culture, oral narrative, and the fantasies of
popular romance which have offered consolation, polemic, and psychic testing-
grounds, through elaborate allegories, for the rites of passage and tribulations of
everyday life since the mediaeval period and earlier. Eri Izawa (2000) describes how
the characters of manga and anime, and of Final Fantasy 3, draw on epic themes in
Japanese folklore, on hero-legends based on historic warlords, and on supernatural
narratives informed by Shinto and Buddhism. Though the historical origins of the
character of Cloud may be obscure for players, a recognition of the legendary quality
of the narrative and its characters is evident in fan writing, as in this fan
reconstruction of Cloud’s backstory:
Sephiroth had a power unseen and unrivalled by anyone at that time. To the
people of Nibelheim, he was a living legend. All the children had dreams of
becoming as powerful as the Great Sephiroth, but Cloud was the only one with
the motivation to join SOLDIER. (Innocente, 2002)
Janet Murray makes a suggestive link between computer game characters and
Homeric heroes (Murray, 1997), citing the early 20th century scholars who revealed
the structures of the oral formulaic tradition, and pointing out that a game character
might be formulaically constructed in similar ways to the Homeric poet’s formulaic
construction of Achilles, a comparison which at least radically shifts the ground on
which conventional aesthetic objections to game texts are ritually made.
In the protagonists of both oral narrative and game, there is a predictability about their
appearance, the tools of their trade, and their actions. The dynamic of the texts is to
see how improvisatory flair on the part of the poet can stitch together and adapt the
formulae; and how the player can stitch together the given repertoires into the
sequence that will gain the desired goal. This kind of improvisatory work can be seen
in Rachel’s account of how she explores the world of Final Fantasy, how she looks
after Cloud when he’s sick, how she fights the battles with him. In both game and oral
narrative, the thing which is the text (a better expression is text-event: the particular
nexus of representation, narrative, affect, causal chain experienced at that moment) is
woven on the spot by the poet/player. The word text, Walter Ong reminds us (Ong,
2002), derives from the Latin word to weave (texere); and he further invokes the idea
of rhapsody as a possible description of oral performance, from the Greek rhapsodein,
to stitch together.
Ong’s ‘psychodynamics of oral narrative’ include a number of features which are
arguably also characteristics of games; some also apply more generally to modern
popular narratives. These include, firstly, ‘Heavy heroes’: oral narratives require
larger-than-life, stereotypical heroes who can be formulaically constructed, easily
recognised and remembered by audiences, and made to represent one or two key
characteristics. Secondly, oral narrative is ‘agonistically toned’: it revolves around
conflict externalised in the form of physical or verbal combat. Thirdly, it is
aggregative rather than analytic – narrative sequences are added and stacked up,
rather than organised hierarchically. This is related to a fourth point; that oral
narrative is high in redundancy, and in what rhetoric calls copia – it repeats the same
thing many times, in different ways, to give the listener the best chance of purchase
on it, as well as buying time for compositional effort for the performer. And finally,
oral narrative is ‘empathetic and participatory’ – the performer and audience are both
immersed in the narrative, to such an extent that, in an example given from African
narrative, the narrator slips from third to first person, his identification with the hero,
Mwinde, completed in the grammar of the telling.
The similarities between Cloud and, say, Achilles, are striking. He is formulaic – like
Achilles, he always fights in the same way, always wears the same clothes, and is
partly controlled by gods in the shape of players. Achilles is “infused with strength”
by Apollo, nourished with nectar and honey by Athena, and given high-quality
armour by the god Hephaistos. Cloud is infused with health points, and equipped with
weapons, protective devices, and magical properties by the player-as-god. He is a
‘heavy hero’: exaggeratedly attractive, good with his sword, and equipped with a
mysterious myth of origin, combining ordinary mortal and supernatural features, like
Achilles. He operates agonistically – his problems are expressed in terms of physical
combat or the overcoming of physical obstacles. He moves in a world replete with
redundancy – the experience of playing him is to keep revisiting the same places
again and again until familiarity shows us the next step; or fighting the same monsters
over and over until we learn their weak points.
In certain important ways, games obviously depart dramatically from traditions of oral
narrative. The differences of commodified and electronically-mediated culture, of
texts moving rapidly across and between global audiences, and of the dependence of
these texts on a wide range of kinds of literate practice, are among the issues that need
to be considered. The argument to be made here is not that games, in some simple
way, are a continuation of the oral tradition, but rather that its residues, in terms both
of narrative and character types, and of performative, improvisatory rhetorics, might
appear in games as what Ong describes as the ‘secondary orality’ of high-technology
societies – an evolution of the oral mindset in ways dependent on literate and
technologically-mediated culture.
Of course, this is an extremely broad historical take. In the more recent history of
popular narrative as mass-mediated discourse, we can locate Cloud in a tradition of
comic-strip heroes, specifically Japanese in this case, but belonging to a wider global
tradition of popular media with its roots in the American comicstrips of the early 20th
century, which created superheroes with dual identities which explored the banality
and anomie of urban life while projecting a flipside fantasy in costumes which were
the polar opposites of the suits worn by Clark Kent and Bruce Wayne, and in bodies
with Renaissance musculatures offering aspirational ideals to those who in real life
sported only the skinny frame of Peter Parker. (Perhaps the latter day descendant of
the comicstrip superhero as RPG avatar has no need for the dual identity, the player
providing the everyday alter ego?). The postwar manga comicstrip superheroes, and
their moving image descendants in anime and live action television and film, were
directly influenced by the US tradition, borrowing the structures of aggrandised heroic
powers and bodies, as well as dual identities; but adding specifically Japanese motifs
such as martial arts skills and weapons, enemies composed of monsters and atomic
power plants, and eventually superhero teams (see Allison, 2000). It is from this
tradition that the Final Fantasy designers descend: Final Fantasy 7 saw the arrival of a
new designer from a popular mainstream manga tradition, Tetsuo Nomura.
Though an important difference between the visual semiotic of comicstrip and film
and the oral narrative tradition is that the heroes become to some extent fixed in visual
form, they are, of course, at the same time extremely visually versatile, plastic,
adaptable. A look at variations on Batman and Superman over the years confirms this
– while retaining key iconic attributes, they adapt to suit variations in aesthetic
preference, social concerns, audience demands, in successive decades. The extreme
semiotic hybridity of games produces a more concentrated kind of variety, however.
Cloud’s appearance varies across a range of artistic and technical design contexts in
FF7, as we shall see. Furthermore, his design spills out into the fan cultures which
adopt and develop the game, so that fan art produces further variations. These vary
from fan pictures more or less faithfully adapted from the iconography of the game, to
more extreme transformations such as the production of doujinshi (amateur) manga in
the Yaoi (male/male slash) tradition. Here, texts depict eroticised relations between
Cloud and Sephiroth in which the hero and villain of the game are represented as
dominant (seme) or submissive (uke) sexual partners, their roles reversing in different
texts. (Fig. 2)
[INSERT FIGURE 2 ABOUT HERE]
The guise of the game, then, offers complex cultural resources such as the sexual
ambiguity of Cloud’s design, a booted warrior but with a feminised face and hair, so
that girl-gamers can either appropriate him affectionately, as Rachel seems to do; or
reject the stereotypical gender relations of the narrative, as this Final Fantasy girl-
gamer reviewer does while acknowledging the ambiguity of the visual design:
“What about Aeris and Tifa?!” [female characters in FF7] … you cry
indignantly. Well, simply put, they are not the lead character in the game. By
“star”, I mean a Cloud … - the person you play principally throughout the
game. Not that some of the lead characters haven’t LOOKED like girls. … I
am tired of the usual RPG plot: a (male) hero with humble roots is chosen by
fate to defend human freedom with his manly strength and courage.
(Squaregrrl, 2000)
At the same time, this is not only a narrative, but a game. Though in the guise of the
game Cloud operates as superhero protagonist, in the system of the game he
embodies, like any RPG avatar, the symbolic and technical mechanisms through
which the player performs actions within transitive sequences of the text, as the next
section will suggest. Altogether, the game operates around some of the immersive,
agonistic, episodic, aggrandised structures of both the traditional oral narrative and
modern popular superhero narrative, fusing them with the rule-based system of the
game.
Cloud – Digital Dummy?
The substance of Cloud as a larger-than-life, highly specific protagonist within the
guise of the game, then, is overlaid on the game engine’s entity module – the skeletal
set of programmed repertoires within the game as system. In this respect, as well as in
all the ambiguities of his design, he is, like all RPG avatars, what Stephen Poole has
called ‘a comparatively blank canvas’ (Poole, 2002), on which the player can project
imaginary structures of his or her own. He is a kind of puppet, and we pull his strings,
as Achilles is manipulated by the Olympian gods. When we press the Playstation
buttons or PC keys, it is this programmed entity we engage with and control.
He is a bundle of semiotic resources, or affordances for the player’s engagement with
the game’s system, equipped to move us through the game’s links and nodes,
landscapes and events. He is a set of economies: health points, hit points, experience
points, weapons and magic with quantified capacity. He is a kinaesthetic grammar,
with a limited set of actions for us to deploy – talk, walk, run, jump, get, fight.
The word ‘dummy’ raises some pertinent questions. We use it to suggest a puppet,
and one with ventriloquial qualities, though these are limited in Final Fantasy 7 – we
can actually ‘say’ very little, though we can act quite a lot. But dummy also suggests
stupidity. This is absolutely not want we want to imply; but it does usefully alert us to
to the critical view of game narratives and characters in popular perception. Janet
Murray notes this (Murray, 1997); and it is part of her strategy to compare game
characters with Homeric heroes in order to demonstrate that such characters are two-
dimensional for good reasons, demanded by the kinds of narrative of which they are
part. The further implication of this is that we cannot simply set up an opposition of
the narrative protagonist, replete with story-content, and the digital avatar, empty of
all such content. In fact, the two are closely related, as we shall see.
Of course, Cloud, the character/avatar, is both heavy hero and digital dummy.
Furthermore, the two roles, though presented here for the sake of contrast in a
polarised way, are interdependent, and leak into each other, just as the game’s system
and guise affect each other. The “Heavy hero”, for instance, is the kind of protagonist
ideally suited to be constructed by rules and formulae, being already predictable in his
behaviour and formulaic in his nature. However, though it is tempting to regard a
textual construct like Cloud as a fixed object, this would miss the point of the player-
avatar relation. This text is more like a series of processes, beginning with the design
and production of the text (itself a complex multiple articulation of different
communicative modes), which draws on the provenance of images, sounds, and
narrative patterns from both recent and distant previous cultural histories, and on a
constantly-developing game engine common to the Final Fantasy series. The process
in which any text is realised is the meeting of text and reader, or in this case, player.
One reason for comparing the playing of a computer game with a performance of oral
narrative is that it foregrounds text as event, rather than as object. It is easy enough to
find texts which behave like and materially exist as objects (print texts, film,
computer text). But some disconfirming examples, equally comparable in many
respects to computer games, raise questions about text as object: a performance of
Hamlet (as opposed to the printed play); an oral narrative; a girls’ clapping game.
Where is the text-as-object here? What is the value of regarding the performance of an
actor as an object; or the orally-transmitted formularies that oral narrators or girls
playing clapping games draw on? Or the combination of song and physical rule-
governed action that a clapping game consists of? Furthermore, the playing of games
is iterative – it is many text-events, all different, with a dynamic relation between the
computer-game as a textual resource or text in potentia, the player as a dynamic
textual element, whose fingers and skills become no less part of the game-system than
the avatar’s strings of code, and the player as cultural resource, interpreter, and
adapter of the game’s resources in the production of fan art and writing.
Playing the avatar
Cloud as Heavy Hero and Cloud as Digital Dummy offer different sets of semiotic
resources from which the player makes her experience of the avatar. The Heavy Hero,
in many respects derived from conventional narratives, and constructed through non-
interactive modes (visual design, music, animation), is largely read by the player
(along with the game guise in general). The Digital Dummy, mostly made up of
interactive textual forms, is largely played by the player (along with the game system
in general).
The sense in which the player is, and is not, the avatar, is central to the experience of
the game, and the pronoun-slippage in Rachel’s account of her experience of the game
directly represents this ambiguous relation. This ambiguity extends to the symbolic
and social meanings which might be attributed to the game-play. Agency, in the
Cultural Studies tradition, is generally presented as a positive aspect of active
readership; and it can, in this spirit, be read into avatar play also, where a simple
equation relates the degree of cultural power to the degree of control over the avatar’s
actions. However, as Perry Anderson observes (1980), agency has two opposed
meanings – one in which we are autonomous, powerful social actors; and one in
which we are merely the representative of another (as in FBI agent). Both meanings
can be read into the player-avatar relation: an unprecedented degree of participative
agency for the readers within the text, celebrated as wholly positive in Brenda
Laurel’s image of the audience moving onto the stage to become actors in the digital
play (1991); or a sense in which players merely accept and play out the roles
determined for them by game-texts devised by global corporations, dominated by
patriarchal narratives and what Sutton-Smith calls the male-dominated power
rhetorics of combative play (1997). The question of player agency in Final Fantasy 7
is quite ambiguous, and it is not clear that the dynamic experience of play directly
affects the spectrum of adoration to resistance observable in the fan sites of this, as
any non-interactive, popular text.
As mentioned above, Walter Ong demonstrates the participatory nature of oral
narrative by the pronoun slippage of the Mwinde narrator, suggesting a slide from
objective oversight of the narrative to empathetic, performative identification with the
protagonist. Similarly, Rachel’s account of Cloud, as we have seen, is characterised
by pronoun-switching. This switching is not random: the avatar is “he” when he
escapes, has flashbacks, is pursued. This is Cloud most located in the guise of the
game, most dominated by the offer aspect of the text. Midway through her account,
the mood changes: “he has to defeat”. This introduces the quest-based imperative of
the game, and the mood changes from declarative to imperative: the demand
structures of the game as system. Immediately afterwards, the pronoun switches, so
that the avatar becomes “you”: you have to go and try to stop Sephiroth, you do this
by collecting materia, and so on.
The player’s dual engagements with offer and demand structures inform each other,
producing a sense of dynamic play and of involvement with a fictional character. As
different moments in the game move more in the direction of offer or demand,
however, it seems likely that the kind of engagement will change. The battle scenes,
perhaps, are the most demand-dominated scenes, where the system of the game would
seem to be all that matters, the economies of health, hits, and magic become critical,
and the temporal elasticity of the game shrinks to realtime conflict.
Rachel’s account of the battles gives some clues about how player agency is
constituted here:
R: Well you kind of get a choice of what to do in battles, and you have to learn
how to defeat some monsters some ways and you have to learn how to defeat
them this way and you have to learn what order to put the stuff in, and it just –
it’s really quite good when you’ve built up your character because for every
battle you get – experience points – and so, um, after a while you’ve built up
your character, and so you know how to use everything more efficiently – and
it’s – the camera angles are cool too –
AB: In the battle scenes?
R: Yeah.
AB: How are the camera angles different?
R: Cos they zoom – it zooms right into your character, and they have different
angles, – one sometimes looking up at the beast, or across, or down – it’s –
really spectacular.
AB: How does it feel then, to be in that?
R: Exciting! Cos it kind of, right – what the game does is, it has a little
sequence where it actually spirals into the battle scenes, and the music changes
and the tempo changes and it really kind of, actually kind of gets you a bit
more excited.
In the relations between player and game, the agency is clear here, reflecting Rachel’s
engagement with the demand of the game. In the first part of her account, the Actor –
literally, the subject of the clauses she speaks – is the player: “You” – and the actions
you are performing are represented as imperatives, as in the triple repetition of “have
to learn”. It seems quite clear that these reflect her engagement with the procedurally-
authored system of the game.
In the second part of her account, the Actor becomes the text: “It”, and its actions are
textual ones: it “zooms”, “has different angles”, “has a little sequence”, “spirals”. The
player becomes the Goal of these actions: “it gets you a bit more excited”. This would
seem to be more to do with offer – the actions of the text here are conventional
cinematic ones, designed to position the spectator and to work for particular kinds of
affective engagement.
However, though Rachel’s account precisely represents the two-way interactive
function of the game text – you do something to it, it does something to you – the
demand/offer structures cannot be so simply separated. How do they work together;
and how are they multimodally realised?
The demand exercised by the text is realised in different ways by the different modes
combined within it. For instance, the music described by Rachel is specific to the
battle scenes; and she describes it accurately – the tempo does change (it speeds up);
and the rhythm changes to a regular 4/4 time, with the mix of midi voices including a
martial snare drum. The orientation of the music to the player, then, operates as a kind
of musical imperative – a call-to-arms, as it were. At the same time, the swirling
graphics which introduce the battle scene produce a giddy, disorientating sense, a
feeling of risk, of danger, in combination with the music. As the battle scene appears
on the screen, the player sees the characters lined up against the enemy, with the
battle statistics represented graphically at the bottom of the screen (Fig. 3). The
readiness of each character to attack is shown by a thermometer-style bar, which fills
up. This specific graphic operates, again, as a form of visual demand, effectively
instructing the player to wait, but get ready. When the bar fills up, a yellow triangle
appears above the head of the character, indicating that it can attack – a visual
imperative equivalent to “Attack now!”
[INSERT FIGURE 3 ABOUT HERE]
In ideational terms, the method of attack is very like the composition of a clause, in
strict sequential form. When the yellow arrow appears, clicking OK selects the
character – the Actor. The next choice is the means of attack, a specifying of the
process, which determines what the character will actually do – whether he will slash
with a sword, fire a lightning bolt, or throw a grenade, for instance. Finally, a white
hand appears, which can be moved by the player to select the enemy at whom the
attack is aimed – the Goal. This particular sequence, then, is a transitive structure
made up from a restricted set of elements, forming a classic restricted language of a
kind typical of many games: Halliday cites the choices available in the bidding
process in contract bridge, for instance (Halliday, 1989). In terms of the player-avatar
relation, the player’s function here is dual. In one sense, the player fuses with the
avatar – both of them are the Actor, both do the attacking; in another sense, the player
is like puppeteer, exercising a dramatic authorial function, pulling the character’s
strings; or even a kind of author, composing a sequence within a restricted language
as part of a rule-based structure of causality.
What, then, does the cinematic element add? This functions as part of the
interpersonal work of the text, positioning the player in a kind of spectatorial grammar
(Burn and Parker, 2001). Whereas in the rest of the game, we are usually positioned
above the characters in a fixed position, here we are positioned much lower down,
alongside the characters, as if fighting with them. At times, the swooping camera
angles even place us lower than the characters. This feels as if you’re fighting with
them, helping stock up health points, or re-charge their weapons.
Though this is an offer – it is distinct from the function of those parts of the text
which are demanding specific actions – it fuses with our response to those demands,
changing our sense of how we act. In effect, it mutes the puppeteer feeling that the
demand-response structures create. If we were given these powers and simultaneously
placed high above the characters, the feeling of pulling strings from a distance would
intensify. The low angles and close-ups bring the player closer to the avatar at exactly
the moment when the demand structures are at their most urgent.
While Rachel shows in this retrospective account of the game how demand structures
are represented as imperatives, there is some evidence that this kind of linguistic
transformation is even more marked during game-play. Gareth Schott’s (2002)
observation of boys engaged in collaborative play of Soul Reaver 2: the Legacy of
Kain (Eidos), showed how the language used during play was marked by a dominance
of imperative forms, functioning as what Halliday (1970) terms the regulatory mode
of language. Schott’s observational data revealed that, during game-play, players
showed no disposition toward describing the game and its objectives in narrative
terms. Instead, players maintained a preference for directive-based instructions such
as, “go there”, “jump on to one of those”, “run away from the other one fast”, “land
on that” and “push that”. Alone, auditory analysis of game-play would not have
imparted any information on the character/avatar or his position on his journey
through the clan territories of the game. Reliance upon ‘regulatory’ modes of
communication also extended to joypad action buttons, where Schott found players
commonly advised each other to press X or Y buttons rather than ‘Push’, ‘Jump’,
‘Shoot’ and so on.
Beyond the battle scenes in FF7, the feeling of offer rather than demand is reinforced
multimodally. The music of these sequences is much less stark rhythmically, either
using unmeasured rhythms or using regular duple times muted beneath flowing
melodies, which either chime with cheerful characters and locations, or evoke the
kind of mysterious sorrow which Izawa notices in Final Fantasy 3 (Izawa, 2000). In
any case, the music suggests that you’re being offered an event and a mood; if there is
any trace of demand, the modality is that of the weakened form of enticement. This
musical enticement, though part of the guise of the game, operates in tandem with the
system, which invites you to make a move.
Similarly, you explore and progress through the gameworld in a fixed camera
environment. Here, you are positioned above the action, with the avatar and other
characters rendered as chunky, polygonal figures. This design distances them from the
player; or perhaps, during these parts of the game, makes them more puppet or doll-
like, developing a tamagotchi-like relationship in which the player trains and nurtures
the avatar like a pet (one the teenagers interviewed at the same time as Rachel draws
this comparison). You are linked to Cloud by your control of his movements, but look
down on him as the Olympian gods look down on Achilles and Hector, controlling
their actions from above. (Fig. 4) This is quite distinct, as we have seen, from the
battle mode, which is temporally fixed but characterised by swooping camera
movements, locating you much closer to the avatar, as if fighting alongside him. It
may be worth remarking that this mode was also available to the gods of the Iliad, as
they could send their avatar to earth in disguise to intervene in the action, as Athene
tricks Hector while disguised as his brother Deiphobus. The Olympian position, by
contrast, is a spatial and visual reinforcement of the offer mode – it detaches the
player a little, and offers stability, unlike the destabilizing battle camera, buttressing
the demand acts of the system.
[INSERT FIGURE 4 ABOUT HERE]
However, the sense that the exploration of the game-world is characterised by a
weaker demand modality – enticement rather than command – depends on player
perception as much as on semiotic design. Janet Murray proposes two kinds of game-
labyrinth: the puzzle maze, a series of solvable obstacles which inexorably lead
towards a defined outcome; and the rhizome (modelled on Deleuze and Guattari’s
metaphor), a network of nodes and links, in which the player’s traversals are open,
unpredictable, and not constrained by linear sequences. Final Fantasy 7 seems to
offer the potential for both, to some degree. Ben, in interview 1, points out that “One
of the problems with Final Fantasy is, is, it seems – it is really linear, but they make it
seem like it’s not.” In fact, he says, there is “only one place you can go to” – so the
appearance of a world where all experiences are causally related to the narrative is an
illusion. This echoes a similar perception in a review of FF7: “As is typical of the
Japanese RPG form, the game is extremely linear. You may not see the train tracks,
but the feeling that you’ve been railroaded is unmistakable” (van Cleef, 1997).
Rachel’s experience of the game, by contrast, emphasises the rhizomic qualities: “…
it’s fantastic cos you can just explore everywhere, and you just never get bored cos
there’s just so much stuff to look around and find out, basically.”
The modality of the game seems here to be quite ambiguous. In fact, the requirement
to explore the game, insofar as it opens an interactive dialogue, is a form of demand –
in effect, “Explore!”. However, as noted above, it is a weakened demand, more of an
enticement or plea, and clearly experienced by the player as a modality in which the
agency of the player is accentuated rather than diminished.
In terms of the modality of the “explore” mode, then, it looks as though Ben reads the
game as demand: as puzzles demanding to be solved; while Rachel plays it as a weak
demand: a rhizomic world to be explored, the strong demand act being kept for key
moments of progression or battle. This ambiguity of Final Fantasy 7 contrasts with
the spatial organisation described by Diane Carr (2003) in relation to Planescape
Torment, where the rhizomic organisation of the gameworld is associated with the
structures determined by the classic D&D-derived RPG.
The most direct responses to the demand structures of the game, then – to the battle
scenes, or the nodes of the puzzle maze – are those when the player is most likely to
report their experience in the second person, when the pronoun slippage is most
likely. And these are the aspects of the game driven by the system, where the avatar is
most empty, most like a vehicle for the dynamic action of gameplay, most simple in
their characterisation, reduced to a sword, or to the sliding economies of health and
experience points. But this kind of involvement, most similar to the agonistic patterns
Ong reports of the oral tradition, is overlaid with other kinds of engagement, provided
by the offer structures of the game’s guise, marked by the third person in the player’s
account. Though the times when the text is least open to player action would seem to
offer least in terms of engagement, it is these times when the character is filled out –
when the declarative mood of the cut scene or interpolated dialogue fills out part of
Cloud’s history, his murky past, the uncertainty about his mercenary motives, his
obscure love affairs, his ambivalent relationship with Sephiroth.
Demand and offer in secondary texts
There is a kind of dialectic, then, between the demand structures of the procedural
text, agitating the causal chain, pressing you over the puzzle hurdles, catching you in
affective tensions and anxieties; and the offer structures which lay out the context,
landscape, backstory, motivation, psychology, engaging the reader-spectator in the
empathetic networks and imaginative extensions of the text which also operate in
conventional narratives. Though it is the combination of these which provides the
kinds of pleasures Rachel describes, in secondary texts surrounding the game the two
structures separate out startlingly. Analysis of fan texts is an extensive project in its
own right; but two examples will make the point clearly enough here. The first is a
walkthrough for FF7, by Kao Megura, whose fan status has become exalted by his
detailed expertise in the game to something of an independent online authority.
Because the walkthrough is generically not interested in the offer structures or the
guise of the game, it omits all reference to the backstory, lovelife, appearance, music,
and so on of the game. Its interest is purely in relaying the procedural demands of the
game-system. Accordingly, it is structured almost entirely as a demand act itself,
written in the second person, dominated by the imperative mood:
Once you leave the train, check the body of the closest guard twice to get two
Potions. Then head north. You'll be attacked by some guards.
Take them out with your sword (you may win a Potion for killing them)
and then move left to go outside. Now, talk to your teammates (Biggs,
Wedge, and Jessie), then name yourself and Barret. Make your way to the
northwestern door, and head up in the next room to enter the heart of the
power plant.
For this player, the thrill of the game seems very much bound up in his exhaustive
expertise in the properties of the puzzle maze, and in the game as system. Anything
incidental to this is omitted or reduced to minimal expression. The social motivation
for this particular development of player preference is clearly bound up in the very
public status that such a position wins in return for his hard work. His attitude to this
status is quite ambivalent, however, and is torn between obvious pleasure in the
recognition such status brings (“I recall that some other people were translating this
FAQ into Spanish, Portugese, and other languages. If they could mail me the URLs
of their translated FAQs, I'll add them here”) and exasperation with online relations
with people who don’t measure up to his notion of minimal competence: “I _WILL
NOT_ answer any gameplay related questions about this game. It's not because I'm a
prick (haha, I know), but because you wouldn't believe the types of questions I get.”
This kind of social role is comparable to the role of expert in the group of boys
observed by Schott (2002). In his account of the ‘actuality’ of play his analysis
unearthed examples of the way that game-play can be successfully orchestrated and
structured under the guidance of peers. Comparable to the metaphor of ‘scaffolding’
(Wood et al., 1976), which is used to describe the nature of support offered within
tutor student interactions, Schott provides evidence of collaborative game-play
fulfilling several key functions of tutoring. Namely, the demonstration of how to
achieve goals and highlighting critical features of the task that a solitary player may
overlook or take time to unearth. However, the main obstacle to the effectiveness of
collaborative play came from the ‘group expert’, who was unable to provide a holistic
account of the game’s structures from his gaming experience, in contrast to
observations relating to the guise of the game offered by the group’s ‘watchers’ (Orr
Vered, 1997). These deficiencies led the group expert’s consistent venture to take-
over and demonstrate his ability to respond to the procedural demands of the game-
system. Another instance of ‘expert’ making little reference to the relevant aspects of
the task beyond ‘response modality’ necessary for the movement and progression of
the character/avatar.
As against this, here is an extract from a fan spoiler from the Final Fantasy Shrine
website, in which Cloud’s story is rendered as a kind of literary narrative:
Originaly [sic], Cloud did leave Neiblhiem [sic] to join SOLDIER. However,
he was found to be unfit, and so he became nothing more than a common
grunt. During this period, he became friends with a SOLDIER member named
Zack. As luck would have it, these two were assigned to accompany Sephiroth
to the Neiblhiem reactor. Cloud, too ashamed of his failure to admit it to his
friends and neighbors, kept his mask on when they arrived.
Because this is no longer a game, the demand function is eliminated, unlike in the
walkthrough, whose function was to deliver the wisdom of the expert to steer novices
through the game. This narrative, like the cut scenes, has nothing to do with system,
everything to do with guise: it is about filling in the gaps, developing the replete,
heavy hero of popular narrative provenance, and is a communicative act of the offer
variety, entirely dominated by the declarative mood. As a monomodal text, it
combines elements which were multimodal in the game: moving image, the
procedural steps of the game, and dialogue all become transformed into written
narrative. There is no sense of interactivity: Cloud is firmly located as the Actor, and
the modes which in the game separately render his actions, his thoughts, his past and
his speech are fully integrated as game becomes story, complete with character
psychology.
Conclusion
The player-avatar relation, then, is hybrid. The engagement with the character is in
many ways developed as in conventional narratives, in response to the guise of the
game, which offers a narrative statement through an unrestricted semiotic of visual
design, animation, text and music, to compose the character as visible, audible
presence, his narrative role and affective appeal drawing on the provenance of popular
narrative, both folk and mass media. The immersive experience of roleplay, by
contrast, is engaged through the specific rule-based demands of the game, and the
player’s improvisatory deployment of the restricted set of actions offered; though this
is infused by the imaginative engagement with the character and gameworld, so that a
highly-restricted set of actions becomes elaborated and deepened by a semiotic
merger with other modes. In some cases, this happens through synchronic
syntagmatic relations, as when the system-driven movement of the avatar forms part
of a visual design made up also of background; or when the music of the battle scene
fuses with the system-driven draining of life-points. At other times, it happens
through diachronic syntagms, such as when we see an elaborate FMV of Cloud on a
motorbike, and this image leaves a residue in our minds as we play the tiny, blocky
figure in the ensuing bike chase mini-game. Perhaps, as Lemke points out (2002), this
employs different kinds of reader-perception on the part of the player – a gestalt
perception which takes in the whole of a complex synchronic syntagm; and an
iterative perception which stacks up successive sequential meanings.
Interactivity, then, means a specific combination of semiotic processes here, and an
interplay between the representational and interactive functions of the text. The
playing of the avatar means to physically assume the Actor role in the system-driven
transitive sequences of the narrative; while this experience, adrenalin-fuelled through
the time-pressure and the urgent economies of the system, is imaginatively infused by
imaginative engagement with the character as image, sound, dialogue, and popular
narrative type.
This mutual infusion, however, is not the kind we find in films or books, where a
more intense engagement would imply a deeper psychological grasp of the character –
there is no evidence that fan commitment to Cloud is any more profound in that
respect than fan art or literature devoted to Buffy the Vampire-Slayer. The difference
is in the play, where rule-based system of the game and the dramatic, performative
engagement of a secondary orality energise the familiar semiotics of the narrative
character. The system brings the kinaesthetic dynamic of play to the engagement with
the character; the guise brings the semblance of dramatic protagonist to engagement
with the avatar. We continue to read; but we make material signs in the text-event of
the game, signs which are a kind of language, a kind of action; but the language and
action of play.
References
Allison, Anne (2000) ‘A Challenge to Hollywood? Japanese Character Goods Hit the
US’, Japanese Studies, Vol. 20, No. 1, pp 67-88
Anderson, Perry (1980) Arguments within English Marxism, London: Verso
Burn, Andrew & Parker, David (2001) ‘Making your Mark: Digital Inscription and
Animation, a New Visual Semiotic’, Education, Communication, Information, Vol. 1,
No. 2, Autumn 2001
Burn, Andrew & Parker, David Parker (2002) ‘Tiger’s Big Plan: multimodality and
the moving image’, in Kress & Jewitt (eds) Moving Beyond Language: explorations
of learning in a multimodal environment, New York: Peter Lang (in press)
Carr, Diane (2003) ‘Play Dead: genre and affect in Silent Hill and
Planescape Torment’, Game Studies, Vol. 3, issue 1, May 2003
Genette, Gerard (1980) Narrative Discourse, Oxford: Blackwell
Halliday, M. (1970) “Relevant Models of Language”. Educational Review, 22: 26-37
Halliday, MAK (1985) An Introduction to Functional Grammar, London: Arnold
Halliday, MAK, (1989) Spoken and Written Language, Oxford: Oxford University
Press
Innocente, Enrique (2002) ‘Cloud’s Strife’,
www.rpgfan.com/fanfics/ffvii/cloud1.html
Izawa, Eri (2000) ‘The Romantic, Passionate Japanese in Anime: A Look at the
Hidden Japanese Soul’, in Craig, T (ed) Japan Pop! Inside the World of Japanese
Popular Culture, New York: ME Sharpe
Jenkins, Henry (1992) Textual Poachers: Television Fans and Participatory Culture,
New York: Routledge
Kress, Gunther & van Leeuwen, Theo (1996) Reading Images: The Grammar of
Visual Design, London: Routledge
Kress, Gunther & van Leeuwen, Theo (2001) Multimodal Discourses, London:
Arnold
Laurel, Brenda (1991) Computers as Theatre, Menlo Park, Ca: Addison Wesley
Lemke, Jay (2002) ‘Travels in Hypermodality’,
http://academic.brooklyn.cuny.edu/education/jlemke/papers/hypermodality/travels.ht
m
Linderoth, Jonas (2002) ‘Making Sense of Computer Games: learning with new
artefacts, conference paper at the International conference on Toys, Games and
Media, London University Institute of Education, August 2002
Megura, Kao (2000) Final Fantasy 7 walkthrough, http://www.the-
spoiler.com/RPG/Square/Final.fantasy.7.html
Murray, Janet (1997) Hamlet on the Holodeck; The Future of Narrative in
Cyberspace, Cambridge MA: MIT Press
Ong, Walter (2002) Orality and Literacy: the Technologizing of the Word, London:
Routledge
Orr Vered, K. (1997) 'Blue Group Boys Play Incredible Machine, Girls Play
Hopscotch: Social Discourse and Gendered Play at the Computer', in J. Sefton-Green
(Ed.) Digital Diversions: Youth Culture in the Age of Multimedia. London: UCL
Press
Poole, Steven (2002) ‘Character Forming’, in King, Lucien (ed) Game On: the history
and culture of videogames, London: Laurence King
Propp, Vladimir (1970) Morphology of the Folktale, Austin, TE: University of Texas
Press
Schott, Gareth (2002) ‘Moving between the Spectral and Material Plane: Interactivity
in Social Play with Computer Games’, conference paper at Playing with the Future,
Manchester University, 4th-7th April 2002
Squaregrrl (2000) Article Discussion Forum, www.womengamers.com
Sutton-Smith, Brian (1997) The Ambiguity of Play, Cambridge, MA: Harvard
University Press
Van Cleef, David (1997), Review of Final Fantasy 7, Games & Software Review,
Vol. 2, Issue 3
Van Leeuwen, T (1999) Speech, Music, Sound, London: Macmillan
Wood, D., Bruner, J.S. and Ross, G. (1976) “The Role of Tutoring in Problem
Solving”, Journal of Child Psychology and Psychiatry, 17: 89-100
FIGURE 1: CLOUD STRIFE, THE HERO OF FINAL FANTASY 7
Figure 2: Cloud and his nemesis Sephiroth depicted in a Yaoi manga comicstrip, with
Sephiroth shown as the dominant sexual partner (seme), and Cloud as the submissive
partner (uke).
Figure 3: A battle screen from Final Fantasy 7
Figure 4: the Olympian perspective in Final Fantasy 7
