Will New Media Produce New Narratives

8/8/2019 Will New Media Produce New Narratives

http://slidepdf.com/reader/full/will-new-media-produce-new-narratives 1/12

12Will New Media Produce New Narratives?

1\

II

III

li~ II

Marie-Laure Ryan

From the very beginning of the revolution that turned computers from

business machines into poetry engines, the relation between narrative and

digital media has been the object of contradictory opinions. Who should

we follow: George Landow, who claims that hypertext will reconfigure the

narrative experience by turning readers into coauthors; Janet Murray, who

regards digital media as a new stage on which old narratives will be re-played in new dimensions (as the title ofher book, Hamlet on theHolodeck,

suggests); Espen Aarseth, who thinks that the future of cybertexts lies not

in storytelling but in computer games; or Katherine Hayles, who equates

digital meaning with complexity, fragmentation, 8uidity, resistance to to-

talization, aporia, paradox, emergence, or self-organizing capabilities-

featuresmore likely to bring in a post-narrative, post-human literature than

to transform the basic conditions of narrativity?

To start this discussion of the narrative potential of digital media on solid

ground, three issues must be covered. First, we need to define narrative.



338 Ryan

The second preliminary issue concerns the distinctive properties of dig-

ital media. To make a list of these properties does not mean that digital

media form a unified field and that each of their idiosyncratic features

is available to every application. On the contrary, there are several genres

within digital textualiry, and different genres exploit different properties. I

would like to single out the folIowing fiveproperties of digital media astht'

most fundamental. I These properties affect narrativiry in either a positiw

or a negative way.

I. Reactive and interactive nature, By this I mean the abiliry of dig

ital media to respond to changing conditions. Reactivity refers \O

responses to changes in the environment or to nonintentional US('j

actions; interactivity is a response to a deliberate user action.

2. Multiple sensory and semiotic channels, or what we may call "mul

timedia capabilities," ifwe are not afraid of the apparent paradox 01

talking about multimedia media.

3. Networking capabilities. Digital media connect machines and peoplracross space and bring them together in virtual environments. '1'111_

opens the possibiliry of multi-user systems and live ("real-time") ,I~

welI as delayed communication.

4. Volatile signs. Computer memory is made of bits whose value l.11I

switch back and forth between positive and negative. Unlike book~

or paintings, digital texts can be refreshed and rewritten, widlOUI

having to throw away the material support. This properry explaillL

the unparalleled fI.uidiryand dynamic nature of digital images.

5. Modulariry. Because the compute r makes it so easy to reprodu"

data, digital works tend to be composed of many autonomous 01,jects. These objects can be used in many different contexts and ('0111

binations, and undergo various transformations, during the 1"1I111the work.

While the fulI expressive power of digital media cannot be dcsnilll d

without mentioning alI of these properties, I believe that the fil'sl 0111,

interactiviry, is the truly distinctive, and consequently fundamenlal. 0111

A novel can be digitized, made available on the Internet (propeny I). ,11111

even daily updated (properry 4) while remaining a traditional 110vd, ...

the recent publishing experiment by Stephen King has shown. Sifllilollh'cinemaoffersmultiplechannels(properry2) and fluid imagcsrh:1I11,,,11111one another easily on the screen (properry 4) j2 morcovcr. :1movi(' 1.111"

shown on the Internet (property J) withOlH signifi t'aIU ( '011$( '<)111'1111111

Will New Media ProduceNew Narratives? 339

i ts narrative potential. But, when interactiviry is added to the text or the

movie, its abiliry to telIstories, and the stories it can telI, are deeply affected.

The third issue to be addressed before we can begin our discussion is

the refinement of the concept of interactiviry. This essay will be based

on a rypology of user participation in digital media that involves two

dichotomies, internal versus external involvement and exploratory versus

ontological involvement.3

Internal/External involvement

In the internal mode users project themselves asmembers ofa virtual (or

fictional) world, either by identifying with an avatar or by apprehending

the virtual world from a first-person perspective. In the external mode

readers situate themselves outside the virtual world. They either play

the role of a god who controls the fictional world from above, or they

conceptualize their activiry as navigating a database. This opposition is

not strictly binary: the position of the user may be more or less inter-

nal or external, or the same text may give rise to different imaginativeacts. Some users will spontaneously situate themselves inside the textual

world; others prefer a distanced point ofview.

Exploratory/Ontological involvement

In the exploratory mode users are free to move around the database,

but this activiry does not make history, nor does it alter the plot; users

have no impact on the destiny of the virtual world. In the ontological

mode, by contrast, the decisions of the user send the history of the virtual

world on different forking paths. These decisions are ontological in the

sense that they determine which possible world, and consequently which

story, will develop from the situation in which the choice presents itself.

This opposition is much more binary than the preceding one, though a

hybrid casewill also be discussed here.

The cross-classification of these two dichotomies yields four rypesof user

participation in the text: internal/exploratory, internallontological, exter-

nallexploratory, and external/ontological. I do not claim that my rypology

cxhausts the field of possibilities; for instance, interactiviry can be described

aseither selective (clicking on a link) or productive (participating in a nar-

rativeaction through dialogue and gestures). Nor do I wish to saythat every

tcxt fits neady imo one of these classes:sometimes the user's role changes inrhe run of the pro~rarni somctimes the user'smode of participation can be

vicwcd ill IWo «1111"1"'111ays.I havechosenthese four categoriesbecausethcy providC'.I"IIIV, 1111'1111111111'or tht' prcscnrarion ofthc various modes



ofimcraclivc narrativity. 1krc: I wil l dis('uss fiw digital w..nn's: hY!'I'III-1

text-based virtual environmcnts, interactivc drama, compUll'" 1-\,11111'\,1111

live Internet image transmission tluough Wcbcams.

Hypertext

Bynow the idea of hypertext should be quite familiar to studcl1ls 01111.

ature: hypertexts are netWorks of textual fragments, called "Icxia" 01 11.

trons," connectedby links.Readersmovethrough the text bycliddllV,111buttons, and, sincemost fragmentscontainmanybuttons, reaclcrsh.,VI"

choice of many different itineraries. The significance of this multiplllll\

has been an object of endless theorizing. Of special relevance to our 10l'h

is the claim that, since every reading follows a different path, hypl'nol 1\

capable of endless self-regeneration. I call this interpretation thc AI('phit

conception ofhypertext, by analogy with "The Aleph," the short slmy I'1

Jorge Luis Borges in which the scrutiny of a cabalistic symbol enab"" dll

experiencer to contemplate the whole of history and of reality, dowlI 111

its most minute details.The Aleph is a small,bound objectthat CXP:IIIII.,

into an infinity of spectacles, and the experiencer could therefore dcvolI ,I

l ifet ime to i ts contemplat ion. Similar ly, hypertext has been conceivcd ,1\ ,I

matrix that expands into a multitude of texts, asreaders unravel new siri111-','of signs from its finite database of discrete lexia.

If we equate these strings of signs with "narrative," hypertext beco11H".

a machine for the production of stories, just as the grammar of a langw'r\1

is a machine for the production of sentences.It is in these terms 11..11

Michael Joyce envisions the novelty of hypertext with respect to prilll

narrative: "Reordering requires a new text; every reading thus becornt's ti

new text . . . Hypertext narratives become virtual storytellers" (193).Joyn"~

nowclassichypertextnovelafternoonallegorizeshis ideaof hypertextas .1

matrix of different stories by proposing several different versions of dllfictional world. The common theme of ali these variations is the narralOl\

witnessing of a cal'accident. In one version the accident is fatal, and tlH'

narrator's ex-wife and son are the victims. In another version the victirm

are strangers. In a third the accident isnot serious. In a fourth the narrallll

himself causes the accident. 01' everything could have been dreamed 01

hallucinated. For those who endorse the Alephic interpretation of hypcl'

text, every reading session leadsto different lexia, creates different semantil

connections between them, and consequendy construCts a different storyaround the theme of the accident.

As seductive as this conception appears-aren't we alI enamoured with

II i/I Nr'/I' M,.,It" 1'/'(I(/lIrrNr/lJ Nrll'nlll/lr.lr 34 I

lhe idc:l of ali Opt 'lI , 1IIl IM;l lIt ly Ii l, lf. transforming work?-i t cannot be

takcn lilcrally, Fil'sl, il is 1101so much because of the interactive nature

IIf hypertcxt Ihm (iftt'rnoon proposes different versions of the same evenr

bllt because Michael Joyce deliberately chose to include lexia with contra-

dictory content in his database. He could have done the same thing in a

print environment. There are indeed many postmodern novels that refuseto construct a solid actual world based on an authoritative version of facts.

Second, the conception of hypertext as a story-generating machine purs

lJuestionable emphasis on linear sequence and the narrative significance oflhe link. If we take literally the claim that every traversal of the database

determines a different story, readers who encounter three segments in the

order Athen B then c will construct a different story than readers who

cncounter the same segments in the order B then A then c. If readers

could place the information given by each lexia wherever they wanted in a

developing narrative partern, irwould not matter in which order they visit

the lexia themselves, and the sequences ABCwould yield the same Story as

Bc A.Take the caseof readers who first encounter a lexia telling them that a

certain character isdead and later discover another lexia inwhich the same

character is still alive. Readers have tWochoices. If linking and sequencing

are narratively significant, they will assume that the character has been

resurrected-an interpretation that presupposes a supernatural world that

may clash with the semantics of the text as a whole. (There is nothing

supernatural about the world of afternoon, for instance.) Alternatively, they

may decide that the sequence established by the links does not represent

causal and temporal order. They willthen treat the lexia telling of the death

as a prolepsis (flash-forward), and they will reconstruCt the same story as

teaders who encounter the tWofragments in the opposite order.

If narrativity is amental representation constrained by logical principIes,

it issimply not possible to construCt a coherent Storyout of every permuta-

tion of a set of textual fragments, because fragments are implicidy orderedby relations of presupposition, material causality,psychological motivation,

and temporal sequence.1t isonly in hypertexts with avery simple map, such

as the tree-shaped diagram that underlies the children's stories known as

Choose Your Own Adventures, that narrative continuity can bemaintained

for every traversal. On a tree diagram different readings follow different

branches, but on a given branch a lexia isalways preceded and followed by

the same lexia.This makes it easyfor the author to control the progression

of the reader and consequendy to guarantee proper logical sequence. But

the vast majoriry ofliterary hypertexts are based on more complex netWorks

that make it possible for a given lexia to appear in different contexts. The



342 Ryan

author may control the path of the reader out of a cerrain node, but aftc'l

a few transitions the path becomes unpredictable.

In keeping with his well-known theory of readers as coauehors, Georw

Landow puts the burden offilling in the logical gaps between fragments 011

readers' imaginations: "In a hyperrext environment a lack of linearity doc'\

not destroy narrative. In fact, since readers always, bue particularly in 11menvironment, fabricate their own structures, sequences or meanings, the'~

have surprisingly litde trouble reading a story or reading for a story" (11)7)

But it would take a mind with angelic-or, rather, post-human-powc'l ~

to fit lexia in a narratively coherent pattern for every order of appearamt'

For merely human minds what hyperrext offers is not a story-generatill~,

machine but something much closer to the narrative equivalent of a jigs:ly,

puzzle: readers try to construct a narrative image from fragments that COlHI'

to them in a moreor lessrandom order,by fitting eachlexiainto agloh.dpattern that slowlytakesshape in the mind. Just aswe can work for a ti lHl

on a puzzle, leave it, and come back to it later, readers of hyperrexl dllnot starr a new story from scratch every time they open the program 11111.

rather, construe a mental representation over many sessions, compk'líIlV.

or amending the pictUre pue together so faroIr is by creating what Espe'lI

Aarseth has called a "game of narration" (94), a scrambled picture Ih.1I

readers try to put back together, that hyperrext narrative takes advantagc'III

the interactive properries of its medium. Out of new syntactic featUr('~

fragmentation and linking-hyperrext thus creates a new type of discolIr'r

The role of readers in this game of narration caI}be described by the'ptlrameters of external and exploratory interactivity. Involvement is extl'f'IIód.

because readers are not cast asmembers of the textual world and becall~('11

takes a perspective akin to a god's-eye view to appreciate the design 01lIutextual network. Readers regard the text more as a database to be seard1('d

than asa world inwhich to beimmersed.4 And, in spite ofGeorge Landow',

theory of readers as coauthors, involvementis exploratory,rathl'r 1111111

ontological, because readers' paths of navigation affect not the narrallVI

events themselves bue only the way in which the global narrative palle'II'

(if there is one at all) emerges in the mind. Similarly,with a jigsaw 1'111/1.

the dynamics of the discovery differ for every player, but they do nOIaIJc'l1

the structure that is pue together. Just as the jigsaw puzzle subordinalc', Ilu

image to the construction process, external/exploratory interactivilY di

emphasizes the narrative itself in favor of the game of its discoV('ry.f\.tllHscholars(for example,Davenporrand Sloane)havcinclt'(.'dohsl'!'wd111111

hyperrextis not a good medillmfor lhe crcationof comlwlling plol\ 11,,11

live fcom sllspcnsl' and emmional p:lflicip:ltion in Ih(' lill(' 0(' dlillólelei

WillNew Media ProduceNew Narratives? 343

Thematically speaking, the external/ exploratory interactivity of classical

hyperrext isbetter suited for self-referential fiction than for narrativeworlds

that hold the reader under their spell for the sakeof what happens in them.

Ir promotes a metafictional stance, at the expense of immersion in the

fictional world. This explains in part why so many literary hyperrexts offer

a collage of literary theory and narrative fragments.5In recent years, however, hyperrext has taken a new direction that shifts

its conceptualization from the model of the scrambled narrative to whatRaine Koskimaa has called the model of the searchable archive. This new

direction is t ied to the improving multimedia capabili ties of digital sys-

tems.6In the multimedia phase hyperrext can retUrn to more solid narrative

structures, and to amore linear presentation, without reverring to the mode

of signification of the standard novel, because interactivity can now take

the form of moving from one medium to another, rather than jumping

around a texto Here I must fundamentally disagree with Roberr Coover,

who thinks that rhe golden age of digitalli teratUre carne to an end whenhyperrext ceased to be purely verbal. Hyperrext can learn from the arrist's

book, pop-up children's books, activity books, advent's calendar, and arr

CD-ROMSo spread many surprises along the visitor's way.Visible or hidden

links can be used to givethe tactile pleasure of mousing over hot spots and

of making something happen-the expansion of the textual world into a

diversified sensory experience. Readers of these texts will be cast into the

role of an investigator who digs into the history of the textUal world by

freelyexploring a collection ofdocuments. The type of topic and structUre

best suited to this idea of searching an archive will be collections of li tde

stories, such as family sagas, narratives of cultUral memory, local history

(for instance, the communal story of a village) or biography. These subjects

lend themselves parricularly well to the relatively free browsing ofhyperrext

because the story ofa lifeor a community isnot a dramatic narrative aimed

at a climax but an episodic narrative made ofmany self-sufficient units that

can be read in many orders.

Text-Based Virtual Environments (Moas and MUDS)

A text-based virrual environment is a social meeting place accessible

through a network. Users log on to the system and interact with one an-

other undcr the mask of a fictional character. This character, known in

the jargon as "avatar," iscreated by posting its description, just asa novelist

crcates cha r.1eIc'" 1i 1'011h lhe performative valuc of fictional discoursc.Thc

samc' 1I\('lIulIl" 11\1'.1ly IIH'Imildl'rs of the systelll to Cfl'all ' a "crmanent



344 Ryan

setting, typically alarge building with many rooms furnished with textualJy

described objects. In both the building of the setting and the performancc

of identities, MOOSare largely dominated by fantastic themes.7When they

are not used as platforms for serious business, they provide a forum for

free flights of fancy, black humor, and surrealist incongruities. Most 01'

the interaction that takes place on the MOOSconsists of small talk and

gestures, hardly the stuff of narrative, but this small talk easily develops

into conversational storytelling:

Carrot grinsCarrot waves

Turnip waves to Carrot

Carrot saysHi

Turnip saysWhat's up

Carrot saysWant to hear a good joke :-;

Turnip says

Carrot tells joke

The joke told by Carrot to Turnip is a standard example of diegetic sto

rytelling. Ir istold in writing but according to the real-time pressures alldstylistic conventions of oral interaction: you have to be a fast typist aswdl

as a fast mind to be a good performer on the MOOS.Even the gestul'l"

that traditionally accompany storytelling can be textually simulated. Frolll

a discourse point of view, this hybrid status between oral and wriu( '1I

communication is the truly distinctive feature of MOOstorytelling. WlwlI

the users are sufficiently imaginative, however,'MOOinteraction rises to 1lu

level of a dramatically enacted narrative. For instance:

Bek throws Panther a box, wrapped prett ily. "Open ir! I bought ir jmlfor you."

Lilypad gets the box open and takes out a puppy.

Lilypad (to Bek): She's a wonderful puppy . . . Where did you gCI11("~Bek (to Lilypad): I found her in an old warehouse. I took her hOJ11l'llId

cleaned her up. I hope you like her. Here, I have a toy for her, (Adnpll'dfrom Kolko II5)

MOOparticipant s have been known to construct imaginary ob jccrs. h( '"

the puppy-and to build elaborate scenarios around these props, WIIC'II

thi s c reat ive role playing actual ly takes place , MOOSbecomc rhc SI; lI -\(111

a collaborativelycreatednarrativeperformance,Sincc rhe panicipalll' 1111

provise this script for t heir own gratificarion. rhey are ai Ih (' snll1(' 111111

authors and specmtors. acrors and characle rs , On t ll l' prngm.1Ii l k 'Vl'1t i"

WillNewMediaProduceNewNarratives? 345

singularity of the MOOexperience can be described as an alternation be-

tween three different forms of interactivity:

· Ontological-external: creating a character or building a room by post-

ing its description. (Out-of-character behavior)

· Ontological-internal: interacting with other users by performing ac-

tions or posting dialogue. (In-character behavior)

· Exploratory-internal: wandering around the MOO,visiting rooms, andlooking at objects. (Neutral behavior)

Can we call MOOSa new form of narrative? The problem does not reside

with the very obvious novelty of the platform but with the narrativity of

the performance. MOOSreadily offer two of the three basic elements of

narrative: setting and characters. The question mark concerns the plot:

as Elizabeth Reid has suggested, MOOScreate a stage but not a script. 8

Most of the time MOOvisitors are satisfied with small talk. Ir isup to the

improvisational skills, willingness to pay roles, and cooperativeness of the

participants to produce a dramatic trajectory retellable as a story.

Interactive Drama in VREnvironments

While text-based virtual environments aremulti-users platforms, virtual re-

ality installations can only accommodate a limited number of participants.

If the technology isever perfected, VRwill enable users to take their body

into three-dimensional simulated worlds and to experience these worlds

through most oftheir senses. In the wildest dreams of developers these sim-

ulated environments will support an interactive form of drama. According

to Brenda Laurel, "The user of such systems [willbe] like audience mem-

bers who can march up onto a stage and becomevarious characters bywhat

they sayand do in their roles" (16).Janet Murray conceives the future drama

form on the model of the Holodeck ofthe popular TVshow "StarTrek." The

Holodeck isa kind of VRcave, in which the crewmembers of the starship

Enterpriseretreat for relaxation and entertainment. In this cavea computerruns a three-dimensional simulation of a fictional world, and visitors-Iet's

calI them "interactors"-become a character in a digital novel. The plot

of this novel is generated live, through the interaction between human

participants and computer-created, AI-operated virtual characters. In the

cxample discusscd by Murray, Kathryn Janeway, the female commander

ar rhc srarshipVtJY(lgc'r,neaksinto rheHolodeckand becomesLucy,rhe

gOVl"'l1l'SSf ti" ,l1l1dl'l'lIil1.111ristOcraricVicrorian houschold. Lucy f.'llIsill loVl'with tllI 1.11111'111lH'(hildn'l1. Lord Burll'y, :lnd Ihcy cxchangc



346 Ryan

passionate kisses, but the very responsible Kathryn realizes that this lovC'

for a virtual human is detrimental to the fulfi lIment of her duties in tht'

realworld, and she eventually orders the computer to delete the charactl'l,

Murray interprets this action as evidence that vR-based interactive drama

can match both the entertainment and the educational value ofliterary naI'

rative: "The Holodeck, like any literary experience, is potentially valuahlc

in exactly this way.It provides a safeplace in which to confront disturbil1~feelings we would otherwise suppress; it allows us to recognize our mOM

threatening fantasies without becoming paralyzed by them" (25).

The viability of the concept of the Holodeck as a model of a digil' ll

narrative is questionable for both technological and algorithmic reasom,

we don't have the hardware to produce truly lifelike three-dimensiol1,d

virtual worlds, and wedon't have the AIto produce complexcharactcr~,

The closest attempts so far to implement the Holodeck experiencc a"c

the projects in interactive drama currently developed at Carnegie Mcllol1

University,under the direction of JosephBates(until 1999)andMichuC'1

Mateas. These projects use a strongly Aristotelian script (folIowing tlu

curve prescribed by the Feytag triangle), and they are meant for a fiftcc'llminute visit of intense emotional involvement by a single human playcl' (II~

Mateas calls the visitor). Anything longer would strain the system as nlllcli

as the participant. Players impersonate a character and interact, mmtll '

through dialogue, with AI-animated characters. The system allows a hull

dozen plot variations, alI triggered by the behavior of the player. Mter ti111

many visits, the player will consequently feelthat alI the narrative possihil

ities are exhausted. Although the ultimate goal of developers is to stagl' tllI

projects in three-dimensional VRenvironments with fulI-body immcrsic"',

at the present t ime the interface is a computer screen, a keyboard, alld ,I

mouse. (SeeMateas and Stern for a technical description; and Ryan, chill '

10, for a narratological discussion.) Here is the plot of Mateas's ClI 'IC'IIproject.

Grace andTrip areapparentlya model couple, socialIyand final1l,tll~

successful, welI-liked byall. Grace and Trip both know the plaYI'"11'11111

work. Trip and the player are friends; Grace and the player havegOIIC"'I'1

know each other only fairly recently. Shortly after arriving at thei.. ho,",

for dinner, Grace confesses to the player that she has fallen il1lovc'wlIl,

him.Throughout the restof theevening,the playerdiscoversIhm( ; '1111

and Trip'smarriage isactualIy falling apart. Their marriage has hC'c'11IIII

for years;deep differences, buried frustrations, anellInspol<c'l1l1lidrllllc

have killcd their love for each orher, Ilow lhe VC'!1l't'1'r dwir 111111"ltllJ

"


cracks, what is revealed, and the final disposit ion of Grace and Trip's

marriage, and Grace and Trip's relationship, depends on the actions of

the player. (Mateas and Stern 2)

This plot evidently strives toward high emotional drama, but its feasibility

is questionable: how could a lifelong relationship be resolved in the fifteen

minutes allowed for the project? In Whos Afraid ofVirginia WóolfEdwardAlbee needed no lessthan two hours to break down a marriage.1t isadmit-

tedly the essence of dramatic art to make long-simmering problems reach

a crisis and resolution in the limited time frame of the stage action. But i t

would be an extraordinary achievement to bring the marital problems of

Grace and Trip to an outcome, and to do so in a believable manner, in afraction of Albee's time.

The predominantly affective nature of the plots suggested by Murray

and Mateas presents a serious emotional problem: what kind of gratif i-

cation wilI experiencers receive from becoming a character in a drama

or a story? The entertainment value of the experience depends on how

interactors relate to their avatar: wil l they be like an actor playing a role,internalIy distanciated from their character and simulating emotions they

do not reallyhave, or will they experience their character in the first-person

mode, actually feeling the love, hate, fears, and hopes that motivate the

character's behavior? The destiny of most literary characters is so unpleas-

ant that interactors would have to be out of their mind-literally and

figuratively-to voluntari ly experience it in the first person mode. If we

derive aesthetic pleasure from the tragic fate of Anna Karenina, Hamlet,

or Madame Bovary,ifwe cry for them and fulIy enjoy our tears (aswelIas

theirs), it is because our participation in the plot isa compromise between

the first-person and the third-person perspective. We simulate mentally the

inner life of these characters, we transport ourselves in imagination into

their mind, but we remain at the same time conscious of being external

observers. Any attempt to turn empathy, which relies on self-conscious

mental simulation, into first-person, genuinely felt emotion would in the

vast majority of cases trespass the fragile boundary that separates pleasure

from pain. I suspect, therefore, that the aesthetic gratification of players

of Mateas 's project will be lessa matter of emotional involvement than a

matter of curiosity about the cleverness of the system. It will take the fulI

six or SCVCI1isits for playersto appreciatethe dramaticarchitectureof the

project,

An l'VC'11111111('110'" pl'llhk'm with the idea ofbecoming a character in

1111Vt'I 01 dlllllhl 1_tI'I li" Olllili.l\iol1or IIsc:rs'frccdom or action wilh lhe



348 Ryan

creation of an aesthetically enjoyable plot. A plot isa global design, imposedtop down on the 6ctional world by a godlike author, while the actions archaracters write the story of the fictional world from within this world

itself. Characters live their life looking forward, while the author arrangcstheir destinies with an eye on the global trajectory of the plot . How caninteractors be coaxed into maintaining the plot on a proper aesthetic courscwhile acting in the name ofa 6ctional persona whose concern issurvival in

a material world, rather than living their lifeaccording to the demands 01'aesthetic teleology?]oseph Batesand hiscolleagues (Kelsoand others) haw

argued that interactive drama is meant to be played, not to be spectated.and that we judge a plot in which we part icipate by different standards

than a plot that we watch. This could mean that the cri teria applying tointeract ive drama may not be as strict as those through which we judgt 'literature and traditional drama. BUt the problem of how to script uscrs'actions in VRenvironments and gendy guide participants onto the path (li'aesthetic gratifications is far from being resolved.I believe that both the emotional and the design problem of interactiw

drama can beminimized by abandoning the idea of building a full-Aedgt'd

dramatic (that is, Aristotelian) plot around the persona of the interactOl.Most dramatic plots featUre the mind of their characters as the thealt 'r

of uncontrollable passions, and their fate as a struggle against the blindforces of dest iny. But, if we are going to enter a virtual world, i t is to 1)('

agents and not patients. This means that only selected types of emotionalexperiences, and consequently selected types of participation, will It.'ndthemselves to the 6rst-person perspective bfinteractive drama. Rather thall

becoming a character in a novel or a drama-and thereby losing tlwilidentity-interactors could play a counterpart of themselves in a forcignenvironment. Ifwe consider the whole gamut of 6ctional characters, whilh

ones would we rather emulate: (I) Hamlet, Emma Bovary, Gregor Sams,1in The Metamorphosis, Oedipus, Anna Karenina, the betrayer Brutlls illJu/ius Ceasar;or (2) the dragon-slaying hero ofRussian fairy tales, AIÍt'cInWonderland, Harry Potter, or Sherlock Holmes? Asfar as I am conccrr1t'd,I would pick a character from list (b), which means a rather fIatchal'at1"1whose contribution to the plot isnot a matter ofrich inner !ifeand illlt 'lI\I

affective experience but, rather, a matter of exploring a world, pcdlll' llling actions, solving problems, competing against enemies. and, ahovt .dl.dealing with interesting objects in a visuallystimulating environmc'nl, "'h I~kind ofinvolvement ismuch doser to playing a compUteI'gamc Ihanlollving aVictoriannovelor a Shakcspcarcandrama.011Iht:(>lht'rhand. ir1111

aUthors of thc fUture insist 011staging lhe'c'(]lIivak'llIof high lilc'raryplol~

t


in VRenvironments, interactors will be better off playing the marginal roleof observer. They will exercise their agency by navigating the virtual worldand byselecting their point ofview on the events that unfold in i t, ratherthan by being existentially entangled in these events. I see, therefore, twopossibilities for interactive drama in VRenvironments: ontological/internalinvolvement when the plot focuses on adventure and problem solvingj orexploratory/internal participation when the plot focuses on interpersonal

relations and deeply affective experiences.

'I

I

Computer Games

The third geme, compUter games, may be the least adventurous in thedomain of narrative theme and structure, bUt, asmillions of game addictshave proven, it is the most successful in terms of tUrning users into char-acters. The secret to the narrative success of games lies in their ability toexploit the most fundamental of the forces that move a plot forward: thesolving of problems. The player pursues the goal speci6ed by the game byperforming a seriesof moves that determine the destiny of the gameworld.

This destiny is created dramatically, by being enacted, rather than diegeti-cally,by being narrated. But, in contrast to standard drama, the enactmentisautotelic, rather than being directed at an observer: performing actions isthe point of the game and the main source of the player'spleasure. Playersare usually toa deeply absorbed in their task to reAect on the 1'101'hatthey write through her actions, but, when people describe their sessionswith compUter games, their reports typically take the form of a story.Consider, for instance, this review by Peter Olafson of the game CombatMission, which simulates the German campaign in Russia during WorldWar 11:"My two panzer IVGtanks gol'lucky. Approaching the crossroads,they deared a rise and caught two Sherman tanks oUt of position, oneobstruct ing the aim of the orher . Concentrat ing rheir f ire , they quickly

took oUt rhe Allied unirs and rhe surviving crews abandoned rhe Aaminghulks and rerreared into rhe woods nearby" (New J0rk Times, Ocrober 5,2000). Many people will righdy argue rhar computer games are played forrhe sake of solving problems and defeating opponents, ofre6ning srraregicskills and of participaring in online communities, and nor for rhe purposeof crearing a trace rhar reads likea story. In contras r 1'0rhe gemes discussedso far (with Ihc' possihle exceprion of socialMOas),rhe narrarivity of gamesisnot an t'nd in Ílsdfhm a means toward a goal.lOThe most sophisticatedgamcs do nol IIl'l'd 10dre'ss 111'in narrative garb to attracr players OlHOIhei.. fi, ,1I1.In ,11"~h II.IIIIC'~mdl ;ISCO, 'It,tris, Chcss, and Pac-Man lISl'rs

I

I~

I

II



350 Ryan

manipulate wholIy or partly abstract objects, and the game livesfrom thcstrategic eleverness ofits design, rather than from the imaginative impact 01'its world. The purpose of narrative scenarios isto make up for the absenct'of an original, truly superior design by providing what Kendall Walton ha~called "a prop in a game of make-believe." Scenarios create diversity Ol!the level of the imaginative experience, when rules fail to create sufficienldiversity or novelty on the level of strategy.

The importance ofthe narrative background varieswith the geme of th{'game. There isin principIe no reason whya complex fictional plot cOllldnot be presented in game form and constitute the focus of playerlreadt'l' \'interest. Players would be solving problems or accomplishing certain t:I~ll'to be allowed to get to the next episode. Experience has shown, hOWl'YrIthat the formula is not very successful. When readers are really intercsl!.din "what happens next," they do not want to find unnecessary obstadc' -thrown in their way.The narrative element of computer games istherefolltypically subordinated to the playing action. Plot is the most visiblc, alldelaborate, in the so-called RPG(role-playing) games to which I alludc ill 1111MOOsection. In these games participants spend a lot of time creating .11111

customizing their own character, they encounter many "NPGS"nonplnYII'~,characters) during their wandering in the fictional world, and the gallll~present many "cut-scenes," that is, lengthy movie elips. But the d('ve.lo!,ment of an elaborate plot cuts into the player action time, sinc{' 11111\ '1 .'elips and the dialogue of nonplaying characters can only be speclale.d 111the pure action games plot is merely a pretext for fast-paced action (havllll4something to do alIthe time seemsto be a prerequisiteforsucccs~),11111players quickly forget, in the fire of combat, the narrative purpOSl'01IIlIltmoves. Since the narrative scenario of action games is dictated by SII.IIII,I,design and since design types are limited, action games offcr 111<'1111111,.variations of the same master plots: rescue the princess from Iht' dl.I~I,OI'1savethe earth from evil aliens; disarm terrorists or be a terrorist YOIII'\dfI.

isindeed an urgent problem in the game industry to gain largl'l 'alldll 'llIby developing new narrative schemes.Computer games represent several distinct gemes, and ('hcis~1IC'111,',.

rative configuration and mode of participation must be trl'all,d "'Ihll 11..1,for each of them. Let me briefly discuss the three princip~1 typn,

Adventure Games

The best-known representatives of advenrure games are tll l' .~oI,alk.! "I~"

person shooters, such as Doam, QlIake, and II:tIr.l.ife, AdVC'lIll lIt I 'Il lItiI

iIIlIstrate the case ofinll' l' I1al and onwloHkal pnnidpatioll, PI.I}/III111'"ti'


a character in the fictional world, and their playing skilIs determine thefate of their avatar. The interaction betWeen users and the fictional world

produces a new life for the character, and consequently a new life story,for every run of the system. The preferred narrative structure of the ad-venture game is the archetypal plot of the quest of the hero, as describedby Vladimir Propp and Joseph Campbell. AsTorben Grodal has observed,these games stretch their plot endlessly in t ime through the pil ing up of

levels, episodes, and action cyeleswith similar structures. Because of theirrepetitive nature, the narrative scripts of typical adventure games wouldnever sustain interest in a nonparticipatory environment, but in this caserepetitiveness is an asset, since it is by performing the same actions overand over again that players acquire the physical skilIsnecessary to excel atthe game.Repetition, in its modular form, isalsothe adventure game's solution to

the conflict betWeen user freedom and narrative designoIr is because users'choices are quite limited in every situation and because every opportunityfor action forms a relativelyself-contained episode that games maintain theplot on the proper trajectory. In a shooting game, for instance, the choices

of players consist of the directions in which to move, of deciding whetherto shoot or to flee when an enemy appears, and, in the former case, ofselect ing and aiming weapons; The only memory needed by the systemin computing these choices is keeping track of the resources available toplayers: how many weapons, how many soldiers, are left? In the complexplots of novels , by contrast, the options of characters at every decisionpoint are both much richer and much more tighdy constrained-richerbecause their range is that of life itself but also more constrained becausethe future is produced by the past and because every life intersects with,and is influenced by, multiple other destiny lines.

Simulation Games

The elassicsof simulation games are Simcity, Civilization, Caesar, Babyz,and The Sims. Here participat ion is ontological and external. Users arecast as a powerful but not quite omnipotent god who holds the stringsof the members of a complex and dynamic system, such as a city, anl 'mpire, or a hllman group. The elements of the system react to players 'dccisions accordil1g to bllilr-in behaviors specified by artificial intelligence.dgorithms. Thl'OlIp,hdlC'I11nniplllationof individual objects players wrirethl' history 01',IIolle'(IIVIt'lIlÍty.Thc true hero of rhe srory has no COI1-sdollsness 01' 11' O\VII 11I~1"\1 11\('1\11111f Inllltiple microprocesses, TI\t'

plll'pOSt ' 01' pl,IY('I'~' , li 11011'I III IIhllll lail l IIl t' systc'l11 in a ,~ta( t' 01' 1 '( '1 .11vc'





354 Ryan

of their characters' lives, viewers sample the steady output of Webcams

in the hope of catching the truly excit ing episodes-the moments when

memorable events "walk," so to speak, into the cameràs field of vision.

While checking is spotty, the lives we imagine from our peeks into the

system are continuous and full; the members ofthe world facing the cameramay not always be visible, but they are always, in some sense, available,

since we can alwaysvisit their living space and look at the traces of their

presence.The Webcam narrative experience can be pragmatically described

as running in real time and customized by users grabbing images from ali

archive of transitory materiaIs. Its interactivity is exploratory and external,

since users look in from the outside and do not control the fate of tht'denizens of the fishbowl.

Condusion

If we opt for a universalist conception of narra tive and if we think 01

narrativein terms ofsemantic requirements, the answer to the question th:1I

forms the tide of this essay ispurely rhetorical: digital media have no moI'('

impact on the cognitive model through which we filter texts and malH'sense of human action than the experiments of postmodern fiction. Th('

texts supported by digital media may satisfy to various degrees the universal

cognitive model, or they may produce creative alternatives to a narratiVl'

experience, but they do not and cannot change the basic conditions (lfnarrativity.

But there is more to narrative theory than the formulation of basil

conditions. A complete grammar 2f lang!l\lge C~P1:ises tb.reeelemCllI\:

semantics, syntax, and .E!ag,p,atics.In narrative theory semantics beCOI1H'~

th~-s~dy ;f plot, or story; syntax beco~e~ the study ofdisc;urse, 01'1111I

rãtí~~ techniques; and pragmatics becomes the~tudy of the u~esof storytelling and of the mod,eof participatiort of llUman ag~nts in the narratiw

performance. Digital media affect narrative in three ways. (See tabk' I), Ifor a summary.)

On the pragmatic levei they offer new modes of user involvemel1talui

new things to do with narrative: exchange stories in real time; imperS0I1,1I1

a character; participate in the collective creation of a story; and explort II

world in the pursuit of a story. (See the columns labeled "type of ir lln

activity" and "user role" in table 12.1.)They also attributc variolls dt'WI'c.~of prominence to narrative in the total communicativc CVCIltsC't'1111'I~Icolumn in the table).

On thc discoursc leveithey produce I1CWways 10 IH('S('nlstoriC',\,Wllhli

c.gê..cu~.0 b.Ou C~=.- ua..=:.

11''o'

"'êcuU

"'êcuE

5]

j

§"

oS

.5

bOC.~o

..cti>



356 Ryan Will New Media ProduceNew Narratives? 357

Notes

(27-48). Which properties are considered essential depends on the purpose of thewriter aswell ason the criteria used in the selection: should these lists be restricted

to properties unique to digital media, or should they include features that these

media implement particularly efficiently but share with other media (for example,

Murray's spatiality and encyclopedic scope); should they be concerned with aspects

of technological implementation hidden from the user (for example, numerical

representation); or should they limit themselves to openly displayed features? Inmy own list I favor features that have an impact on narrativity; that are either

unique to digital media or taken by them to a new levei; and that the user can

perceive directly.

2.This holds of the screen image; the film from which the image isprojected

cannot be easilyupdated, unless it isa compute r file.

3.These two pairs are adapted from Espen Aarseth's typology of user functions

and perspectives in cybertexts, which is itself part of a broader cybertext typology

(Cybertexts 62-65). But I use different labels that shift the emphasis toward theuser's relation to the virtual world.

4. See Lev Manovich's definition of a database in the introduction to this

section.

5.For instance, Michael Joyce, afternoon; or Mark Amerika, Grammatron.

6.The best examples ofthis type ofwork are two hypertexts byM. D. Coverley,

Califia (Eastgate, 2000), and the work in progress The Book ofGoing Forth ByDay.

7. MUDstands for Multi-User Dungeon and MOOfor Multi-User Dungeon,

Object Oriented. Object Oriented refers to the programming technique.

8.The earliest MUDSwere textual game environments with a built-in plot. (The

acronym refers indeed to the role-playing game Dungeons and Dragons.) In the

I980s and I990S MOOSdeveloped into chatrooms and social meeting places, and

the system-defined plot was lost. But the idea of a combination of goal-driven,

emplotted game action and free talk was resurrected in the late I990S with enor-

mously popular games, the so-called massively multi-player role-playing games,

such as Ultima Online and EverQuest . In contrast to the earliest MUDS,these

environments offer textual communication in a graphically represented world.

Players, who number in the hundred thousands, no longer need to create their

characters through verbal description; they can construct the appearance of theiravatar from a menu of visual elements.

9. Selmer Bringsjord, a computer scientist who has developed a state-of-the-

art story-generating program called Brutus, has argued, with the support of log-

ical proofs, that AIwill never produce characters approaching the complexity of

human-generated literary characters. His argument offersa sobering rebuttal to thc

prophecies of cyber gurus such as Ray Kurzweil, who claims that by 2029 many

of the lcading artists, including novelists, will be machines (223). For Kunwcil.

howcVl'r, tlw mll"hincs mkc a shortcut that renders thc dcvdopmcnt of AIalgo.

rithms unlll'\ "~~III tllIYIII'Cahle to writc novcls bccausl.'n:1no(ed1l1olo!-;yllowb

the dOWlllollllillJ101tlu 11111111111l',till imo dil-liral dl' (' lI ilS, TIIl' miml 01' PI'OIISI

necessitate new interpretive strategies on the part of users. For instance, the

"chunking-linking" technique of hypertext, as Hayles calls it, leads to the

jigsaw puzzle mode of reading. (See the column labe!ed "discourseltech-

niques.")

On the semantic leve!, finally, the impact of digitali ty on narrative is

not a matter of deve!oping a new logic but, rather, a matter of finding

the right fit between the medi um and the form and substance of the

narrative contento Each medium has particular affinities for certain themes

and certain types of plot: you cannot tell the same type of story on the

stage and in writing, during conversation and in a thousand-page nove!, in

a two-hour movie and in a TVseria! that runs for many years. The most

urgent of the issues that faces deve!opers of new media narrative is to find

what themes and what kinds of plots take proper advantage of the built-

in properties of the medium. The fourth column of the table, themes and

structures, proposes the beginning of an answer to this questiono As my

survey has shown, combining the inherent linearity of narrative structures

with interactive protocols is not an easy thing to do, bur the task willbe much less daunting if we remember that there is no need for digital

narrative to emulate Victorian nove!s or Shakespearian drama.

If we look back at the history of narrative, we can see it has survived the

transition from orality to writing, from manuscript to print, from book to

multimedia, and from the stage to moving pictures. Each of these techno-

logical innovations has liberated new narrative energies and exploited new

possibilities. Given itswell-demonstrated resiliency, narrative should easily

weather the digital revolution. Bur I may be asking the wrong questiono

The surviva! of narrative does not depend on its abili ty to adapt itse!f to

new media; narrative has been around solong that it has little to fear from

compurers. Rather, it isthe future of new media asa form ofentertainment

that depends on their ability to deve!op their own forms of narrativity.

I. Many theorists of digital media have proposed lists of distinctive propertics,

and each of them comes up with a different l ist. But the different labels oft l' lI

cover related ideas.Janet Murray lists,for instance, the "four essential propertics 01

digital environments" as being (I) procedural (that is, being operated by computl'1'

code); (2)participatory (my "interactive"); (3)spatial (but why singleout spatialily

and omit temporality?); and (4) encyclopedic (71-90). Lev Manovich lists: (I)

numerical representation; (2) modularity (a category I borrow dirccdy from hinl):

(3) automation (Murray's "procedural"); (4) variability (my "volatility"); and (~)

transcoding (the technical property responsible for my "mulriplicilY or chanllc'ls")



358 Ryan

preserved in silicon will be able to create literary masterpieces forever.But will this

silicon Proust qualify as a machine?

10. Not alIgame developers would agreewith this statemem. For Brenda Laurel,

whose now defunct company Purple Moon developed games for girls that tried to

address issues specific to the experience of growing up female, narrative content is

not instrumental but cemral to the gaming experience. The ultimate purpose 01'

the Purple Moon games was to provide "cultural comem" through stories, asdid

myth in ancient societies (Utopian Entrepreneur 61).

11.Theresa Senft's term (qtd. byMcLemee 7).

References

Aarseth, Espen. Cybertext.PerspectivesonErgodicLiterature.Baltimore: Johns Hopkins University Press, 1997.

Arnerika, Mark. Crammatron. <http://www.grammatron.com> (April 28, 2002).

Barthes, Roland. SIZ Trans. Richard MilIer. New York: HilI and Wang, 1974.

Bringsjord, Selmer. "Is Ir Possible to Build Dramatically Compelling Digital En

tertainmem (in the form, e.g. , of compurer games)?" Camestudies I (2001)

<http://www.gamestudies.org/oIoI/bringsjord/index.html>(April 30,2002).Coover, Robert. "Literary Hypertext: The Passing of the Golden Age." <httf'."11

www.feedmag.comldocumentldo29Ilofi.html> (March 24, 2000; no longer available) .

Coverley, M. D. Califia.Watertown MA:Eastgate Systems, 2000. (Hypertext sof,ware.)

-. TheBook olCoing Forth by Day. <http://califia.hispeed.comIEgypt>

(April 28, 2002).

Davenport, Glorianna. "Your Own Virtual Storyworld." ScientificAmerican (November 2000): 79-82.

Grodal, Torben. "Stories for Eye, Ear,and Muscles: Computer Games, Media. ,11111

Embodied Experiences." The Video Came TheoryReader. Ed. Mark J. I~WIIII

and Bernard Perron. London: Routledge, 2003. 129-55.

Hayles, N. Katherine. "The Transformation ofNarrative and the Matcrialil)1 111

Hypertext." Narrative 9.1(2001): 21-39.

Joyce, Michael. Afternoon, a Story.Watertown MA:Eastgate Systems, 1987. (llypl'ltext software.)

-. 0ITwo Minds: Hypertext, Pedagogy,and Poetics.Ann Arbor: UniVl'I'NiI'"

Michigan Press, 1995.

Kelso, Margaret Thomas, Peter Weyhrauch, and Joseph Bates. "Dr:lImuil PII_

ence." Presence:Teleoperatorsand Virtual Environments 2.1 (1993):I I~.

Kolko,Beth. "Buildinga World with Words:The NarrativeRcaJity01'Vil ""Ij

Communities." Wórksand Days 13.1-2 (1995):105'-26.

Koskimaa,Raine.DigitalLiterarurc."FromlCxIto Hyperlt'xl,111.1Bt'YOIIII."PieIIdiss.,UnivcrsityorJyv1iskylHFinl:tnd). 2000.


Kurzweil, Ray. TheAge olSpiritual Machines. New York:Viking, 1999.

Landow, George P.Hypertext 2.0: The ConvergenceolContemporary Critical Theory

and Technology.1992.Reprim. Baltimore: Johns Hopkins University Press,1997.

Laurel, Brenda. Computersas Theatre.Menlo Park CA:Addison- Wesley, 1991.

-. Utopian Entrepreneur.Cambridge: MITPress, 2001.

Mateas, Michael, and Andrew Stern. "Towards Imegrating Plot and Character for

Imeractive Drama." Working Notes of the Social Imelligence Agems: The Hu-

man in the Loop Symposium. AAAIFalI Symposium Series.Menlo Park CA:AAAI

Press, 2000. Version used here online at: <http://www-2.cs.cmu.edu/-michaelm/

publicationslsIA2ooo.pdJ>(April 24, 2002).

McLemee, Scott. "I Arn a Camera." Lingua Franca (February 2001): 6-8.

Murray, Janet E. Hamlet on the Holodeck: The Future 01Narrative in Cyberspace.

New York: Free Press, 1997.

The Nerdman Show. <http://www.nerdman.com> (April 28, 2002).

Olafson, Peter. "Game Theory." New YorkTimes, October 5,2000.

Ryan, Marie-Laure. Narrative as Virtual Reality: lmmersion and lnteractivity in

Literature and ElectronicMedia. Baltimore: Johns Hopkins University Prcss,2001.

Sloane, Sarah. Digital Fictions:Storytellingin aMaterial Wórld.Stamford CT:Ablcx.2000.

Walton, Kendall. Mimesis asMake-Believe: On the Foundations 01the Representtto

tionalArts. Cambridge: Harvard University Press, 1990.

White, Hayden. "The Value ofNarrativity in the Represemation ofReality." On

Narrative. Ed. W. J. T. Mitchell. Chicago: University of Chicago Press, 1980.

1-24.

Documents

Will New Media Produce New Narratives