The Third Element; or, How to Build a Storyworld

(iii) Worldmaking/world disruption. The events represented in

narrative are such that they introduce some sort of disruption or

disequilibrium into a storyworld involving human or human-like

agents, whether that world is presented as actual or fictional, realistic

or fantastic, remembered or dreamed, etc.


Narratives as Blueprints for Worldmaking

Storyworlds can be defined as the worlds evoked by narratives; reciprocally, narratives can be
defined as blueprints for a specific mode of world-creation. Mapping words (or other kinds

of semiotic cues) onto worlds is a fundamental – perhaps the fundamental – requirement for
narrative sense-making; yet this mapping operation may seem so natural and normal that no

“theory,” no specialized nomenclature or framework of concepts, is necessary to describe and
explain the specific procedures involved. In the present chapter, I argue for the need to slow down
and de-automatize the rapid, apparently effortless interpretive processes involved in experiencing
narrative worlds. Exploring the third basic element of narrative necessitates taking the measure of
these processes – that is, identifying what is distinctive about narrative ways of worldmaking as
opposed to other methods for using symbol systems to make and unmake worlds.


The classical, structuralist narratologists failed to come to terms with the referential or
world-creating properties of narrative, partly because of the exclusion of the referent in favor of
signifier and signified in the Saussurean language theory that informed the structuralists’
approach (see chapter 2). Over the past couple of decades, however, one of the

most basic and abiding concerns of narrative scholars has been how readers of print narratives,
interlocutors in face-to-face discourse, and viewers of films use textual cues to build up
representations of the worlds evoked by stories, or storyworlds. Such worldmaking practices are of
central importance to narrative scholars of all sorts, from feminist narratologists exploring how
representations of male and female characters pertain to dominant cultural stereotypes about gender
roles, to rhetorical theorists hypothesizing about the kinds of assumptions, beliefs, and attitudes
that must to be adopted by readers if they are to participate in the multiple audience positions
required to engage fully with fictional worlds, to analysts (and designers) of digital narratives
interested in how interactive systems can remediate the experience of being immersed in the virtual
worlds created through everyday narrative practices. New ways of characterizing the third basic
element of narrative, its intrinsic concern with more or less richly detailed

storyworlds, have arisen from this re-engagement with the referential, world-creating potential of
narrative. That re-engagement has received additional impetus from foundational theoretical studies
of narrative worlds – studies that I discuss later in this chapter and that draw on

ideas developed by philosophers, psychologists, linguists, and others concerned with how people use
various kinds of symbol systems to refer to aspects of their experience. In parallel with the
account developed in Herman (2002a: 9–22), I use the term storyworld to refer to the world evoked
implicitly as well as explicitly by a narrative, whether that narrative takes the form of a

printed text, film, graphic novel, sign language, everyday conversation, or even a tale that is
projected but never actualized as a concrete artifact – for example, stories about ourselves that
we contemplate telling to friends but then do not, or film scripts that a screenwriter has plans

to create in the future. Storyworlds are global mental representations enabling interpreters to
frame inferences about the situations, characters, and occurrences either explicitly mentioned in
or implied by a narrative text or discourse. As such, storyworlds are mental models of the
situations and events being recounted – of who did what to and with whom, when, where, why, and in
what manner. Reciprocally, narrative artifacts (texts, films, etc.) provide blueprints for the
creation and modification of such mentally configured storyworlds.


1 Storytellers use the semiotic cues available in a given narrative medium to design these
blueprints for creating and updating storyworlds.


2 In print texts, the cues include the expressive resources of (written) language, including not
just words, phrases, and sentences, but also typographical formats, the disposition of space on the
printed page (including spaces used for section breaks, indentations marking new paragraphs, etc.),
and (potentially) diagrams, sketches, and illustrations. In graphic novels such as Ghost World, by
contrast, the nonverbal elements play a more prominent role: the arrangement of characters in
represented scenes, the shapes of speech balloons, and the representations of the scenes in panels
that form part of larger sequences of images and textual elements, can convey information about the
storyworld that would have to be transmitted by purely verbal means in a

novel or short story without a comparable image track. Likewise, interlocutors in contexts of
face-to-face storytelling, readers of short stories and novels, and members of the audience
watching a film draw on such medium-specific cues to build on the basis of the discourse (or
sjuzhet) a chronology for events (or fabula) (what happened when, or in what order?); a broader
temporal and spatial environment for those events (when in history did these events occur, and
where geographically?); an inventory of the characters involved; and a working model of what

it was like for these characters to experience the more or less disruptive or noncanonical events
that constitute a core feature of narrative representations, which may in turn be more or less
reportable within a particular discourse context or occasion for telling.


3 At the same time, as discussed in chapter 3, interpreters seeking to build a storyworld on the
basis of a text will also take into account complexities in the design of the blueprint itself –
complexities creating additional layers of mediation in the relationship between narrative

and storyworld. Such mediation affects the interpretive process in, for example, cases of
unreliable narration such as Browning’s My Last Duchess, where the teller of a story cannot be
taken at his or her word, compelling the audience to “read between the lines” – in other words,

to scan the text for clues about how the storyworld really (or probably) is, as opposed to how the
narrator says it is. Likewise, in Ghost World, during a sequence in which Enid fantasizes about one
of her teachers, Mr. Pierce, the use of a distinctive font or typeface within the speech balloons
(not to mention the content of the sequence – e.g., Enid naked in the shower with Mr. Pierce clad
in a formal suit) indicates that the represented scenes and utterances are ones that Enid has
imagined, rather than events that took place within the storyworld to which the characters orient
as actual or real (Clowes 1997: 32). Both of these examples entail complex processes of
worldmaking. For its part, the Browning poem compels readers to sift out from the Duke’s
elliptical, distorted version of events a divergent or rather more complete account of what

happened, affording through these indirect means a blueprint for building the domain of factual (or
at least probable) occurrences. The world that emerges through this process is one in which the
Duke, despite or rather because of his own best efforts at spin or damage control, figures

as an insanely jealous, homicidally possessive, and controlling spouse. Meanwhile, Enid’s erotic
fantasy demonstrates in another way the multifacetedness of storyworlds, which typically encompass
not just worlds that are socially and institutionally defined as “given” but also

private worlds (Ryan 1991) or subworlds (Werth 1999) consisting of characters’ beliefs, desires,
intentions, memories, and imaginative projections. Some of these subworlds may never be expressed
outwardly to other characters, as is likely the case with Enid’s fantasy – hence

Clowes’s use of a typeface that distinguishes this sequence from other conversational exchanges
represented in the text. But what would a more general account of how narratives evoke storyworlds
look like? And how do narrative ways of worldmaking differ from other representational practices
that involve the construction or reconstruction of worlds, in a broad sense? In other words, when
it comes to world-creation, what distinguishes narrative representations from other contexts in
which people design and manipulate symbol systems

for the purpose of structuring, comprehending, and communicating aspects of experience? I explore
these issues in my next section.


Narrative Ways of Worldmaking

To capture what is distinctive about narrative ways of worldmaking, this section begins with an
overview of Goodman’s (1978) broad account of “ways of worldmaking.” The building of

storyworlds involves specific procedures set off against this larger set of background conditions
for world-creation. I start to outline these procedures by developing an account of narrative
beginnings as prompts for worldmaking. This in turn sets up my next section, where I survey a range
of approaches to world-creation in narrative contexts, moving from accounts that characterize the
experience of narrative worlds in a relatively macrostructural or gestalt way toward more
microstructural approaches that seek to anchor types of inferences about storyworlds (including
their temporal and spatial dimensions) in particular kinds of textual designs. In his study Ways of
Worldmaking, the philosopher Nelson Goodman develops ideas that afford context for my own analysis.
Adopting a pluralist instead of a reductionist stance, Goodman argues that “many different
world-versions are of independent interest and importance,

without any requirement or presumption of reducibility to a single base” (Goodman 1978: 4), for
example, the world-version propounded in physics. As Goodman puts it, “[t]he pluralists’ acceptance
of [worldversions] other than physics implies no relaxation of rigor but a recognition that
standards different from yet no less exacting than those applied in science are appropriate for
appraising what is conveyed in perceptual or pictorial or literary versions” (1978: 5). More
generally, Goodman asks, In just what sense are there many worlds? What distinguishes genuine

from spurious worlds? What are worlds made of? How are they made? What role do symbols play in the
making? And how is worldmaking related to knowing? (Goodman 1978: 1)


Arguing that worldmaking “as we know it always starts from worlds already on hand; the making is a
remaking,” Goodman goes on to identify five procedures for constructing worlds out of other worlds:
composition and decomposition; weighting; ordering; deletion and supplementation; and deformation
(1978: 7–16). Brief definitions andexamples of each procedure follow:


• Composition and decomposition: “on the one hand . . . dividing wholes into parts and partitioning
kinds into subspecies, analyzing complexes into component features, drawing distinctions; on the
other hand . . . composing wholes and kinds out of parts and members and subclasses, combining
features into complexes, and making connections” (1978: 7). Ethnographic investigation of an
indigenous population, for example, may uncover the presence of several subcultures where only one
had been recognized previously; conversely, the formation of new “hybrid” disciplines or
subdisciplines (algebraic geometry, biochemistry, information design) results in new, more complex
world-versions.


• Weighting: “Some relevant kinds of the one world, rather than being absent from the other, are
present as irrelevant kinds; some differences among worlds are not so much in entities comprised as
in emphasis or accent, and these differences are no less consequential” (1978: 11). From a
macrohistorical perspective, the shift from a religious to a secular-scientific world-version
entailed a reweighting of the particulars of the phenomenal world, which came to occupy

a focus of attention formerly reserved for the noumenal or spiritual realm.


• Ordering: “modes of organization [patterns, measurements, ways of periodizing time, etc.] are not
‘found in the world’ but built into a world” (1978: 14). As suggested in chapter 4, taxonomies of
plants,animals, or other entities are in effect world-versions built on a hierarchical systems of
categories that may be more or less finely grained(and more or less densely populated), depending
on whether onehas expert or only a layperson’s knowledge of a given domain. My

world-version currently contains names for (and concepts of) only a few common types of insects, in
contrast with the world-version of an entomologist.


• Deletion and supplementation: “the making of one world out of another usually involves some
extensive weeding out and filling – actual excision of some old and supply of some new material”
(1978: 14). I might study entomology, and supplement my world-version with new knowledge and new
beings; alternatively, if because of climate change an insect species becomes extinct, the
entomologist’s world-version will undergo compulsory excision.


• Deformation: “reshapings or deformations that may according to point of view be considered either
corrections or distortions” (1978: 16). Here one may think of arguments for a new scientific theory
in favor of an older one (e.g., the geocentric vs. the heliocentric models of the solar system)
from the perspective of those who are parties to the debate.


As my examples of each worldmaking procedure indicate, there is nothing distinctively story-like
about the worlds over which Goodman’s account ranges, though there is nothing about the analysis
that excludes storyworlds, either. Narrative worlds, too, might be made through processes of
composition and decomposition: think of allegories fusing literal and symbolic worlds, or
decomposition in texts such as The Canterbury Tales, where the narrative ramifies into a frame tale
that constitutes the main diegetic level and embedded or hypodiegetic levels created when
characters within that frame tell stories of their own. Weighting may also be a generative factor:
consider postmodern rewrites that evoke new world-versions by reweighting events in their precursor
narratives, as when Jean Rhys’s Wide Sargasso Sea generates a new storyworld on the basis of
Charlotte Brontë’s Jane Eyre by using as a metric for evaluating events not Jane Eyre’s or Edward
Rochester’s perspective (as refracted through Jane’s telling) but rather Antoinette Cosway’s. So
too with ordering: narrative worlds can be made when new time-scales are deployed, as when Alain
Robbe-Grillet as a practitioner of the nouveau roman in France produced novel worlds by drastically
slowing the pace of narration (Robbe-Grillet [1957, 1959] 1965), or when the average shot length in
Hollywood films diminished over time to produce more rapid cuts between scenes (Morrison
forthcoming). Deletion and supplementation likewise find their place in the building of
storyworlds. I may tailor my recounting of my own life experiences to adjust for differences among
groups of interlocutors, going into more detail among close friends and less detail when asked a
question during a job interview. And as for deformation, the film version of Ghost World can be
viewed as a reshaping of the graphic novel version, and more generally any adaptation of a prior
text in another medium for storytelling will result in alterations of the sort that Goodman
includes under this rubric (cf. Genette [1982] 1997).


In short, Goodman’s is a broad, generic account of worldmaking procedures, operative in both
non-narrative and narrative contexts. The basis for distinctively narrative ways of worldmaking
must thus be sought in other, more specific procedures set off against this larger set of
background conditions for world-creation. In my next subsection, I discuss how story openings
trigger particular kinds of worldmaking strategies that cut across storytelling media and narrative
genres, but that are also inflected by the specific constraints and affordances

of various kinds of narrative practices.