NARRATOLOGY

Introduction to the Theory of Narrative

Fourth Edition

Since its first publication in English in 1985, Mieke Bal’s Narratology has
become the international classic introduction to the theory of narrative
texts. Narratology is a systematic account of narrative techniques and
methods, and their transmission and reception, in which Bal distills
years of study of the ways in which we understand literary works.
In this fourth edition, Bal updates the book to include a greater
focus on literary narratives while also sharpening and tightening her
language to make it the most readable and student-friendly edition to
date. With changes prompted by ten years of feedback from scholars
and teachers, Narratology remains the most important contribution to
the study of the way narratives work, are formed, and are received.

MIEKE BAL is a cultural theorist, video artist, and Professor Emeritus

in Literary Theory at the University of Amsterdam. Her website is at
www.miekebal.org.
MIEKE BAL

NARRATOLOGY
Introduction to the Theory of Narrative

Fourth Edition

UNIVERSITY OF TORONTO PRESS

Toronto Buffalo London
© University of Toronto Press 1985, 1997, 2009, 2017
Toronto Buffalo London
www.utppublishing.com
Printed in Canada

ISBN 978-1-4426-50329 (cloth)

ISBN 978-1-4426-28342 (paper)

First Edition 1985

Reprinted 1988, 1992, 1994
Second Edition 1997
Reprinted 1999, 2002, 2004, 2006, 2007
Third Edition 2009
Reprinted 2012, 2014
Fourth Edition 2017
This fourth edition is a substantial revision of Narratology: Introduction to the Theory
of Narrative, which was a translation, revised for English-language readers, by
Christine Van Boheemen, of the second, revised edition of De Theorie van vertellen en
verhalen (Muiderberg: Coutinho 1980).

Printed on acid-free, 100% post-consumer recycled paper with vegetable-based inks.

______________________________________________________________________________

Library and Archives Canada Cataloguing in Publication

Bal, Mieke, 1946−

[Theorie van vertellen en verhalen. English]
Narratology : introduction to the theory of narrative / Mieke Bal. −
Fourth edition.

Translation of: De theorie van vertellen en verhalen.

“This fourth edition is a substantial revision of Narratology: Introduction
to the Theory of Narrative, which was a translation, revised for English-
language readers, by Christine Van Boheemen, of the second, revised
edition of De Theorie van vertellen en verhalen.”
Includes bibliographical references and index.
ISBN 978-1-4426-5036-7 (hardcover) ISBN 978-1-4426-2837-3 (softcover)

1. Narration (Rhetoric). I. Boheemen, Christine van, translator

II. Title. III. Title: Theorie van vertellen en verhalen. English.

PN212.B313 2017 808.3’93 C2017-903696-3

______________________________________________________________________________

University of Toronto Press acknowledges the financial assistance to its publish-

ing program of the Canada Council for the Arts and the Ontario Arts Council, an
agency of the Government of Ontario.

an Ontario government agency

un organisme du gouvernement de l’Ontario

Funded by the Financé par le

Government gouvernement
of Canada du Canada
Introduction

I begin with a series of working definitions. My purpose in doing this is to

empower the student of narrative, who can fall back on such defini-
tions, test them against analyses and interpretations, and check their
consistency. They are meant not to hold the truth of their object but
rather to make it accessible and discussable.
Narratology as a field of study is the ensemble of theories of narra-
tives, narrative texts, images, spectacles, events – of cultural artefacts
that tell a story. Such theory helps us understand, analyse, and evalu-
ate narratives. A theory is a systematic set of generalized statements
about a particular segment of reality. That segment of reality, the cor-
pus, about which narratology attempts to provide insight consists of
narrative texts of all kinds, made for a variety of purposes and serving
many different functions.
If characteristics of narrative texts can be defined, these characteris-
tics can serve as the point of departure for the next phase: a description
of the way in which each narrative text is constructed. Then we have a
description of a narrative system. On the basis of this description, we can
examine the variations that are possible when the narrative system is
concretized into narrative texts. This last step presupposes that an infi-
nite number of narrative texts can be described using the finite number
of concepts contained within the narrative system.
Readers of this book are offered an instrument with which they can
describe and, hence, interpret narrative texts. This does not imply that
the theory is some kind of machine into which one inserts a text at one
end and expects an adequate description to roll out at the other. The
concepts that are presented here must be regarded as intellectual tools
for interpretation. These tools are useful in that they enable their users
4 Narratology

to formulate an interpretive description in such a way that it is acces-

sible to themselves as well as to others. Furthermore, discovering the
characteristics of a text can also be facilitated by insight into the abstract
narrative system. But above all, the concepts help increase understand-
ing by encouraging readers to articulate what they understand, or think
they understand, when reading or otherwise processing a narrative
artefact.
The textual description achieved with the help of this theory can
by no means be regarded as the only adequate description possible.
Someone else may use the same concepts differently, emphasize other
aspects of the text, and, consequently, produce a different description.
This variation is inevitable and felicitous, because reading is a funda-
mentally subjective activity. The point is, if the description of a text is
understood as a proposal that can be presented to others, the fact that
the description is formulated within the framework of a systematic the-
ory facilitates discussion of the proposed description. This is a demo-
cratic use of a theory. This democratic nature of the joint activities of
analysis, description, and interpretation I call intersubjectivity.
Of what does this corpus of narrative texts consist? At first glance,
the answer seems obvious: novels, novellas, short stories, fairy tales,
newspaper articles, and more. But, intentionally or not, we are estab-
lishing boundaries – boundaries with which not everyone would agree.
Some people, for example, argue that comic strips belong to the corpus
of narrative texts, but others disagree. The underlying difference may
be that, for the proponents, the fictional nature of the narrative is the
standard; for the antagonists, the conception of literature inheres in the
idea of narrative texts. Neither of these underlying ideas pertain to nar-
rative, however. If these people hope to reach agreement, they will wish
to explain how they have arrived at their decisions. And once they try,
more differences will come up. For example, those who consider comic
strips to be narrative texts interpret the concept text broadly. In their
view, a text does not have to be a linguistic text. In comic strips, another,
non-linguistic, sign system is employed – namely, the visual image.
Others, sharing a more restricted interpretation of what constitutes a
text, reserve this term for texts presented solely in language.
As this simple example demonstrates, it helps to define the concepts
we use and to restrict their content to singular ideas. This is not always
obvious; think of very common and seemingly obvious notions such as
literature, text, narrative, and poem; and art, popular culture, context. A dis-
agreement about the status of comic strips would quickly be settled if
 Introduction 5

the definition of a text were first agreed on. In other words, definitions
are like a language: they provide something of a dictionary, so that one
person understands what another means. But the definitions proposed
here are provisional, serving before all else the purpose of being explicit
and transparent enough that we can know what is relevant and what is
not for the discussion at hand.
Presenting a theory about narrative texts entails defining a number
of central concepts. Within the scope of this introduction, then, a text
is a finite, structured whole composed of signs. These can be linguistic
units, such as words and sentences, but they can also be different signs,
such as cinematic shots and sequences, or painted dots, lines, and blots.
The finite ensemble of signs does not mean that the text itself is finite,
for its meanings, effects, functions, and backgrounds are not. It only
means that there is a first and a last word to be identified, a first and a
last image of a film, a frame of a painting – even if those boundaries, as
we will see, are provisional and porous.
For this book I have selected the following definitions. A narrative text
is a text in which an agent or subject conveys to an addressee (“tells”
the reader, viewer, or listener) a story in a medium, such as language,
imagery, sound, buildings, or a combination thereof. A story is the con-
tent of that text and produces a particular manifestation, inflection, and
“colouring” of a fabula. A fabula is a series of logically and chronologi-
cally related events that are caused or experienced by actors. These three
definitions together constitute the theory this book elaborates.
These key concepts imply other ones. Take the last one, the fabula, for
example. Its definition contains the elements “event” and “actor.” An
event is the transition from one state to another state. Actors are agents that
perform actions. They are not necessarily human. To act is defined here as
to cause or to experience an event. And this series of definitions can go on.
The basis of this theory’s usefulness for analysis is the three-part
division it proposes. The assertion that a narrative text is one in which
a story is told implies that the text is not identical to the story, and the
same holds for the relationship between story and fabula. Take the
familiar fairy tale of Tom Thumb, about a small boy who outsmarts
a dangerous ogre. Not everyone has read that story in the same text.
There are different versions, that is, different texts in which that same
story is related. Some texts are considered to be literary while others
are not; some can be read aloud to children while others are too difficult
or too frightening. Narrative texts differ from one another even if the
related story is more or less the same. Here, “text” refers to narratives in
6 Narratology

any medium. I use this word with an emphasis on the finite and struc-
tured nature of narratives, not the linguistic nature of text.
“Tom Thumb” can also help explain the next distinction, that between
story and fabula. This distinction is based on the difference between the
way in which the events are presented and the sequence of events as they
occur in the imagined world of the fabula. That difference lies not only
in the language used. Despite their having read different texts, readers
of “Tom Thumb” would agree, I expect, as to which of the characters
deserves sympathy. They applaud the clever boy, and they rejoice at the
ogre’s misfortunes. In order that Tom might triumph over his enemy, read-
ers are quite prepared to watch unabashedly as Tom exchanges crowns
so that the blind ogre unwittingly eats his own children. Evidently, in all
of the texts this rather cruel fabula is presented in such a way that read-
ers are willing to sacrifice one group of children for another. When “Tom
Thumb” is told in another sign system – in a cartoon film, for example –
similar reactions are evoked. This phenomenon demonstrates that some-
thing happens with the fabula that is not exclusively language-based.
These definitions suggest that a three-layer distinction – text, story,
fabula – is a good basis for the study of narrative texts. Such a distinc-
tion entails that it is possible to analyse the three layers separately. That
does not mean that these layers exist independently of one another. They
do not. The only material we have – that can be said to exist – is the text
before us. Readers have only the book, paper and ink, or the strokes of
paint on a canvas, the light in a dark (movie) theatre, the sound coming
out of speakers, and they must use this material to establish the structure
of the text themselves. Only the text layer, embodied in the sign system
of language, visual images, or any other, is materially accessible. That a
text can be divided into three layers is a theoretical supposition based on
a process of reasoning, of which I have given a summary above.
Layers serve as instrumental and provisional tools to account for par-
ticular effects the text has on its readers. The theory being presented in
this introduction is based on the notion of distinct layers, such distinction
being necessary for a detailed analysis. It is, therefore, inevitable that what
is in effect inseparable should temporarily be disjoined. This goes to show
that reading and analysing are distinct activities, even if they cannot be
separated either. In order to analyse, one needs to read; and reading inevi-
tably entails interpretive moments that will inform an analysis made later.
As with the three layers, the activities involved in the processing of narra-
tives cannot be separated, but must be distinguished in theory in order for
us to understand the process.
 Introduction 7

The fabula, understood as material or content that is worked into a

story, has been defined as a series of events. This series is constructed
according to certain rules. We call this the logic of events. If human
behaviour is taken as the criterion for describing events, then the ques-
tion of the function of the agents of action, the actors, arises. The actors
can be described in relation to the events. However, two other elements
in a fabula can be discerned. An event, no matter how insignificant,
always takes up time. This time is often important for the continua-
tion of the fabula and deserves, consequently, to be considered. If Tom
Thumb had not had seven-miles boots at his disposal, he would never
have been able to escape from the ogre in time. The difference between
the time that Tom needs to escape from the giant’s grasp and the time
that the giant needs to wake up is, in this case, decisive for the happy
ending of the fabula. Furthermore, events always occur somewhere, be
it a place that actually exists (Amsterdam) or an imaginary place (C.S.
Lewis’s Narnia). Events, actors, time, and location together constitute
the material of a fabula. I will refer to these as elements.
These elements are organized in a certain way into a story. Their
arrangement in relation to one another is such that they can produce
the effect desired, be this convincing, moving, disgusting, pleasing, or
aesthetic. Several processes are involved in ordering the various ele-
ments into a story. These processes are not to be confused with the
author’s activity – it is both impossible and useless to generalize about
the latter. I distinguish the following:

1 The events are arranged in a sequence that can differ from the
chronological sequence.
2 The amount of time allotted in the story to the various elements of
the fabula is determined with respect to the amount of time these
elements take up in the fabula.
3 The actors are provided with distinct traits. Thus they are individu-
alized and transformed into characters.
4 The locations where events occur are also given distinct characteris-
tics and are thereby transformed into specific places.
5 Other relationships – symbolic, allusive, traditional – may exist
among the various elements.
6 A choice is made from among the various points of view from
which the elements can be presented. The resulting focalization, the
relation between who perceives and what is perceived, colours the
story with subjectivity.
8 Narratology

The result is a specific story that is distinct from other stories. I will refer
to the traits that are particular to a given story as aspects.
A fabula that has been ordered into a story is still not a text. A nar-
rative text is a story that is told, conveyed to recipients, and this tell-
ing requires a medium; that is, it is converted into signs. An agent who
relates, who utters the signs, produces these signs. This agent cannot be
identified with the writer, painter, composer, or filmmaker. Rather, the
writer withdraws and calls upon a fictitious spokesman, the narrator.
But the narrator does not relate continually. Whenever direct speech
occurs in the text, it is as if the narrator temporarily yields this function
to one of the actors. When describing the text layer, the key question is
who is doing the narrating.
A text does not consist solely of narration in the specific sense. In
every narrative text, one can point to passages that concern something
other than events, such as an opinion about something, for example,
or information by the narrator that is not directly connected with the
events – perhaps a description of a face or of a location. It is thus possi-
ble to consider what is said as narrative, descriptive, or argumentative.
Such an analysis helps us assess the ideological or aesthetic thrust of a
narrative. There is often a noticeable difference between the narrator’s
style and that of the actors. As a result of this division into three parts,
some topics that traditionally constitute a unified whole will be treated
separately in different stages of this book. An example of this is the
agent who performs the activities pertinent for each layer. This agent
is called “narrator” or “speaker” in the study of the text, “character” in
the study of the story, and “actor” in the study of the fabula.
Ideally, the characteristics of narrative text should be as follows:

1 Two types of “speakers” utter the signs that constitute a narrative

text; one does not play a role in the fabula whereas the other does.
The narrator who does not appear and never refers to itself, and
thus is reduced to being a voice, has the same status as one who
constantly interferes, comments, or identifies with an actor.
2 We can distinguish three layers in a narrative text: the text, the story,
and the fabula. Each of these layers can be described. The difficulty
of distinguishing these is also a tool for understanding how they are
intertwined.
3 That with which the narrative text is concerned, the contents it
conveys to its readers, is a series of connected events caused or
experienced by actors presented in a specific manner.
 Introduction 9

Logically speaking, the reader first sees the text, not the fabula. The
fabula is really the result of the mental activity of reading; it is the inter-
pretation by the reader, influenced both by the initial encounter with
the text and by the manipulations of the story. The fabula is a memory
trace that remains after the reading is completed. And how writers pro-
ceed we simply cannot know. Writers work mostly alone; filmmakers
are part of a team. Narrative is a cultural phenomenon, one of the many
cultural processes by which we live. The conditions of possibility of
those processes are what constitute the interest of narrative analysis;
there lays its cultural relevance.
Together, these characteristics produce a definition: a narrative text is
a text in which all three characteristics are found. These characteristics
are not exclusive to narratives. The third feature also applies, for exam-
ple, to films and dramatic texts. Moreover, there are texts that display
all three characteristics but that nevertheless, on the basis of either tra-
dition or intuition, people do not regard as narrative texts. This is true
of many poems. The Waste Land by T.S. Eliot is one of the numerous
examples. A poem such as this may be termed a narrative poem, and
its narrative characteristics may be narratologically described. That this
does not often occur can be attributed to the fact that the poem displays
other, more salient characteristics, such as poetic ones. Hence, the char-
acteristics do not lead to an absolute delimitation of the corpus.
This in turn implies that a narrative theory facilitates description
only of the narrative aspects of a text and not all the characteristics,
even of a clearly narrative text. It is, therefore, as impossible as it is
undesirable to delimit a fixed corpus. This is an issue of relevance – the
answer to the always useful question “So what?” We can demarcate a
corpus of texts in which the narrative characteristics are so dominant
that their description may be considered relevant. Alternatively, we can
use the theory to describe segments of non-narrative texts as well as
the narrative aspects of any given text, such as, for example, the poem
by Eliot. Why bother with such analyses? A preliminary answer to the
question of relevance is that narrative – or rather, the narrativity that
makes artefacts narrative – grabs and holds the attention.
A number of descriptive concepts follow from the development of
the theory of such a narrative system. These concepts make possible
a description of texts to the extent that they are narrative. In practice, the
analyst will always make choices. Intuition often brings together a
striking aspect of an artefact and a relevant theoretical element. On the
basis of a careful reading of the text, as well as a careful attention to
10 Narratology

one’s reader’s response, one selects those elements of the theory that
one thinks particularly relevant to the text. Those will be, I presume, the
features that triggered one’s interest in the first place. This is why this
theory is compatible with reception-oriented approaches.
The textual description that results provides the basis for interpretation,
from which it cannot be firmly distinguished. Interpretation is involved
every step of the way. Precisely for that reason a systematic theory is help-
ful, not to eliminate or bracket interpretation but to make it arguable. An
interpretation is a proposal. If a proposal is to be accepted, it must be well
founded. If it is based on a precise description it can be discussed, even if,
in practice, the intuitive interpretation preceded the analysis. The theory
presented here is an instrument for making descriptions, and hence inter-
pretations, discussable. That, not objectivity or certainty, “being right” or
“proving wrong,” is the point.
Such discussions are possible and relevant because interpretation is
both subjective and susceptible to cultural constraints. Those constraints
define each reader as a cultural being, as a participant in a continuous
discussion about meaning. Endorsing that view entails an interest in
framings – those constraints that make the process of interpretation of
more general interest. This turns narrative analysis into an activity of cul-
tural analysis, for the subjectivity in analysis is a larger cultural issue. Sub-
jectivity, understood as the crossing, in culture, of individual and social
existence, also characterizes the concepts themselves. The provisional
definitions given above, and the more elaborate ones that follow, have in
common a special focus on agency. To talk about narrators, for example,
is to impute agency to a subject of narration, even if this subject is not to
be identified with the narrator. I will explain that focalizers, in the story,
are the agents of perception and interpretation. Actors, in the fabula, are
the subjects of action. This attention paid to agency – and, hence, to sub-
jectivity – is indeed the basic tenet of the theory presented in this book. It
insists on the complex manner in which narrative communicates.