Text

The Narratology of Comic Art
By placing comics in a lively dialogue with contemporary narrative

theory, The Narratology of Comic Art builds a systematic theory of
narrative comics, going beyond the typical focus on the Anglophone tra
dition. This involves not just the exploration of those properties in com
ics that can be meaningfully investigated with existing narrative theory,
but an interpretive study of the potential in narratological concepts and
analytical procedures that has hitherto been overlooked as well. This
research monograph is, then, not an application of narratology in the
medium and art of comics, but a revision of narratological concepts and
approaches through the study of narrative comics. Thus, while narrato
logy is brought to bear on comics, equally comics are brought to bear
on narratology.
Kai Mikkonen is Associate Professor of Comparative Literature at the

University of Helsinki, Finland.
Routledge Advances in Comics Studies
Edited by Randy Duncan, Henderson State University
Matthew J. Smith, Radford University
1 Reading Art Spiegelman

Philip Smith
2 The Modern Superhero in Film and Television

Popular Genre and American Culture
Jeffrey A. Brown
3 The Narratology of Comic Art

Kai Mikkonen
The Narratology
of Comic Art
Kai Mikkonen
NEW YORK AND LONDON

First published 2017
by Routledge
711 Third Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Routledge is an imprint of the Taylor & Francis Group,
an informa business
© 2017 Taylor & Francis
The right of Kai Mikkonen to be identified as author of this
work has been asserted by him in accordance with sections 77
and 78 of the Copyright, Designs and Patents Act 1988.
The Open Access version of this book, available at
www.taylorfrancis.com, has been made available under
a Creative Commons Attribution-Non Commercial-No
Derivatives 4.0 license.
Trademark notice: Product or corporate names may be
trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
CIP data has been applied for.
ISBN: 978-1-138-22155-0 (hbk)
ISBN: 978-1-315-41013-5 (ebk)
Typeset in Sabon
by codeMantra
Contents
List of Figures vii

Acknowledgements xi
Introduction: Comics, Narrative, and Medium 1
PART I
Time in Comics 31
1 Time in Comics 33
PART II
Graphic Showing and Style 71
2 Narration as Showing 73
3 Character as a Means of Narrative Continuity 90
4 Graphic Style, Subjectivity, and Narration 109
PART III
Narrative Transmission 127
5 Narrative Agency (in Jiro Taniguchi’s A Distant

Neighborhood) 129
6 Focalisation in Comics 150
7 Characterisation in Comics 174
vi Contents
PART IV
Speech and Thought in Narrative Comics 201
8 Presenting Minds in Comics 203

9 Dialogue in Comics: Medium-Specific Features
and Basic Narrative Functions 220
PART V
Narrative Form and Publication Format 243
10 Picture Story and Narrative Organisation

in Early Nineteenth-Century British Caricature
and Comic Strips 245
Afterword 277
Bibliography 279
Index 293
List of Figures
I.1 “After that it was my turn” © Matti Hagelberg. 1

1.1 April 20th, 2005. Building Stories (2012) © Chris Ware 35
1.2 Extrait de l’ouvrage. A Taste of Chlorine (2011).
Bastien Vivès © Casterman. Avec l’aimable
autorisation des auteurs et des Editions Casterman 39
1.3 Matti Hagelberg. Kekkonen (2004) © Matti Hagelberg 44
1.4 Fred. Philémon. La mémémoire (1977) ©
Philémon—tome 11, La mémémoire DARGAUD
by Fred. All rights reserved www.dargaud.com 52
1.5 Fred Philémon. La mémémoire (1977) © Philémon -
tome 11, La mémémoire DARGAUD by Fred. All
rights reserved www. dargaud.com 59
1.6 Guy Delisle. Pyongyang: A Journey in North Korea
© 2006 L’Association/Guy Delisle; Courtesy of
Drawn & Quarterly 62
2.1 John Leech. “Substance and Shadow” (1843) 78
2.2 H.M. Bateman. “The One-Note Man” (1921) 79
2.3 H.M. Bateman. “The One-Note Man” (1921) 80
3.1 Brian K. Vaughan and Fiona Staples. Saga © 2014
Brian K. Vaughan & Fiona Staples 93
3.2 Robert Kirkman, Charlie Adlard, and Cliff Rathburn.
The Walking Dead4 (2005) © Robert Kirkman 95
3.3 John Tenniel. “How Mr. Peter Piper Enjoyed a Day’s
‘Pig-Sticking’” (Near Burhampoor, Bengal)’ (1853) 98
3.4 John Tenniel. “Mr. Spoonbill’s Experiences in the Art
of Skating” (1855) 100
4.1 Tommi Musturi. Walking with Samuel (2009)
© Tommi Musturi 114
4.2 Winshluss Pinocchio (2008) © Winshluss 117
4.3 Winshluss. Pinocchio (2008) © Winshluss 118
4.4 © Blast - tome 4, Pourvu que les bouddhistes se
trompent DARGAUD by Larcenet. All rights reserved
www. dargaud.com 121
viii List of Figures
4.5 Tommi Musturi. Walking with Samuel (2009)
© Tommi Musturi 123
5.1 Graphic Novel Excerpt from PERSEPOLIS 2:
THE STORY OF A RETURN by Marjane Satrapi,
translation copyright © 2004 by A njali Singh. Used
by permission of Pantheon Books, an imprint of the
Knopf Doubleday Publishing Group, a division of
Penguin R andom House LLC. All rights reserved 140
5.2 Taniguchi, Jiro. A Distant Neighborhood. Vol. 1.
(2009) © by Jiro TANIGUCHI 141
6.1 Jarmo Mäkilä. Taxi van Goghin korvaan (2008)
© Jarmo Mäkilä 163
6.2 Emmanuel Guibert, Frédéric Lemercier, and Didier
Lefèvre. The Photographer © 2009 by First Second 163
6.3 Emmanuel Guibert, Frédéric Lemercier, Didier
Lefèvre, The Photographer, 2003. © Emmanuel
Guibert, Frédéric Lemercier, Didier Lefèvre 169
7.1 Gary Panter. Jimbo’s Inferno © 2006 by Gary Panter 181
7.2 Guido Martina & Angelo Bioletto. Mickey’s Inferno
(1949/2006) © Disney 182
7.3 Guido Martina and Angelo Bioletto. Mickey’s Inferno
(1949/2006) © Disney 187
7.4 Glyn Dillon. The Nao of Brown © 2012 Glyn Dillon 191
8.1 Rodolphe Töpffer. Monsieur Vieux Bois (1827) 205
8.2 Charlier & Giraud, Blueberry: Le Spectre aux balles
d’or (1972) © Bluberry – tome 12, Le spectre aux
Balles d’or DARGAUD by G iraud & Charlier. All
rights reserved www. dargaud.com 207
8.3 Madame Bovary par Daniel Bardet et Michel Janvier
© Éditions Adonis, 2008 211
9.1 Abel Lanzac and Christophe Blain Weapons of Mass
Diplomacy (2012/2014). Trans. Edward Gauvin
© 2014 SelfMadeHero 224
9.2 Aapo Rapi. Meti (2008) © Aapo Rapi 228
9.3 Garth Ennis and Steve Dillon. PREACHER. Book
One (1995) © Garth Ennis and Steve Dillon. All
characters, the distinctive likenesses thereof, and all
related elements are trademarks of Garth E nnis and
Steve Dillon 235
9.4 Jérôme Mulot & Florent Ruppert. Barrel of Monkeys
© 2008, Ruppert, Mulot & L’Association, Rebus
Books for the english translation 237
List of Figures ix
10.1 George Cruikshank and J. Pxxxy’s “Back & Front
View of the Ladies Fancy-Man, Paddy Carey O’Killus”
(1822). Courtesy of The Lewis Walpole Library, Yale
University 254
10.2 © The British Library Board, LOU.LON 1052.
Charles Jameson Grant. Fisherman and the fish.
Every Body’s Album (September 1834) 259
10.3 William Heath. “Life of a Soldier.” Northern Looking
Glass. No. XII. 12 December 1825 261
10.4 William Heath. “Life of an Actress”. Northern
Looking Glass X. 14 November 1825. 263
10.5 Robert Seymour. “Colonial Slavery.” The Looking
Glass No. 8. 1 August 1830 267
10.6 Robert Seymour. “St. Michael of London.”
The Looking Glass No. 8. 1 August 1830 270
Acknowledgements
The writing of this book would have been impossible without the
University of Helsinki Visiting Fellowship at Clare Hall College, University
of Cambridge, 2014–2015. I am deeply grateful to Maria N ikolajeva,
Professor of Education and Director of Children’s Literature Centre of
Homerton College, for helping to make that happen. My special thanks
are also due to John Calton, Matti Hagelberg, Essi Varis, Philip J enkins,
and Mark Shackleton, and the anonymous readers of Routledge Advances
in Comics Studies, for reading earlier drafts of this work. Thank you,
Matti, for your artwork and the enjoyable experience of teaching a
course together on comics storytelling at the University of Helsinki in
2013. Over the past decade, my work has profited greatly from the input
from colleagues and friends involved in the Nordic Network for Comics
Research (NNCORE) and the comics studies research committee of the
International Comparative Literature Association (ICLA). I am very
grateful for the sense of scholarly community and real friendship that
I have enjoyed with the members of these two groups.
Portions of earlier versions of some chapters have first appeared in the
journals Partial Answers 6.2 (2008), American Studies/Amerikastudien
56.4 (2011), International Journal of Comic Art 13.2 (2011), and
Scandinavian Journal of Comic Art 1 (2012), and in the anthologies
The Rise and the Reason of Comics and Graphic Literature (Eds. Joyce
Goggin and Dan Hassler-Forest, 2010), From Comic Strips to Graphic
Novels (Eds. Daniel Stein and Jan-Noël Thon, 2013, 2nd edition 2015),
and Dialogue across Media (Eds. Jarmila Mildorf and Bronwen Thomas,
2017). I would like to express my gratitude to the publishers and the editors
for their permission to reproduce this material here. In the same vein,
I would like to extend great thanks to all the copyright holders who have
generously given the permission to reproduce the images included here.
Introduction
Comics, Narrative, and Medium
I first learned to read by reading comics with my father when I was

around 5 or 6 years old. For all I know, my father never read comics
by himself, except perhaps when browsing comic strips in his morn
ing newspaper, but he wanted to teach me how to read. That’s why we
ended up reading my comic books. He would hyphenate all the words,
syllable by syllable, in the speech balloons with his marker pen on some
randomly chosen page and then read the words to me one syllable at a
time (Figure I.1).
Figure I.1 “After that it was my turn” © Matti Hagelberg.
Later I have often wondered what it was in these children’s comics,

possibly Donald Duck or Bugs Bunny, or in my father’s method, that
helped me to learn. One reason, definitely, must be the images that
I could follow on my own and that I had already encountered. It was
exciting to make the connection between the words and the drawings,
the characters who were speaking and the situation at hand. The words
added new significance to the images and shifted my previous under
standing, but I am convinced that the narrative form of these comic
books also had something to do with it. Learning the words enhanced
the story, making it fuller, perhaps funnier, more thrilling, and more nu
anced, while in pleasant symbiosis, the continuing story told in images
2 Introduction
made it easier to figure out what the words meant and how they related
to everything that I could see in the images.
***
This book has two main objectives. First, by placing comics into lively
dialogue with narrative theory, The Narratology of Comic Art builds a
systematic theory of narrative comics. This involves not just the identifica
tion of those properties in comics that can be meaningfully investigated by
means of narrative theory, but a study of the potential in the narratolog
ical approach that has hitherto remained overlooked in this field as well.
Comics studies is experiencing an exciting period of growth and diversi
fication, and as interest in exploring the connections between comics and
narrative theory increases, it is all the more important to carefully think
through what is specific to the medium. A comprehensive narratological
engagement with comics will allow us to do just that, and contribute to our
understanding of comics’ narrative devices, conventions, and strategies.
Second, the development of the narratology of comic art entails the
identification of the kinds of problems that we encounter in employing
narrative theory in comics studies. Many of these problems stem from
the historical fact that narratology—the formal, systematic study of nar
rative representation—has been more extensively developed in literary
research than in other fields. This book, then, is not a mere application
of existing narratology to the medium, but it also offers a revision of
narratological concepts and approaches through the study of narrative
comics. Thus, while narrative theory is brought to bear on comics, com
ics are equally brought to bear on narrative theory.
In this investigation, the relation between narrative, or narrativity
(i.e. the capacity to inspire a narrative response), and medium is of
crucial interest.1 One key finding of this study is that narratological
insights into the organisation, presentation, and mediation of stories
cannot be transferred from one medium to another without due modi
fication. Thus, it is hoped that this investigation can contribute to nar
ratology in general, for instance, with regard to the emerging field of
transmedial studies that looks at narratives in different forms of expres
sion, communication, and art. 2 Furthermore, the challenges that the
medium of comics poses for narrative theory can potentially shed light
on related problems elsewhere, specifically in other forms of visual and
multimodal narratives. 3
Narratology and the Narrative Medium

From its beginnings in the late 1960s, narratology has developed for
mal accounts of narrative and its functioning. In the early stages of the
discipline, inspired by structuralist linguistics, narratology sought to
Introduction 3
identify the major constitutive elements of narrative. Most importantly,
the French literary scholar Gérard Genette defined in his study N arrative
Discourse (“Discours du récit: Essai de methode”, 1972), based on the
analysis of the narrative forms in Marcel Proust’s Remembrance of
Things Past, the main research issues of narrative analysis, such as nar
rative time, voice, and mood. In this study, Genette invented several
much-used typologies of narrative technique, in particular concerning
the narrator, narrative level, focalisation (perspective), and temporal or
der that will be discussed here in relation to comics.
Today, narratology has been established as a discipline that has both a
theoretical and an application-oriented approach to narrative, while still
focussing on narrative qua narrative: “What is typical to n arratives?”, or
“What is a narrative?” (Prince 2003b, 3). The various contemporary nar
ratologies, often grouped under the superordinate term of post-classical
narratology,4 have broadened the field considerably, both with regard to
the research questions and the corpus investigated. These changes have
pushed the field in new directions and revitalised the theoretisation, re
search, and analysis of narrative representation, communication, and
expression.
The broader approach to narratological problems in recent decades
has included, for instance, the question of the reading process, the his
torical context of narratives and their interpretation, and the history
of narrative forms. In part, we can see this development as a reaction
to the popularity of transmedial storytelling and story systems, involv
ing large-scale production in various media, such as The Lord of the
Rings, the Harry Potter series, Star Wars, and Doctor Who that create
a storyworld through multiple documents in various media platforms.5
Therefore, narratology can investigate, for instance, the transferability
of narratives between various media. This kind of inquiry may ask how
the medium affects narrative form and meaning, or how a given medium
may be better equipped to carry out some kinds of narrative functions
rather than others (Ryan 2004, 34–35).
These changes have been significant for the study of comics. For one
thing, the expansion of the narratological field does not just involve ‘the
corpus argument’—that is, the incorporation of new kinds of narratives
into the discipline—but also ‘the medium argument’, i.e. the question of the
applicability of narratology across the media, and the interest in exploring
the relation between narrative and the medium. A seminal part of this in
vestigation has also been the applicability of key narratological concepts,
such as focalisation, narrator, or voice, across media. In fact, through
out the history of the field, narratologists have frequently tested their
concepts, theorems, and analytical approaches with other-than-literary
narratives, including comics. The impulse in narratology to conceptual
ise whatever is general to narrative, irrespective of its various manifesta
tions, has driven research in this field towards intermedial questions and
4 Introduction
comparisons. Many early French narratologists, such as Roland Barthes,
Tzvetan Todorov, and Claude Bremond, held that the same story may
be realised in different forms of art and communication. For instance,
Todorov, who coined the term narratologie in his 1969 study Grammaire
du Décaméron, calling ‘narratology’ the science of narration (la science
du récit), was interested in developing a theory of narration that could
be applied to all domains of narrative, including literary texts, popular
tales, myths, films, and dreams (1969, 10). Later, in the American con
text, Seymour Chatman’s central works in this field, Story and Discourse
(1978) and Coming to Terms (1990), featured a significant amount of
reflection on film narratives and their medium-specific ways to tell a story.
A clear shift of emphasis away from literature to other forms of
narratives—or narratives across media—is reflected in the Dutch cultural
theorist Mieke Bal’s works from the late 1970s to the 1990s. Bal’s 1977
introduction to narratology, Narratologie (translated as N arratology:
Introduction to the Theory of Narrative, 1985), which helped to estab
lish the name of the discipline, analyses in its second and fully revised
edition (1997) several visual narratives, from Ken Aptekar’s paintings to
Steven Spielberg’s filmic adaptation of Alice Walker’s The Color Purple
and the Indian seventh-century bas-relief Arjuna’s Penance. Bal’s narra
tological attention to visual storytelling, such as the question of perspec
tive, is an important precedent for the present-day interest in narratives
across the arts and the visual media.6
During the 1980s, narratology established important inroads in film
studies. In the French-language context, notable among the pioneering
works in film narratology were Francois Jost and André Gaudreault,
who, inspired by Genette’s concepts and theorems, sought to develop
the theory of film narratives. Their work is directly relevant to the
narratology of comics, in particular in relation to questions of perspec
tive and narrative mediacy. In English-language film theory, narratology
came into its own in the influential works of David Bordwell, Kristin
Thompson, and Edward Branigan.7 Today, given its transmedial rele
vance, for instance, with regard to questions of narrative comprehen
sion, style, agency, temporality, characterisation, and perspective, film
narratology is enjoying an increasing influence on the narratological
study of visual storytelling in general.
The Beginnings of Comics Narratology

Narratology has often been applied in comics research, and it has regu
larly informed comics theory, but no comprehensive narratological the
ory of comics has been proposed to date. Typically, comics theorists and
scholars have made use of narratological notions whenever they have
seen it fit for their research question. One rare exception in this regard
is Martin Schüwer’s Wie Comics erzählen (“How Comics Tell a Story”,
Introduction 5
2008), where Schüwer explicitly discusses the relation between narrative
theory and comics, and applies Genette’s categories of focalisation and
narrators to comics. Schüwer’s discussion of movement, space, and time
is also widely informed by narratological insights into temporality and
narrative communication. Furthermore, Schüwer calls for the modifi
cation of the existing narratological concepts to better suit the study of
comics (2008, 17–26). The Narratology of Comic Art takes up precisely
this challenge.
Narratology has also been one of the recurring points of theoretical
reference in the French–Belgian ‘school’ of comics theory since the 1970s,
reflected in the works of scholars such as Pierre Fresnault-Deruelle,
Philippe Marion, Thierry Groensteen, Harry Morgan, Jan Baetens, and
Pascal Lefèvre. In his doctoral dissertation Traces en cases (1993), one
of the most cited works in this tradition, Marion adapts narratology
creatively, specifically in relation to the notion of narrative enunciation.
We will return to Marion’s theory in chapters on time, graphic style, and
narrative agency. Thierry Groensteen, in his influential studies Système
de la bande dessinée (1999) and Bande dessinée et narration (2011),
translated into English as The System of Comics (2007) and Comics and
Narration (2013), applies certain basic principles of narratology. More
precisely, Groensteen makes use of Todorov’s principles of narration,
succession, and transformation, filtered through André Gaudreault’s
film narratology (1999, 122, 135). In the introduction to The System of
Comics, however, Groensteen also observes that general semiological
theory, with which he associates narratology, has often poorly justified
its linguistic bias, i.e. emphasised the verbal component at the expense
of pictures and visual meaning. He notes, moreover, that the dogmas
of literary narratology are all-too-often applied mechanically to other
forms of narrative (1999, 12).
Although Groensteen’s approach is broadly informed by n arratological
theoretisation, it does not constitute a narratology of comics. Groensteen
characterises his theory with the term ‘neo-semiotic’, by which he refers
not only to the semiotic foundation of his approach but also—with the
neo-prefix—to his focus on the language and intelligibility of the me
dium, or the poiesis of comics, understood in the sense of the condi
tions of their meaning-making. In a later addition to his neo-semiotic
system, Comics and Narration, Groensteen makes the relationship be
tween his approach and narratology somewhat more explicit, especially
in his discussion of the narrator concept and subjectivity. Nevertheless,
Groensteen reiterates his warning here (2011, 86–87) that the extra
polation of narratology, such as the theory of the narrator, to the field of
comics studies, will not succeed without due modification and revision
of these concepts.8
The constant, but also often problematised, references to narrato
logy in the German- and French-language comics theory have been an
6 Introduction
important dimension in the multi-faceted historical relationship between
comics studies and narratology so far. Perhaps even more importantly,
however, narratology has had an indirect effect on comics theory, also
in the English-language research, through general notions of narrative
form and technique, for instance concerning voice, perspective, and nar
rative agency. One particular dimension of this relation is to be found in
the introductions to narratology, published before our ‘transmedial’ era,
which have appropriated comics, and especially the short format of the
comic strip, to illustrate the theory. These readings are often revelatory
of the tendency to make general claims about narratives without paying
attention to the relationship between narrative and medium.
Three examples will serve to make the point. First, in Story and
Discourse, one of the most influential introductions to narratology in
North America, Seymour Chatman analyses at length Frank O ’Neal’s
newspaper strip “Short Ribes” in order to illustrate his theory of the
narrative situation (1978, 37–41) and, in particular, to explicate the ba
sic narratological distinction between story and discourse in the tempo
ral organisation of a story. For Chatman, the comic strip demonstrates
what he calls the process of ‘reading out’ narratives, that is, how readers
may read “the relevant narratives out of or through one sort of non
verbal manifestation” (1978, 41). Most importantly, this means the
readers’ abstraction of the ‘story’, that is, what is told, out of ‘how’ the
story is transmitted (i.e. ‘discourse’). Second, in their Telling Stories:
A Theoretical Analysis of Narrative Fiction (1988), Steven Cohan and
Linda M. Shires offer an extended discussion of Cathy Guisewite’s
comic strip from the long-lived “Cathy” series to illustrate the difference
between different categories of plot structure, in particular the so-called
kernel and satellite events. This is the distinction, made initially in Story
and Discourse,9 between events that raise possibilities of succeeding or
alternative events (kernels) and those that “amplify or fill in the outline
of a sequence by maintaining, retarding, or prolonging the kernel events
they accompany or surround” (satellites) (1988, 54). Third, in his 1992
article “Points of Origin: On Focalization in Narrative”, Patrick O’Neill
develops a theory of focalisation, and especially the notions of embed
ded focalisation and covert global external focalisation (the implicit
perspective of the so-called implied author), through a reading of Bill
Watterson’s Calvin and Hobbes strip (1985–1995). O’Neill argues that
the implied author’s covert perspective can be detected in the graphic
style of Watterson’s strip, that is, in the reality of the characters drawn
as cartoon characters.
These three approaches are alike in their use of comic strips as illus
trations for the basic tenets of narratology and, at the same time, for
their lack of interest in the specificities of the medium. What is prob
lematic, especially, is that in all of these examples, comic strips are per
ceived, independent of their medium, to represent a narrative in its most
Introduction 7
rudimentary and, at once, most illustrative form. The fact that a comic
strip includes pictures, or is structured around panels in a sequence, sup
posedly has no significance for the way the story is told.
What makes comics so illustrative of the basic tenets of narrative the
ory? Two immediate explanations come to mind: the visual and succinct
form of these strips. In other words, the strip format that tells a simple
story in a compact space of a few panels is attractive to use in a textbook,
an article, and a course on narratology. Moreover, differences between
literature and comics can be neglected since it is presumed that a comic
strip is, after all, only a simple narrative for fast and easy consumption.
In fact, Seymour Chatman refers to the comic strip as a deliciously edible
new form of hermeneutic interpretation: “a glorious new medium” for
hermeneutics “to munch on, along with Sunday pancakes” (1978, 41,
n16). The discontinuous spatial arrangement of the strip allows Chatman
to identify both that which is left out of the images (the ‘what’ of the story
that the reader needs to infer) and the key events of the story. However,
the discontinuous spatial form itself, or the visual framing of the perspec
tive, and its partial subjectification as the main character’s perspective
are not incorporated into the analysis. This is all the more curious since
part of Chatman’s reading of the strip relies on his notice of a difference
between the media: In the comic strip, the distance between the imagined
perspective of the panel frame and the character—a king who decides to
pawn his crown in order to play at a casino—keeps changing all the while
the given angle shifts between panels. In the comic strip, we can look at a
character looking at something (that we do not see), get gradually closer
to his field of vision (as the perspective alters in each panel), and also liter
ally share part of what the protagonist sees. Perhaps since the differences
between literary and visual narratives are too obvious and palpable in
this regard, the question of perspective is left aside.
The classical narratologists rarely posed the question of the relation
ship between narrative and medium even if they were motivated to con
sider what unites narratives across the media and the arts. Thus, comics
remained a transparent medium, too obvious, surely, to deserve any
narratological discussion of its own. It is, therefore, high time to think
more systematically about the specificities of narrative comics, as well
as about the similarities between comics and other narrative media, in
narratological terms.
The Narratology of Comics: What Should Be Included?

As much recent research suggests, comics are increasingly recognised as
a medium, or form of art and literature, that is worthy of narratological
attention in its own right. One indication of this change is that in the
recent decade, narratology has become a frequent component in text
book introductions to the study of comics.10 At the same time, there has
8 Introduction
been a growing consent that narratology needs to be taken into new di
rections through this medium. Several theorists of narrative and comics
have recently underscored the importance of such a task. For instance,
David Herman notes in his Basic Elements of Narrative (2009a, 74)
that “a fully developed narratology of graphic narratives (…) remains a
goal for the future”. Likewise, in her Contemporary Comics Storytelling
(2013), Karin Kukkonen calls for a “comprehensive and coherent narra
tology of comics” (2013a, 180), and she further explains that such a proj
ect would “work toward an understanding of the design and the effects
of comics narratives, and thereby contribute to the overall development
of narratology” (2013a, 181). Also, Jan Baetens and Hugo Frey state
these objectives in The Graphic Novel. An Introduction (2015), where
they specify that comics studies, and the study of the graphic novel in
particular, can redefine narratology in two significant ways: by showing
us that storytelling cannot be separated from its inherent and medium-
specific materiality, and by reintroducing certain aspects of story analy
sis, such as world-making, space, and characterisation, which have been
neglected by the dominant formalist methodologies of narrative analysis.
But what should be included in a comprehensive narratology of com
ics? What should be its main foci? To return to Kukkonen, for her, a full
narratology of comics would address in detail three problematic issues
that have been much debated in literary narratology: the story/discourse
distinction, the question of the narrator, and focalisation (2013a, 181,
185). To these problems, she then adds the questions of “how comics
establish storyworlds and how they characterise and communicate fic
tional minds” (2013a, 185). In her Reading Bande Dessinée: Critical
Approaches to French-language Comic Strip, Ann Miller dedicates a
chapter (2007, 103–124) to outline those areas of narrative theory that
she sees as being particularly relevant for the understanding of narration
in comics. Comparing comics storytelling to narrative techniques in film
and novels, Miller suggests that relevant narratological questions would
involve, in particular, the issues of narrative time; narrative perspec
tive, or focalisation and ‘ocularisation’ (a character’s visual perspective);
and narrating instance, concerning issues such as voice and the narrator.
Thus, drawing on Genette’s distinction between narrative time, mood,11
and voice and combining this framework with relevant insights from
film narratology, Miller sketches another potential framework for a nar
ratology of comics.
If we were to follow Miller’s and Kukkonen’s proposals, a compre
hensive narratology of comic art should cover the questions of the story/
discourse distinction, the narrator, focalisation, storyworld, and the rep
resentation of the characters’ speech and thought (or their ‘minds’).12 Be
yond deciding upon this set of categories, however, we should consider
the relevance of these questions in relation to the capacities and con
straints of the medium. For instance, given that time is often represented
Introduction 9
through the spatial organisation in this medium, the premises of the story/
discourse distinction should be subjected to a thorough medium-specific
reevaluation. Given also the many ways in which comics tell a story
through images, the narratology of comic art needs to cover issues such
as narration by visual showing and graphic style, aspects that are simply
not relevant in literary narratology. Furthermore, the representation of
perspective in comics and the question over the category of the narrator
in many ways have more in common with visual forms of narration,
such as film, than literature. This means that the narratology of com
ics is cognizant of developments in film narratology, visual studies, and
multimodal studies in this regard. Moreover, as Baetens and Frey point
out, the vast topic of characterisation in comics has not attracted the
narratological attention that it deserves.
A more comprehensive narratology of comics should not simply bor
row its basic concepts, theorems, and approaches from literary or film
studies, but it should also critically reflect on those categories, adjust and
calibrate them, and, where necessary, invent new ones. Beyond identi
fying the basic narrative constituents, conventions, and means of the
medium, we need to reflect back on narratology to evaluate what this
approach allows us to bring into a sharper focus in narrative comics
and how the approach should be modified to better achieve that goal.
One risk that I wish to avoid is to propose a narratology of comics that
would treat the medium as just another form of literature or film, with
additional remarks about the function of images in a sequence, or the
interplay between word and image.
What Is Narratological Analysis?

The point of narratology is that it can make relevant claims about the
narrative qualities of texts. Narratological concepts are not timeless
universals but developed on the basis of particular corpora of narra
tives that are created in a certain time and place, in a particular form
and medium, and for a particular audience. The historicity of narrative
forms is also one reason why the division between narratological ana
lysis and interpretation cannot always be very strict. Surely, for many
scholars, including myself, narratology is, first of all, a heuristic tool that
helps to arrive at a clearer understanding of narrative literature, art, and
communication. If we distinguish between narratology and the proper
interpretation of narrative texts and works, where the aim of analysis is
to relate the narrative text to particular contexts of meaning, narrato
logical analysis can also be seen as a distinct phase in the research pro
cess. In this process, the value of narratology is measured in the ways in
which it enriches the interpretation and understanding of particular nar
rative works.13 Tom Kindt and Hans-Harald Müller, who seek to make
a clear distinction between the narratological process of description and
10 Introduction
interpretation, have argued, more precisely, that narratology serves “as
a tool for preparing, initiating, or backing up interpretations; it is un
derstood as a heuristics but not as a theory of interpreting texts” (Kindt
2009, 37).
That said, however, it is important to keep in mind that narratological
analysis and description is typically part of an interpretative process,
not an objective in its own right. The narratological framework brings
forth certain features of texts, works, and documents instead of some
others, that is, it focusses on their narrative qualities and their qualities
as narrative works. Another problem that we may confront in defining
narratology too strictly as a formal description of narrative is that it
evokes the thorny issue of form versus content. It goes without saying
that narrative forms and devices in comics can be employed to convey
meanings, not just to create an artistic form. For instance, the page lay
out can be used to give a sense of a character’s perspective, the frames
and the type of panelling can support a theme, and the dynamic between
the narrating “I” and the “I” shown in the images can have various
consequences for identity and self-image in autobiographical comics.
Samuel R. Delany has claimed cogently in this respect that the division
between form and content may be a necessary critical fiction, but that
it is only a provisional fiction since “at a certain point in the discus
sion, form begins to function as content—and content often functions
as a sign for the implied form with which that content is conventionally
dealt” (1999, 259). The challenges we face in trying to relate narrative
forms to particular meanings, or in trying to understand the relationship
between style and meaning, are often highly complex. This complexity,
and the way in which form and content work together, is also a major
incentive for research in the first place. Narrative techniques can be used
for different purposes and to different ends, and they can be successfully
connected to a seemingly limitless number of themes, contexts of mean
ing, and interpretations.14
It is important to keep in mind that particular narrative devices, con
ventions, and strategies in comics always have a history, having deve
loped in particular circumstances of artistic production and reception.
For instance, the history of modern comics has been affected by a num
ber of significant changes in artistic production and publication format
with various consequences for the way in which stories can be told in
this medium. Among such major changes can be counted the inclusion
of cartoons and comics in the European nineteenth-century satirical
journals, the emergence of American newspaper comic strips and comic
magazines in the late nineteenth century, the breakthrough of comics
weekly magazines and albums in Europe and comic books in America in
the late 1930s,15 the experiments of underground comics in the 1960s,
and the development of the graphic novel and independent comics since
the 1970s. Many of these transformations have changed conventions
Introduction 11
of storytelling in the medium, such as how characters’ thoughts and
emotions are presented or how to create a sense of dynamic interaction
between action and dialogue, pushing the very limits of the medium.
The historical nature of these changes is not my focus here, but I do not
pretend that I can ignore it.
We can conclude, then, that narratological analysis is not the same
thing as interpretation, but that it requires interpretive skills and its aim
is often to aid interpretation. As a heuristic tool in comics scholarship,
narratology can be conceptualised as a kind of preliminary stage of
interpretation that directs our attention to the narrative features in a
given work and helps to analyse and clarify the significance of those
features. In Film Narratology (2009), Peter Verstraten is keen to point
out that narratology is not a magical solution to all interpretative prob
lems in film analysis. After all, narratology restricts itself to the narra
tive aspects in films, and it can offer a sound foundation for research
only as long as we understand that a more complete narrative analysis
is always ‘narratology plus X’ (2009, 11). Verstraten also emphasises
that no film—and we might add, no comic—is “unproblematically nar
rative in its entirety” (2009, 24). Even the most prototypical narrative
comic, say an action comic or a Tintin adventure, has other dimensions.
Comics use symbols and metaphors; they are descriptive, painting-like,
lyrical, pedagogical, or abstract rather than narrative; they can define
or explain things rather than narrate events; they can put forward
claims about a fictional world or our world; they may function as a
commentary, an argument, or tell a joke; they can inform or teach us
about something; and so on. All this can obviously happen within a
larger narrative compositional frame, but it is not always the case. Not
everything in comics is narration, and not all comics are narratives.
Instructional comics such as guidebooks and instruction leaflets, lyrical
comics, or abstract comics do not always tell a story, or their narra
tivity may be weak. Furthermore, the reading of comics also regularly
involves other aspects than the processing of the story and its telling,
such as paying attention to the visual features of the image or style and
appreciating the narrative drawings or the spatial composition. One of
the challenges I will tackle in this book in this respect is the relation
between graphic style and narration. Style often serves a narrative func
tion in comics, but its effects and implications are never exhausted by
this function.
Narratology may be the best available theory for describing and ex
plaining comics as narratives, but it should not pretend to cover all possi
ble qualities of comics, not even in narrative comics. Therefore, a crucial
aspect raised in this book is the judicious application of narratological
concepts. This will necessitate a self-critical reflection on the limits and
possibilities of these concepts, theorems, and analytical approaches in
close relation to a wide range of works.
12 Introduction
The Key Terms
It is a generally accepted truth in research conducted in the humanities
that a clear understanding of the key terms of the research determines
the shape of the study and gives it perspective. Thus, some definitions
are needed.
Comics
There is not much use for a strictly formal definition of ‘comics’ in the
narratology of comic art. A formal definition would be, for instance,
to state that comics are “the phenomenon of juxtaposing images in a
sequence” (Duncan and Smith 2009, 3) or “juxtaposed pictorial and
other images in a deliberate sequence, intended to convey information
and/or produce an aesthetic response in the viewer” (McCloud 1993, 9).
We could also state, though less strictly, that “the necessary, if not suf
ficient, condition required to speak of comics is that the images will be
multiple and correlated in some fashion” (Groensteen 2007, 19). The
problem with these definitions is not so much that they are too broad
or too narrow, but that they qualify comics by reference to some single
component that is perceived as the necessary element for something to
be called comics. In comics scholarship, the futility of the definitional
project has come sharply into view.16
The core notion in most contemporary definitions is that comics are
juxtaposed pictorial images, a sequence of separate pictures or images,
or interdependent images in a series. Groensteen’s emphasis on what he
calls the foundational order of “iconic solidarity” (2007, 128) highlights
the idea of the interdependence of images in a series as the most relevant
feature of the medium17 and distinguishes comics from single images or
“unique enclosed images within a profusion of patterns and anecdotes”
(2007, 128). Such a distinction is indeed quite reasonable—the idea of
interdependent images seems crucial to the recognition of something as
comics in our contemporary understanding of this word. And yet, it is
also clearly problematic. It is problematic, in particular, in the sense that
the notion of a “unique enclosed image” is not always perfectly distin
guishable from a series of interdependent images, specifically if a single
image depicts several events or an unfolding situation. When does a sin
gle image become a series, or when can a group of images be perceived
as one image? As art history demonstrates to us, a sequence may be em
bedded in a single image; a single-panel image or painting can comprise
a series of images. And, we can find various kinds of image series outside
comics, for instance, church paintings, frescoes, or window art that may
also conform to this principle. In addition, a formal separation between
comics and other related visual media or forms of art, such as picture
books, cartoons (single-panel drawing), or animation, is irrelevant here.
Introduction 13
There will always be borderline cases between these forms of expression,
such as hybrid formations of picture books and comics (for example,
Maurice Sendak’s In the Night Kitchen), books illustrated with comics
and cartoons (Jeff Kinney’s Diary of a Wimpy Kid), or intermedia in
stallations (so-called ‘gallery comics’), which render the question of defi
nition difficult or impossible and point out the historically contingent
and relative nature of such categorisations.
In the various formal definitions of comics, a lot of ink has also been
spilled over the question of the interplay between image and word, and
whether it is the defining feature of the medium. The fundamentally
hybrid nature of the medium in this respect is also reflected in the fre
quent attempt in comics theory to reveal a hidden bias in scholarship in
favour of either the word (literature) or the image (art, visual culture). By
displaying these biases, comics theory has then sought to accommodate
the opposite argument, shifting the emphasis again to either literature or
visual art and leading to some kind of self-perpetuating dialectic.
For instance, Bart Beaty has identified, in his discussion of the
relationship between comics and the art world in the North American
context, Comics versus Art (2012), what he calls a ‘literary turn’, or
literary–narrativist prejudice, in the academic study of comics in re
cent decades. Beaty locates the beginnings of this prejudice in Colton
Waugh’s The Comics (1947), where Waugh highlighted the importance
of certain issues of narrative content, such as continuing characters and
narrative form, including sequential images and speech balloons, for all
comics (2012, 27). More recently, as Beaty claims, a similar bias, which
sees comics as popular narrative literature or mass culture but not as a
form of visual art, has only strengthened. This is mirrored, he goes on to
say, in the coalescence between narrative and literary medium, and the
easy slippage between arguments about comics as a narrative form to
the discourse of their literariness in academic comics studies (2012, 44).
Whenever contemporary comics scholarship has equated narrative form
with literariness (the quality of being literary), this is usually justified as
an attempt to raise the low cultural status of comics in academia and, in
particular, in the field of literary studies. The opposite bias, however, of
treating comics as popular visual culture, instead of literature, has also
been quite common. For instance, one persistent trend in comics theory,
though less common today, has been to see comics as a ‘frozen’ and ‘im
mobile’ form of cinema, or a kind of storyboard for a film that conse
quently does not require a poetics of its own.18 More recently, comics have
also been subjected to broader theoretical frameworks of visual language
and meaning-making—semiotic, linguistic, or cognitive approaches, for
instance, or multimodal studies—in ways that may marginalise the ques
tion of comics as a self-respecting medium, literature, or art form.
The literary or narrativist bias in contemporary comics scholarship is
a complex issue. First, as most narratologists are happy to admit, there
14 Introduction
is no a priori connection between literary value and narrative form. In
literary studies, the concept of literariness—that is, what makes a given
work a literary work—is not associated with the question of narrativity
but, in the main, with the organisation and style of language that devi
ates from standard use, and may thus highlight the stylistic choices or
the act of expression itself.19 Literary value can also be defined through
the concept of imagination for instance, but a narrative form is hardly a
sufficient requirement for some work to qualify as literature. Second, al
though much comics theory today tends to see that comics are narratives,
it also recognises that there exist many predominantly non-narrative
forms of comics, such as abstract, educational, and lyrical comics.
As suggested earlier, the dispute about the biases of comics theory
and research can be taken as another reminder of the truly multimodal
nature of the medium. For instance, in her Comics and Language (2013),
Hannah Miodrag presents almost the opposite argument to Beaty in
challenging what she sees as the preoccupation with the visual content
in comics studies and criticism. Thus, she undermines the perception—
one that she sees as central in much scholarship in the field—that visual
content comes first and needs “to control the narrative and dominate
the text” (2013, 11). She further concludes that there have been a “gen
eral critical reluctance to focus on verbal content” (2013, 18) in comics
and a widespread “logophobia” in comics studies (i.e. a fear of words
subjugating comics’ visual form)20 in order to insist on the visual value
of comics. Miodrag then shows in her insightful analyses how the ver
bal content in comics, from George Herrimann to Lynda Barry, can be
conceived in terms of literary language and aesthetic effect, while she
also seeks to problematise, by focussing on linguistic features in comics,
what it means for a comic to be read as literature.
We could claim that by conceiving comics as literature, Hannah
Miodrag distances the medium from visual art and culture, while Bart
Beaty, by making a case for comics as an important form of visual art
in its own right, distances comics from the art of storytelling, which he
associates closely with literariness, literary value, and literary studies.
My focus on narrative comics is an important qualifier here. Comics can
be perceived as a form of literature and visual art, but to treat them as
narrative also allows us to avoid the word–image dichotomy. I am con
vinced that there is no particularly good reason to juxtapose the cate
gories of literature, narrative, and visual art in this respect. Comics tell
stories by verbal means and by showing images, through their visual and
spatial form, through their many combined visual–verbal signs and con
ventions, and through the interaction of these elements. The communi
cation of a story is, surely, not the only thing that pictures do in comics,
or what comics do, but it can also be a focus of its own.
What I wish to emphasise in this regard is mainly that it is quite pos
sible to avoid the circularity of the argument that has characterised
Introduction 15
some recent comics theory: When the pictorial or the verbal character of
comics (or the idea of images in succession) has been deemed to be too
prominent, the corrective move has shifted the theoretical perspective in
favour of the visual component and graphic art, or vice versa. However,
there is just so much variety in comics—that is, in works that are pro
duced, recognised as and called ‘comics’—in their blendings of images
and words, or their emphasis on one or the other, that the interplay
between words and images does not provide us with any self-evident
starting point for a comprehensive theory of narrative comics. The rich
tradition of wordless comics will also always be hard to accommodate
within such definitions.
This book will not discuss the issue of word and image interaction as a
separate aspect of narrative comics, but it will integrate their interaction
in all of its chapters. The focus on comics as narratives allows us to con
ceive comics in relation to narrative forms, conventions, and strategies,
but this does not necessarily make comics more recognisable as litera
ture or visual art. I do think favourably of both of these endeavours since
they can open up interesting research perspectives on intermedial and
interartistic relations and make us aware of any medium and art-specific
biases in our own aesthetic judgements. Furthermore, it can be import
ant to claim literary and artistic values for comics in contexts where seri
ous scholarly interest in this field still needs to be justified. However, it is
not the objective of a comprehensive narratology of comics to elevate the
value of comics as literature or art, but to study them as narratives and,
subsequently, illustrate their value as narratives. It is also to be hoped
that the literary and artistic merits of comics need no underscoring.
Therefore, instead of a formal, or even a reasonable, definition of com
ics, 21 comics are perceived here in terms of their social institution, that
is, the institution of making and reading comics. In other words, the
term ‘comics’ refers here to a medium or, if you like, a form of art and
literature that consists of works that are commonly recognised, called,
and intended to be read as comics.22 The institutional definition enables
us to avoid the formalist trap of focussing on a particular structure or de
vice as the distinguishing factor. This approach is not without caveats as
there may be some disagreement about what counts as comics today, but
the institutional definition suggests that we can appreciate these diver
gences instead of trying to resolve them according to formal definition.
Narrative and Graphic Narrative

The question of what constitutes a narrative seems an even thornier is
sue than what constitutes comics. This is because the phenomena that
are commonly recognised as a ‘narrative’ are incredibly versatile. In the
everyday usage of the term, a narrative can not only be an account of
events or an act of telling a story, but it may also be equated with fiction,
16 Introduction
ideology, and myth. Sometimes, in comics theory and scholarship, a
‘narrative’ is used synonymously to the verbal component or the literary
aspect of the work, or in the meaning of the story in its verbalised form.
These usages are misleading in that narration in comics usually happens
both through words and images and, for much of the time, in a concep
tual space between them.
One useful distinction for the many dimensions of the term ‘narrative’
is that it can refer to a product, such as a text, document, and perfor
mance; a process, such as narrating, reading, watching, and listening to
a narrative; and a mental model in the sense of perceiving something as
a narrative. Most contemporary narratological definitions of this term
combine these three dimensions in one way or another. A narrative text,
a film, or a comic book, for instance, typically implies some conception
of who does the telling (in reality and/or fictionally), how the telling is
done, and how the audience is supposed to understand the telling (and
not just the story that is told). Moreover, the perception of events and
occurrences as a narrative is something that can be both an element of
the narrative text or document—a character’s, narrator’s, or an author’s
perception—and what happens in the reading of some text or documents
as narrative.
Narratology tends to perceive certain features as constitutive of narra
tives. Among these basic properties are included, in particular, the tem
poral sequencing and the causal connection between the narrated events
(or different phases of one event). 23 However, a chronology of events,
a chronicle, does not in itself constitute a narrative, or it makes only a
poor story; a causal connection between the events is usually required,
related, for instance, through a narrator or an experiencing character, a
sense of a world that undergoes changes, a narrative situation (narrator
and his or her audience) and a communicative structure (being situated
in a specific occasion of telling), or the experiential frame (the author’s,
the narrator’s, or the character’s experience and the emotional evalu
ation of that experience). 24
Here, I am not seeking a formal definition of narrative since I under
stand narrative not just as a representation of the time course and causal
connections of particular, narrated events, but also as a mental image
that, besides being a property of texts and documents or performances,
is a common response to kinds of texts in specific contexts. The crucial
question here is the narrative quality of comics, comics qua narrative,
or what we can also call their narrativity. Storytelling is perhaps the
predominant mode of representation, or text type, in comics in general.
Therefore, it is useful to think of narrativity in comics as a gradient,
a more-or-less quality (see Herman 2009a, 6). Thus, some comics are
more prototypically narrative, while the others are less so, for instance,
due to the relative predominance of visual description, argumentation,
explanation, abstraction, or lyrical expression.
Introduction 17
The term ‘graphic narrative’ that I sometimes use in this book in lieu
of ‘comics’ is not, strictly speaking, synonymous with the medium of
‘comics’. The term has emerged in the recent decade in American aca
demic comics scholarship and literary studies, where it refers to parti
cular types of narratives in the medium. Specifically, the term has been
employed in relation to contemporary sustained nonfiction narratives
(Chute and DeKoven 2006, 767, 779, n1; Chute 2010, 3). One justifica
tion for using this term is the alternative it provides to the now standard
but also in some ways problematic category of the ‘graphic novel’, which
is widely used in criticism, publishing, and marketing. 25 The main prob
lem with the term ‘graphic novel’ is that many narrative works that are
given this label are not fictions at all. Marjane Satrapi’s Persepolis, Art
Spiegelman’s Maus, Alison Bechdel’s Fun Home, or Joe S acco’s Palestine
are nonfictions in different genres, not novels. As Hilary Chute has
pointed out, many nonfictional works in the comics medium “deliber
ately place stress on official histories and traditional modes of transmit
ting history” and thus “are deeply invested in their own accuracy and
historicity” (2010, 3). Therefore, the category of ‘novel’ is a questionable
point of reference. It should also be emphasised that to call some com
ics graphic narratives is not a value judgement, but simply a descriptive
choice. The terms ‘graphic novel’ and ‘graphic narrative’ are sometimes
used in ways that try to distinguish significant works of comics as liter
ature in contrast to ‘mere’ comics. The American comics critic Douglas
Wolk encapsulates the snobbery of what he calls the “if-it’s-deep-it’s-
not-really-comics” gambit in some reactions to the works of Spiegelman,
Satrapi, and Bechdel (2007, 12) by readers who do not feel comfortable
identifying themselves as comics readers, but prefer to call these works
‘graphic narratives’. If I use that term occasionally in this book, there is
no snobbery intended.
Medium
It is a kind of truism in comics scholarship to consider comics as a me
dium of its own. The problem, again, is that the concept of medium can
be defined in various relevant ways. A basic division in the meanings
of the term is that a ‘medium’ can refer, on the one hand, to a channel
for transmitting information, such as television, newspaper, film, or the
internet, and, on the other hand, to the materials, the technical means
and physical instruments of expression, as in the sense of artistic ex
pression. 26 In addition, a medium can mean those social practices that
characterise the use of a particular medium. For instance, media his
torian Lisa Gitelman defines a medium not only as a technology that
enables communication but also as a set of associated “protocols” or
social and cultural practices, normative rules, and default conditions
that gather and adhere around that technology. The latter involve norms
18 Introduction
and standards about how and where the medium is used, and express
a variety of social, economic, and material relationships (2006, 5–7).
Conceived in this sense, the comics medium could then mean the ways
in which comics are made, published, and read; their reading protocols;
and reading culture. Here, ‘medium’ means in particular a system of
communication and information that has a certain material support and
is defined by a set of associated social practices, such as conventions of
reading, genres, the publication format, and channels of distribution.
Finally, different media can also be specified on the basis of their use
of one or more semiotic systems (linguistic, visual, auditory, gestural,
spatial) to transmit their contents, and which are linked to different ap
proaches to narrative. In this sense, comics is a multimodal, or multi
channel, medium that can make a unique contribution to the study of
the relation between narrative and medium.
Medium Specificity
Many of the presupposed medium-specific features of comics have close
parallels across media and narrative art forms. Compare, for instance,
the function of the gutter—the space in between the panels—and the
use of ellipses and gaps of information in written narrative fiction, such
as between paragraphs or scene and chapter breaks, acts and scenes in
theatre, shot transitions in films, the segmented structure of a narrative
mural painting or a fresco, or transitions between lines and stanzas in
a poem. Segmented sequences and fragmentation function as a provo
cation to meaning-making and problem-solving across the arts, invit
ing the audience to make a connection and fill the gap. Similarly, the
alternation between sequential reading and a more global perspective
on the composition can play a significant role in reading picture books,
and graphic design can create complex relations between words and im
ages comparable to comics. Nor are combined forms of text and image,
such as speech balloons, unique to comics, even if comics are famous
for them.
To hold a very strict notion of medium specificity can lead narrative
theory to focus on an unproductive guarding of borders between the
media. However, paying insufficient attention to the question of medium
can result in overly facile equations between the presupposed isomorphic
qualities across media. One simple alternative, therefore, is to study how
comics are, at once, different from and similar to other kinds of nar
rative media. The stance of this book in this respect is twofold. First,
The Narratology of Comic Art questions the idea of hermetic bound
aries between comics and other narrative media or arts, especially the
closely related forms of visual storytelling, such as single-panel cartoons,
picture books, and animation films, on the level of their formal qualities.
Second, this book argues nonetheless that certain qualities, conventions
Introduction 19
and strategies, and their combinations are more expected of the medium
of comics. The challenge, then, is to explain how the (relative) distinc
tiveness of certain devices, conventions, and forms of comics can be un
derstood, without seeking some ahistorical essence or formal purity for
the medium.
The medium-specificity thesis often has a normative dimension,
recommending that artists exploit the distinctive possibilities of the me
dium and, consequently, define artistic accomplishment and skill on this
basis. In its most radical forms, then, the doctrine of medium specificity
can contribute to a species of purism. A more relativising understand
ing of medium specificity, and the one that is sensitive to the historical
changes of intermedial relations, holds that media have a range of repre
sentational, expressive, and formal capacities, some of which are typical
to them, even if subject to appropriation by other media, while they also
have other capacities that they share with other media but may put to
different uses. For instance, in his discussion of the medium-specificity
thesis, the philosopher Noël Carroll has usefully specified “distinctive”
to mean two things in this context: the notion that certain of the effects
of the given medium are managed both (1) better than other things the
medium does and (2) “better than said effects are managed by the media
possessed by any other artform” (2008, 36). For Carroll, a ‘medium’
refers especially to the artistic materials—paint in paintings and sounds
in music—and the material instruments, such as a paintbrush or a film
camera, used in shaping these materials. Thus, for Carroll, media are
the physical media. However, I see no particular reason why the same
notion of specificity could not be extended to refer to the uses and ef
fects of the semiotic resources—words, images, and sounds—that char
acterise a given medium. For instance, many narrative media are capable
of creating the impression of a character’s inner life, sense perception,
and personal voice, but they manage these effects differently, by using
and combining different materials and resources such as words, images,
sounds, and graphic style. Where one medium may have the advantage
over other media and do something better, it may detract in some other
way or must find ways to circumvent the challenges posed by the materi
als, instruments, and resources by which it is characterised.
Medium specificity in this sense, then, is not a determinative cate
gory, but a relative and graded one. The notion of ‘affordance’, derived
from environmental psychology and redefined more recently in multi
modal research, and which refers to the potentialities and constraints
of different resources (or modes) of meaning-making, provides us with
one further specification of medium specificity in this sense. In multi
modal research, affordance means, more precisely, “the materially,
culturally, socially and historically developed ways in which meaning
is made with particular semiotic resources” (MODE 2012. Glossary
of multimodal terms). 27 For instance, an affordance of images is the
20 Introduction
capacity to describe visual experience in visual and spatial detail, while
in language, it may be relatively easier to convey the idea of changing
temporal perspective, negation, or causal relations than in single im
ages. The affordances of a medium are conditioned by those semiotic
resources that the medium uses and that it may maximise to a greater
effect than other media.
Yet another useful way to conceptualise medium specificity, and one
which complements the idea of affordance, is the notion of the con
straint. A constraint in this respect can mean a restriction imposed by
the material form, the semiotic source, the combination of the material
form and the resources, or the publication format. To better illustrate
this, let us take two perspectives on the concept of the constraint, one
from comics scholarship and the other from the poetics of experimental
comics.
In their Power of Comics (2009), Randy Duncan and Matthew
J. Smith argue that the comic book is characterised by the following
formal constraints:
• spatial limitations (number of pages, page size)

• reproduction technologies (paper quality)
• unrealistic images (comics are two-dimensional and lack the photo-
realistic qualities of some other forms of visual storytelling media)
• limited capacity to control the reader (readers can view panels and
pages in any order and for any duration)
• the page as a unit of composition (allows some control over the
reader)
• the conception of images as selected moments
• interdependence of words and pictures
• artistic skill (what the cartoonist is able to achieve)
• the serial aesthetic (most mainstream comic books are published as
episodes in an ongoing saga)
(Duncan and Smith 2009, 119–120)
In fact, many of the constraints included here reflect the general capa
cities of comics as a medium and not just the comic book format. These
qualities comprise, in particular, the limited chance to control the order
or reading, or the ability to read linearly for the panel sequence and
at once freely look at the overall arrangement of the panels. Likewise,
the importance of the page layout as a unit of composition, the notion
of the panel as a unit of time, the cartoonist’s artistic skill, and the in
terdependence of words and images are common features in the medium
across various genres and publication formats. By contrast, the con
straints of unrealistic images and the conception of images as “frozen”
moments of action are open to discussion. I will revisit these questions
later in the book.
Introduction 21
The close connection between a medium-specific constraint and a
capacity becomes evident in experimental comics that have used con
straints as a means to investigate the basic conventions of the me
dium. In particular, the activities of the group of French cartoonists
and comics theorists known as OuBaPO (Ouvroir de bande dessinée
potentielle), founded in 1992, have exposed, by inventing deliberate
limitations according to which comics should be made and proposing
changes to existing norms and practices, some of the medium’s most
persistent norms and limitations. 28 In the first volume of the OuBaPO
group’s works, Oupus 1, published by L’Association in 1996, Thierry
Groensteen enumerated the medium-specific constraints that could
be performed on comics and grouped these constraints into two basic
categories: generative constraints, which can be employed to produce
new works, and transformative constraints, which alter existing works.
The original typology included ten generative and seven transformative
constraints. 29
The deliberate generative constraints are as follows:
• Iconic (or iconographic) restriction: The exclusion of specified picto

rial elements, such as human figures, from the strip.
• Graphic restriction: The restriction or exclusion of some graphic ele
ment or shape, such as a human face or a colour, from the strip.
• Scenic restriction: Constraints of the scene within the strips and the
way strips are framed (for instance, every panel is drawn from the
same viewpoint).
• Iconic (or iconographic) repetition: The repetition of a single image
or sequence of images.
• Multireadability: Comic strips or pages can be read in more than
one direction. This has subcategories such as
Acrostic strips: The panels of the comic can be read both horizon
tally and vertically;
Palindromic strips: The strips can be read both backwards and
forwards.
• Reversibility (upside down): A comic that can be read back to front.
Famously, Gustave Verbeck invented this technique in his 1903
comic The Upside-Downs Little Lady Lovekins and Old Man
Muffaroo.
• Overlapping: The folding of the strip or the page alters the comic,
thus revealing a new meaning, story element, or a story. Also, add
ing elements of other supports, such as a clear acetate sheet, may be
used.
• Random order or consecutiveness: Strips whose panels can be read
in any order.
22 Introduction
• Regulated distribution: Any pictorial element can be regulated by
mathematical, oulippian (i.e. relating to the contraints invented by
members of the literary group Oulipo), or other constraints.
• Geometrical arrangement: The layout of the comic strip frames fol
lows predetermined constraints in some geometrical shape.30
If we examine in a little more detail how these constraints are conceived,

we can see that many of them work with or against certain fundamental
norms and expectations about comics. Several concern the spatial form,
the principle of sequentiality, and the conventional order of reading. For
instance, the repetition of the same pictures, pictorial elements, panels,
or sequences makes visible the expectation that individual panels should
add something new and valuable to the ongoing sequence and, further,
that the diversion from this norm should be motivated, i.e. indicate a
significant change in the story content, perspective, or narrative rhythm.
The constraint of reversibility and experiments with random order
and geometrical arrangement work with and against expectations about
the typical order of reading for comics. One such expectation is that
panel relations should be neither random nor mechanical but motivated
by the visual composition and the narrative content. Scenic restriction
points out that it is expected of comics that the viewpoints keep chang
ing from one panel to another. Other constraints in this category involve
the central compositional structure of the panel frame and the way in
which the frame is related to perspective in the image. Iconic and graphic
constraints, in turn, undermine certain basic expectations concerning
picture content. Specifically, these constraints make visible how the use
of shapes, such as recognisable characters in a sequence of images, or the
depiction of faces increases narrativity in comics.
The constraints in these two lists do not describe the essence of com
ics, but they manifest many key conventions and expectations. The
premise of this book is that only when we create a more comprehensive
understanding of the capacities and constraints of comics, together with
a better understanding of the shared elements between other narrative
media, can we start to modify existing narratology in this field in inter
esting ways. One great challenge in this sense is that the narratology of
comics needs to develop in three different directions at once: through
the analysis of the specificities of the medium, the investigation of shared
elements across narrative media, and by studying the common properties
between related media. The latter involves a kind of mid-level theoreti
cal abstraction between the capacities of the medium and the common
qualities and strategies of visual and multimodal narratives. For Hans-
Christian Christiansen, the deep structure of visual storytelling com
prises techniques that are “based on general models of storytelling or
even on universal human experience”, such as facial expressions and
postures, relatively independent of style, but to which different forms of
Introduction 23
visual art give different prominence due to their particular substances
and constraints (2000, 110). Hence, the narratology of comic art needs
to develop towards both the specific and the general, not forgetting the
shared middle of related visual and multimodal forms of expression.
Choice of Examples
This book focusses on conceptual issues and is theoretically motivated.
As such, the choice of examples is consistent with the need to investigate
the medium-specific capacities, constraints, and expectations of story
telling in comics. Individual works in comics have a seminal role in this
book as the ultimate testing ground for all theoretical claims. A great
number of the examples included here are formally innovative works
that explore the possibilities of storytelling and, in many cases, also take
issue with their medium, i.e. explore the limitations, affordances, and
constraints of the medium. Some examples, from Chris Ware to Fred, go
beyond the expressive potential of current practices. Such works, where
the narrative medium itself becomes in some way a question, serve parti
cularly well to distinguish major narrative conventions, capacities, and
constraints of comic art. However, my main focus is not formal innova
tion in its own right. Also, many popular comics, from Disney comics
to Calvin and Hobbes, are included in the discussion to illustrate how
popular comics may explore their affordances and constraints.
Although my focus is on narrative forms and conventions, I do not
wish the approach to be wholly synchronic vis-à-vis the history of com
ics, that is, limited to concerns about the presupposed narrative univer
sals of the medium in a more or less contemporaneous moment of time.
Thus, in the course of the discussion, I will make some forays into the
history of the medium, including a whole chapter dedicated to the ques
tion of the development of narrative forms in ‘premodern’ comic strips
of the early nineteenth-century British satirical journals.
Certain strong trends that have marked the art of comics in the
European and North American contexts over the past three decades
are reflected in my choice of examples. These trends include, in parti
cular, the independent comics revolution, the emergence of autobiogra
phy and other forms of nonfiction comics, the revision of the superhero
genre, the influence and predominance of manga, and the development
of the graphic novel. The majority of my examples represent European
comics traditions and American comics. In the European context,
I devote particular attention to Franco-Belgian and British comics, but
some examples of Italian and Finnish comics are also included. Most
English-language studies in the history, theory, and culture of comics
focus on American comic books and comics scene, whereas French com
ics theory is concentrated on the French–Belgian tradition of the comic
album. This book clearly goes against the grain in this sense, developing
24 Introduction
extensive dialogue between Francophone and Anglo-American comics
theory and artists. Manga is generally not included in these readings
even if Jiro Taniguchi’s A Distant Neighborhood has a central role in
the discussion of the narrator concept offered here. Note, however, that
Taniguchi’s graphic novel is a Europeanised and Westernised manga to
begin with. Not only is its reading order reversed from the Japanese stan
dard, but also many elements of the visual style in the original Japanese
version are closer to Western comics than manga.
Despite the emphasis on longer, sustained narratives in comics, I con
tend that most of the theoretical claims in this book, exempting ques
tions relating to page layout and the overall appearance on the page, are
general enough to be extended to the shorter form of the comic strip.
Comic strips have their particular generic outlook, expectations, and
narrative poetics that differ, in the main, from longer formats. Perhaps
most typically, in conventional gag-a-day comic strips with three or four
panels, narrative breakdown is based on a specific effect in the given
limited space so that the last panel, or sometimes the penultimate panel,
functions as the gag panel. However, serial or continuity strips follow
two different kinds of narrative logic at the same time—to be meaning
ful in three or four panels and as a longer narrative—and, at least since
the 1950s, reflecting this double logic and expectation, many strips have
been republished as magazines or albums. Therefore, the boundaries
between the comic strip and a sustained narrative work in comics are
not always that clear-cut, and it is not in the interests of this project to
try to enforce that distinction.
Outline of This Book

The Narratology of Comic Art is divided into five main sections. These
sections focus on the sense of time in comics, narrative showing and
graphic style, narrative transmission, the presentation of speech and
thought, and the relationship between narrative form and the publica
tion format. The ‘Time in Comics’ chapter of this book examines the
issue of temporal organisation in comics by evaluating the basic narra
tological distinction between story time (the chronological order of the
events) and discourse time (the order in which these events are presented
in the comic) in this context, and employs the conceptual framework of
temporal order, duration (or rhythm), and frequency that derives from
that distinction. The discussion of temporal order will develop a more
comprehensive understanding of panel relations, the means of connec
tivity, and the levels of sequencing in comics, thus exploring the ways in
which narrative comics exploit the ratio between the compositional units
of the panel, sequence, and layout to convey a sense of time. Sequencing
in comics is treated here as a multilevel issue, involving the dimensions
of chronological, psychological, and presentational sequencing. The
Introduction 25
discussion of narrative rhythm considers rhythmic variation in relation
to the narrative units of a panel and a scene, as well as certain medium-
specific techniques that are based on multidirectional panel relations.
All these determinants of time in comics are further related to the basic
conventions of reading comics that make many of these temporal effects
possible in the first place.
The second part of the book consists of three chapters that address
the issue of narration through narrative drawings and means of graphic
style. Chapter 2, ‘Narration as Showing’, focusses on the ways in which
images tell a story by showing a narrative event, such as some character
in action. The fundamental premise of this chapter is that images in
comics can visually articulate the narrative by showing certain things
in certain ways. More precisely, the concept of ‘graphic showing’ re
fers to the visual content of narrative drawing that the image presents
for looking, specifically figures and figuration, such as characters and
their situations, milieu, or world, which evoke a story-like scenario
and thus inspire a narrative response from the readers. This chapter
develops one of the fundamental arguments of this book, which is that
the development of the narratology of comics requires that we pay at
tention to the various features of graphic showing that interact with
verbal narration and perspective (focalisation). Chapter 3, ‘Character
as a Means of Narrative Continuity’, develops the notion of narration
by showing further by examining the way in which the repeated char
acter figures, in an ongoing situation or action, function as a basic
sequential model and a tool of narrative continuity. The notion of sa
lience is especially helpful in conceiving the characters’ function in this
way. Narrative salience refers to the importance of how something is
shown, and how certain elements, in particular relating to characters
and their actions, are significant for an understanding of the narrative
as a coherent whole. In this regard, theories developed in film studies
that stress the importance of characters’ spatio-temporal paths, and
how those paths can function as spatio-temporal attachment and sub-
jective access for the viewers, have much to offer for a narratological
understanding of comics. Chapter 4, ‘Graphic Style, Subjectivity, and
Narration’, discusses the narrative function of graphic style in channel
ling story information. In particular, this chapter concentrates on two
specific narrative functions of style in comics—stylistic variation and
rupture—as means of presenting the characters’ subjectivity, i.e. their
thoughts, perceptions, and emotions. The concept of ‘mind style’ also
allows us to rethink the relation between graphic style and narration
with regard to the presentation of a character’s mind. Furthermore,
the chapter argues that wordless comics have the unique potential to
undermine the ascription of states of mind or perception to characters
since it may be difficult to know the degree of subjectivity of vision
from images alone.
26 Introduction
The third part of the book focusses on narrative transmission and
mediacy, i.e. the storytelling process that mediates between a narra
tive comic and the reader. Chapter 5, ‘Narrative Agency’, addresses
the question of narratorial authority and enunciation through a case
study of Jiro Taniguchi’s graphic novel, A Distant Neighborhood, and
examines the reasons why it is challenging to describe the source of
narration in such works. By ‘narrative agency’, I mean the conceptuali
sation of a kind of global frame of narration that enables the reader to
estimate the meaning and importance of the various visual and verbal or
visual-verbal elements, their relations, and their alternating perspectives
at the micro-level of the story. The chapter will start with a compre
hensive discussion of the narrator concept. Chapter 6, ‘Focalisation in
Comics’, develops the study of perspective in this multimodal medium,
by examining the distinction and interplay between perceptual and
cognitive focus, the interaction between various simultaneous perspec
tives, as well as the relation between verbal narration and graphic style.
More precisely, the chapter investigates certain hitherto little-studied
aspects of perspective-taking in comics: the relationship between nar
rative voice, mode, and showing; the epistemic access to the point of
perception and of what is perceived, concerning the spatially explicit
(or determined) point of perception in graphic images; the distinction
between perceptual and cognitive focalisation; the issues of embedded
and simultaneous focalisation in a literal sense, i.e. the simultaneity of
different visual focalisers both within and without the image frame; and
the complex scale of intermediate (visual) focal points between inter
nal and non-character-bound positions, or between personal and im
personal viewpoints. Chapter 7, ‘Characterisation in Comics’, addresses
how characters are presented in comics and develops the issue through a
sustained analysis of two adaptations of Dante Alighieri’s Inferno. The
chapter will elaborate on how comics prompt their readers to ascribe
properties, or traits, to their characters, that is, how the reader’s con
struction of a character is cued in the elements of the story. The chapter
focusses on the main parameters of characterisation in comics, in parti
cular, with regard to the characters’ mimetic (realist) function, i.e. how
they can be perceived as possible persons. Moreover, the chapter will
discuss the ways in which characters can achieve a sense of psychologi
cal or conceptual complexity in comics, emphasising the significance of
types and caricature in the history of the medium.
The fourth part of the book focusses on speech and thought repre
sentation and the narrative function of dialogue. Chapter 8, ‘Presenting
Minds in Comics’, discusses three central issues in the presentation of
characters’ (or the narrator’s) minds in comics: the mimetic aspect of the
image, the problem of free indirect discourse, and the interaction be
tween visual focalisation and verbal narration in first-person narration.
The theoretical discussion will be illustrated by representative extracts
Introduction 27
from comics, moving from third-person to first-person narration, both in
fiction and nonfiction. The analysis of these examples shows that graphic
narration subjects to doubt some of the theoretical presuppositions, prev
alent in much narrative theory that is based on literary narrative fiction,
and requires us to critically examine various key distinctions, such as
those between first- and third-person narration; between direct, indirect,
or free indirect discourse; and between telling and showing. Chapter 9,
‘Dialogue in Comics: Medium-Specific Features and Basic Narrative
Functions’, focusses on the dialogue form in comics as a key narrative de
vice and examines the elements and narrative functions that characterise
conversational scenes in this art. The multimodal character of conversa
tional exchange in comics requires that we attend to the interaction be
tween the utterance and the elements of the narrative drawing, that is, the
ways in which the dialogue form (as written and drawn speech) interacts
with what is shown in the image. Crucial aspects in conversational scenes
in comics are facial expressions, gestures, body language and shape, and
participant involvement. Equally, the expressive functions of typogra
phy and pictorial symbols, onomatopoeia, as well as graphic style, panel
framing, and page layout, can play a major role in creating conversational
scenes. The chapter is structured around four questions: the embodied
speech situation in comics, the bond between the speaker and the utter
ance, the temporal and rhythmic functions of speech balloons, and the
narrative function of visual and verbal contrast in dialogue.
The last part of the book discusses the relationship between narra
tive form and the publication format. Chapter 10, ‘Picture Story and
Narrative Organisation in Early Nineteenth-Century British Caricature
and Comic Strips’, is a study in diachronic narratology that investigates
the basic narrative techniques and operative principles of juxtaposition,
sequentiality, and simultaneity in the historical context of the so-called
British Golden Age of Caricature (1780–1820) and the caricature maga
zines published between 1825 and 1835. This investigation will specifi
cally examine the conception of panel arrangement and panel relations,
illustrating how the early cartoonists experimented with the principles
of juxtaposition, sequentiality, and simultaneity in response to the de
mands of the publication format, thus contributing significantly to the
gradual emergence of the modern comic strip.
Notes
1 Mary-Laure Ryan argues that while a narrative is a “semiotic object”, nar-
rativity refers to the quality of “being able to inspire a narrative response”
(2006, 10–11). See also Wolf (2003).
2 See Ryan (2004, 1–40, 2006, 4–7).
3 Kress and van Leeuwen define a multimodal document as “any text whose
meanings are realised through more than one semiotic code” (2006, 177).
See also Stöckl (2004).
28 Introduction
4 See David Herman’s definition: “Postclassical narratology […] contains clas
sical narratology as one of its ‘moments’ but is marked by a profusion of new
methodologies and research hypotheses; the result is a host of new perspec
tives on the forms and functions of narrative itself” (1999, 2–3).
5 See Ryan (2005, 2013), Ryan and Thon (2014, 12–13), Jenkins (2006,
195–196) for the definition of transmedial storytelling.
6 Writing about the contribution that narratology can make in visual studies,
Bal emphasises that “the analysis cannot be limited to the application of
narratological concepts to visual representations (‘How do images tell?’)”
(1990, 744). Rather, she claims, “the confrontation between the narratolog
ical apparatus and the visual image inevitably changes or even subverts the
categories” (ibid.).
7 See, for instance, Bordwell (1985) and Branigan (1984, 1992).
8 In his 1988 article “Narration as Supplement” (2014, 179), Groensteen also
“dreams” of founding a narratology that would be specific to comics.
9 Chatman develops this distinction on the basis of Roland Barthes’s notions
of nuclei and catalysers, explained in “Introduction to the Structural Ana
lysis of Narrative”, in Image Music Text. Trans. Stephen Heath. New York:
Hill and Wang, 1977, 79–124.
10 Such narratologically informed textbooks include, for instance, Saraceni
(2003), Miller (2007), Kukkonen (2013b), and Postema (2013).
11 For Genette, the narrative mood concerns techniques of filtering the narra
tive, such as focalisation and techniques of speech and thought representa
tion (1980, 161–211).
12 Steven Surdiacourt also proposes a model of comics narratology in his PhD
dissertation (2015), divided into the issues of segmentivity/sequentiality/
structure, focalisation, and narrative voice.
13 I side with Marc Singer, for whom “putting narrative theory into practice
is primarily a matter of conducting an informed, close reading of the text,
looking at all the ways the narrative elements contribute to its meaning”
(2012, 59).
14 For instance, as much research shows, forms of speech and thought represen
tation, such as free indirect discourse, can realise a multitude of functions in
the nineteenth-century novel (McHale 1978, 274–284, 1993, 60–62; Cohn
1999, 163–180).
15 See, for instance, Jared Gardner, on how the popularity of the comic book
raised the question of page composition to a new level, both in terms of a
new kind of visual whole and narrative breakdown (2013, 249–250).
16 See, for instance, Groensteen (1999), Meskin (2007, 376), Hatfield (2010, 5),
and Beaty (2012, 31–44). Samuel R. Delany has argued that instead of a
definition, we need a “careful, analytic description of what is vital, intrigu
ing, newly noticed, and wondrous about comics (what they are; how they
work)” (1999, 245).
17 For Groensteen, iconic solidarity means “interdependent images that, partici
pating in a series, present the double characteristic of being separated—this
definition dismisses unique disclosed images within a profusion of patterns
or anecdotes—and which are plastically and semantically overdetermined
by the fact of their coexistence in praesentia” (2007, 17).
18 For instance, Jules Feiffer argues (1965, 68) that “comics are movies on
paper—the final dream!”. Elisabeth Potsch and Robert F. William voice the
same view: “Comics is cinema without motion or sound” (2012, 13).
19 See also Miodrag, who points out that literary quality (literariness) in com
ics has much less to do with the ability to tell a story than using language in
particular self-reflexive and poetic ways (2013, 18).
Introduction 29
20 Miodrag derives the term logophobia from Dylan Horrocks, who argues
that, for Scott McCloud, it is “as if the very presence of words—any words—
in a comic is a potential threat to its identity as a comic. To protect that iden
tity, it is essential for the pictures to dominate the words” (2001, 5).
21 Groensteen conceives his definition of comics as “reasonable”, i.e. one that
pertains not to all comics, but “the totality of historical manifestations of
the medium” (2007, 17).
22 Beaty argues, similarly, for an institutional definition of comics (2012,
31–44). A social and institutional definition of comics is also used in Hague
(2014, 16–18) and Barker (1989, 8).
23 For Gerald Prince, for instance, a narrative is “the representation (as product
and process, object and act, structure and structuration) of one or more real
or fictive events communicated by one, two, or several (more or less overt)
narrators to one, two, or several (more or less overt) narratees” (2003a, 58).
24 Following Monika Fludernik’s definition of the term “experientiality”,
i.e. the communication of anthropomorphic experience (1996, 30).
25 The term has refered to different kinds of long-form comics since Will Eisner
called his A Contract with God (1978) a “graphic novel”. For the debate
over this term, see Labio (2011, 125–126) and Hatfield (2005).
26 See Ryan (2004, 15–20) and Carroll (2008, 35).
27 The notion of affordance derives from James J. Gibson’s interactionist eco
logical theory, where an affordance is a way of perceiving a given environ
ment in relation to what the various elements of that environment afford one
to do—‘one’ referring here to both people and animals (1979). The notion
has been widely applied in the study of multimodality since Günther Kress
adopted the term (2010). See also Kukkonen (2013b, 167).
28 The group’s cartoonists have included, for instance, François Ayroles, Anne
Baraou, Gilles Ciment, Patrice Killoffer, Jean-Christophe Menu, and Lewis
Trondheim.
29 The transformative constraints include the means of manipulating an
already-existing comic by expansion, reduction, substitution graphic rein
terpretation, reframing, and other techniques.
30 I am using Brad Brooks’s translation of these constraints, and some of his
wording in the definition of the terms. Please see http://www.paolacarbone.
com/cultura/2011/oubapo.pdf. Jacques Dürrenmatt claims that the two es
sential constraints in the medium are the page layout and the constraints that
the writer and the cartoonist, even when they are the same person, impose
on each other (2013, 129). He distinguishes these from stylistic constraints
(the expectation of stylistic unity) and self-imposed constraints, such as the
OuBaPo activities.
Part I
Time in Comics
1 Time in Comics
The opposition between erzählte Zeit (story time) and Erzählzeit

(narrative time) (…) is less relevant perhaps in other forms of narrative
expression, such as the roman-photo or the comic strip (or a pictorial
strip, like the predella of Urbino, or an embroidered strip, like the “tapes
try” of Queen Matilda), which, while making up sequences of images and
thus requiring a successive or diachronic reading, also lend themselves to,
and even invite, a kind of global and synchronic look—or at least a look
whose direction is no longer determined by the sequence of images.
Gérard Genette (1980, 33–34)
It is a kind of commonplace in comics studies to argue that space and

time are intricately intertwined in this medium.1 However, the discus
sion of the spatialised forms of temporality in comics has often remained
rather one-sided. It has been common, for instance, to privilege parti
cular elements of the composition, such as the space between the panels
or the panel as a unit of time, to illustrate how comics prompt readers
to make inferences about temporal progression. In this regard, narra
tology can provide a more comprehensive account of time in narrative
comics, and it can also offer a precise vocabulary and a widely tested ap
proach to this topic. Specifically, the narratological distinction between
story-time (the chronological order of the events) and discourse-time
(the order in which these events are presented in the comic) allows us to
investigate the basic forms of temporal structure in comics. While parts
of this model have been applied in earlier comics studies, much of its
potential remains to be explored.
Sometimes comics explicitly reflect on the capacities that affect the
communication of temporal meanings in their spatial composition. Chris
Ware’s graphic novel Building Stories (2012), a boxed set of fourteen
different kinds of printed works in comics, investigates the relationship
between story-time and discourse-time at various levels, thus ambiguat
ing the temporal and chronological relationship between various parts
of the story. Building Stories poses the question of narrative memory by
allowing the reader to “build a story” in any possible order of its parts.
34 Time in Comics
The navigational mock directions given on the back of the box further
encourage the sense of freedom and randomness in this regard. They
indicate, in a kind of ironical version of the method of loci, places in
any “average well-appointed home” where one can “set down, forget
or completely lose” any number of the parts of the composition. 2 Thus,
the title of the work itself can be understood in its literal sense as stories
about buildings, stories told by buildings—indeed, the Chicagoan tene
ment house that is central to the story actually has a narrative voice—
or stories told in relation to buildings and memories about their spaces
(the method of loci). There is also the metaphorical meaning of the
building blocks of a story—the toy set, the game board, or the model kit
analogy—with which the reader can creatively engage in order to create
a story within some temporal frame.
Building Stories sets the different printed materials that it comprises
in a circular network of complementing units, thus suggesting the idea of
a story that needs to be processed through a web of interconnected sto
ries rather than conceived in a strict linear order. All the different story
units have the potential to serve as background memory to be used for
filling in the blanks created by the other parts. The order of the reading
can always be reversed if desired or needed. This freedom highlights the
importance of something that the readers conventionally do with narra
tives regardless of the medium: establish the chronology of their events,
identify the characters, construe their personalities and relationships,
and make note of the significance of the milieu and the circumstances.
Thereby, Building Stories extends the manipulation of the order of the
events from the level of presentation to the order of reading.
That said, however, it must be stressed that the temporal open-
endedness of Building Stories concerns primarily the order of reading
the various printed works rather than the sense of chronology in the
story. The story-time is not random even if the temporal relations bet
ween the different instalments are not clearly defined. How is this pos
sible? First, the reader can make inferences about the order of the events
on the basis of the image contents pertaining, for example, to signs of
aging in the reoccurring characters, their changing relationships, homes,
and milieu. Furthermore, we can detect indications of time in the depic
tion of the protagonist’s thoughts and experiences (how they develop
and change, or imply earlier or foreshadow future events). The main
storyline involves a young woman with a prosthetic leg, a former art
student and later a florist, whose life the reader may follow for about a
decade, with some flashbacks to her childhood, and intersecting stories
about the people who lived in her tenement house in Chicago when she
was in her twenties. Later, we follow her life as a mother in a suburban
home. Second, one of the instalments, the hardcover book with a gilded
spine that is modelled after America’s Little Golden Books for children,
functions as a kind of aid for the construction of a temporal continuum.
Time in Comics 35
Figure 1.1 April 20th, 2005. Building Stories (2012) © Chris Ware.
The book gives an exact date, 23 September 2000, for a one-day story
in a three-floor Chicago apartment building. Consequently, this date,
which is also the title of the instalment, and the flash-forward of the last
page in this instalment (20 April 2005), may be used to determine the
order of the other events in the other instalments, where there is no such
precision, or at least assess what might be the most coherent arrange
ments in this regard (Figure 1.1).
It is possible to read Building Stories as a network of events and mem
ories to be found and links to be made, and perhaps again to be remade
as other instalments offer new information. Yet, there is also much sup
port for a chronological understanding of the events. Thus, chronolog
ical order and the network of interconnected moments are not opposite
temporal structures, but they can be conceived of as complementary op
tions and in terms of their interaction. Building Stories both undermines
and reinforces the distinction between the order of the events in the story
and the order of their presentation. The readers need to be particularly
active in creating a sense of chronology out of the seemingly intercon
nected moments.
In what follows, this chapter will evaluate the relevance of the funda
mental narratological distinction between story-time (the order of the
events) and discourse-time (the order of their presentation) in the con
text of narrative comics. However, this framework needs to be related
to the conventions and expectations of reading comics, such as the con
vention of a kind of synchronic look that Genette mentions above. It is
not merely a matter of formal analysis. Also, Building Stories highlights
the significance of these conventions through its great variety of shapes
and formats of comics, from pamphlets and tabloid-size magazines to
cloth-bound books.
36 Time in Comics
Story and Discourse in Narrative Comics
The temporal structure of narratives has been among the most central
questions of narratology throughout the history of this discipline.3 The
key distinction in this respect is between story-time, meaning the tem
poral order or succession of the events in the story, and discourse-time
(or narrative time), in the sense of how the events and the story contents
are arranged and presented. Narratology has not invented this distinc
tion, but has systematised its study, providing us with a much-tested ap
proach for analysing temporal organisation in narratives on that basis.4
However, what complicates the matter is that the scope of this distinc
tion has always been under some debate. 5 The Genettean understanding
of ‘discourse’ comprises, besides narrative time, the entire expression
plane of narrative mood and voice. Narrative mood refers to the regu
lation of narrative information through distance from the things that
are told, such as by means of perspective, while narrative voice means
the act of narrating through a narrative situation, in particular by a
narrator. Other broader definitions contend that ‘discourse’ comprises
elements of style or what is specific to a medium. Similarly, the ‘story’
(the ‘what’ that is narrated) is not always limited to the chronological or
der of the recounted events of the story, but it may refer to the basic ele
ments of the story content, in particular the characters and their world.
The problems that result from maintaining the broadest definitions
of these two terms have not remained unnoticed by narratologists who
have turned their attention to visual narratives. There are, at least, two
main objections that we can level against the broad definition of dis
course in this context.
First of these is the difficulty in differentiating between graphic style,
or the materiality of the image, and the narrative meaning of the images.
Martin Schüwer has problematised the usefulness of the story-discourse
distinction on this basis (2008, 23), by arguing that this divide can
not be as clear-cut in comics as it may be in the linguistic structures of
verbal narratives. The second challenge is that posed by Genette: the
global or synchronic look, or what is called “tabular” reading in refer
ence to a tableau (picture, painting, table), originally defined by Pierre
Fresnault-Deruelle (1976). More precisely, the notion of tabular reading
refers to features in comics that invite a nonlinear, or not only sequential,
reading of the panels and where, thus, the whole of the spatial arrange
ment merits a more global look and appreciation. There are great differ
ences between comics in this respect. For instance, certain “exploded”
scenes that are typical of Guido Crépax’s Valentina invite a pronouncedly
tabular reading: the panels on the page are not integrated into a logical
continuum in terms of a sequence, but these reflect the protagonist’s
mental state through relations of contiguity that are sometimes quite
complex (see Fresnault-Deruelle 1976, 23). Here, the composition invites
Time in Comics 37
a synchronic look at the whole as a unit of graphic design and narra
tion. Such arrangements defy the story-discourse distinction: is there any
sense of temporal order in scenes that focus on the protagonist’s mental
state or the various perspectives of the situation at hand?
Schüwer’s advice, which I follow here, is to restrict the application
of this distinction to the study of temporal structure.6 The limitation
will be beneficial for our analysis of perspective (focalisation), voice
(narrative mediation, the narrator), characterisation, and style in later
sections of this book. The issues of mood and voice will be discussed,
independently of the story and discourse divide, in chapters dedicated to
narrative agency, style, perspective, and the presentation of speech and
thought.
Order, Duration, and Frequency

Gérard Genette systematised the distinction between story-time and
discourse-time, based on the distinction between fabula and syuzhet in
Russian formalism, by proposing to study this relation through three es
sential determinants of time: order, duration, and frequency (1980, 35).
In the following, I will concentrate, in particular, on the questions of or
der and duration in comics storytelling. In the course of this discussion,
I will also examine a number of important theories of time and space in
comics studies, evaluating especially their narratological relevance.
Temporal Order
The narratological analysis of time focusses on “anachronous” sequences,
or anachronies that depart from the sense of narrative present that is es
tablished in the given narrative7 and thus involve some shift of balance
between the levels of story-time and discourse-time. The basic categories
of asynchrony in this regard are retrospection (flashback), anticipation
(flashforward), and the lack of temporal chronology, or what Genette
calls ‘achronism’.8 Retrospection involves the narration of past events
in relation to the narrative present, and anticipation involves the narra
tion of future events in relation to the narrative present (1980, 40). Such
anachronies can be more or less explicit, or take multiple (or embedded)
forms so that, for instance, an anticipatory passage includes retrospec
tion or a flashback includes further flashbacks (1980, 79). As a case of
‘achronism’, Genette treats certain passages in Marcel Proust’s In Search
of Lost Time, where the told events are cut loose from any temporal
situation, or their relation is random in this respect. For instance, the
succession of train stations that are described at the end of Sodome et
Gomorrhe evokes in the narrator’s mind a series of stories from different
times, only connected by the same space or a theme. Similar to stories
38 Time in Comics
that are connected through the same building in Building Stories, mem
ories are thus organised according to a sequence of places, or spatial
contiguity, rather than temporal continuity.
In narrative comics, anachronies can be identified in any significant
changes with regard to the progress of time in the story world. Tempo
ral gaps are perhaps the most obvious starting point for the analysis.
The gaps, and other types of temporal shifts such as flashbacks, can
take place between formal elements of the composition, such as panels,
strips or tiers of panels, pages, and double spreads, or between narrative
units, such as passages, scenes,9 chapters, and instalments. The follow
ing discussion of temporal order will proceed in three steps by looking
at the temporal function of the panel relation, the means of connectivity
between the panels, and the levels of sequencing.
Temporality in Panel Relations

Perhaps the most obvious element for manipulating temporal order in
comics is the panel relations that invite the reader to construct meaning
ful connections and fill in the gaps in information. The transition between
two panels does not necessarily indicate a temporal shift—a sequence of
panels may, for instance, depict the same character or object at the same
time from different angles—but temporal transitions between panels are
so common in the medium that they may amount to a kind of default
expectation. Notice how in this wordless page in Bastien Vivès’s graphic
novel A Taste of Chlorine (2011, Le Goût du chlore, 2008), which de
picts the glass ceiling of a swimming pool in several panels from a char
acter’s subjective perspective, the content of the images stays nearly the
same, but the relation between the point of perception and the object of
perception changes (Figure 1.2). The perspectival changes indicate slow
temporal progression, relating to the swimmer’s sense of movement on
his back in the pool.
Some of the most popular theories of meaning-making in comics fo
cus on the gaps in information between the panels in a sequence. In these
approaches, the missing information in the ‘gutter’, the space between
the panels that can have a literal form (an empty space) or remain virtual
(a mere frame separating two panels, or frameless panels, for instance),
involves at the same time a distinctive formal feature of the medium, a
convention of reading, and mental activity. In Understanding Comics,
Scott McCloud famously referred to the mental completion of gaps bet
ween panels as ‘closure’. Closure, McCloud specifies, is “the pheno
menon of observing the parts but perceiving the whole” (1993, 63), that
is, the mental activity of completing the missing information between
the panels to make a meaningful whole, or to use McCloud’s wording, to
“mentally construct a continuous, unified reality” (1993, 67). The idea
of gaps in information, which give rise to meaning in the reading process
Time in Comics 39
Figure 1.2 Extrait de l’ouvrage. A Taste of Chlorine (2011). Bastien Vivès

© Casterman. Avec l’aimable autorisation des auteurs et des Edi
tions Casterman.
by provoking the readers to make connections, is fundamental to com

ics. When two images or pieces of information are juxtaposed in space,
very little seems to be needed for the presentation to suggest a meaning
and, perhaps, a story. McCloud’s choice of the term, however, is confus
ing in the context of narrative theory, where ‘closure’ has an established
meaning as “the satisfaction of expectations and the answering of ques
tions raised over the course of any narrative” (Abbott 2005, 65–66).
Here, in other words, closure refers to an outcome in a narrative, and not
to all gaps of information along the narrative.
Yet, from our narratological perspective, the terminological confusion
is perhaps less crucial than the way in which the closure theory limits the
question of panel relations, and consequently the question of time, to a
40 Time in Comics
linear connection between two subsequent panels. As a side product of
this theory, the space between the panels, the gutter, is reified as a kind
of essence of the medium.10 In this respect, Jan Baetens and Hugo Frey’s
critique of what they see as the “inflated” value of the gutter concept is
helpful in two important ways. First, Baetens and Frey point out that the
diegetic function of the gutter can vary widely:
in certain cases, it [the gutter] may be utterly unremarkable (in the

two meanings of the words: unnoticed and also not worth noting);
in other cases, its role can be paramount (and of course, the same
remark applies to the visual, pictorial function of the gutter, which
is by all means a key aspect of page composition).
(2015, 121–22)
Thus, we must recognise the importance of the panel relation, not the
space in between. Second, in their critique of the closure theory, Baetens
and Frey point out that panels in comics, at least in longer works such
as the graphic novel, are structured on various levels simultaneously, in
cluding the strip or tier, which can be organised horizontally, vertically,
or as a combination of both; the page, which can have a wide variety
of sizes and formats; and the book, where the size and format can also
diverge widely (2015, 104–5). To these three levels, we could further add
forms of layout and panel organisation that go beyond the structure of
the strip and the grid, such as the impression of depth through stacked or
superimposed panels and the directionality of the strip/tier (see Bateman
et al. 2016), or the treatment of the whole page as a geometrical shape.
Consequently, the contact between the panels can, on all these levels, be
either sequential or more autonomous, that is, have a painterly function,
or prompt a global look at the composition as a whole.
Let us consider the manipulation of temporal order through some
panel relations in Nicolas de Crécy’s wordless comic album Prosopopus
(2003). The ‘mute’ quality of the story creates certain representational
pressures in its own right and promotes an associative mode of visual
reading that prompts the readers to pay attention to the breakdown of
the story on the visual plane alone.11 Beyond attending to the panel rela
tions, juxtapositions, and transitions, this also involves inferences about
changes in viewpoint, and paying careful attention to various visual
cues, such as facial features, gestures, clothes, and details of the objects.
Specific ‘time props’ and time cues are also highlighted. These include,
for instance, empty panels that indicate the duration of time, symbols
like numbers or the clock face, pictures within pictures, or changing
ratios of light and shadow, details in the landscape, or changing seasons.
Part of the challenge in reading Prosopopus and understanding the
chronology of its events is due to the effects of juxtaposition and montage
between the panels. Events and scenes that are spatially and temporarily
Time in Comics 41
removed from each other can follow each other without any explana
tory frame. What is particularly remarkable in this respect are larger
intrapictorial relations, including tiers of panels, a page or a double-page
setup, or panel-to-panel references throughout the whole album, that
may convey information about temporal relations through repetition,
visual analogy, or other meaningful correspondences between the panels
and their sequences. At times, De Crécy employs an analogous technique
to the cinematic match cut to indicate a simultaneous change of scene
and/or temporal frame: transitional panels that lead from one scene or
temporal frame to another through some visual detail in the image
(a spot on the floor, a tile on the wall, hands holding a tool/weapon,
fragment of a paintings, and so on), or the effect of zooming into a de
tail that ‘connects’ somewhere else, thereby establishing a graphic match
between the panels. In some panels, by contrast, the depiction of a video
camera viewpoint, supposedly held by the story’s namesake, the mon
ster Prosopopus, helps to create the narrative effect of condensed layers
of time. On a page towards the end of the story, one flashback follows
another, including the scene of assassination seen in the beginning of the
story. Here, the panels of the earlier scene are marked off by two (nearly)
blank panels, while the sequence is at the same time sped up by dimin
ishing and narrowing the size of the panels. This is a flashback of a flash
back that, we may assume, takes place in the main character’s mind.
Prosopopus illustrates to us how non-verbal information alone can
serve to construct complex temporal layers and relations in the story
and, thus, require that the reader engages in a lot of back-and-forth
checking of story elements beyond the immediate panel sequence. More
over, it makes manifest the significance of the narrative context, such
as a passage or scene, which allows us to determine the temporal or
other meaning of a given panel transition. The narratological potential
of McCloud’s six types of transition between the panels—moment-to-
moment, action-to-action, subject-to-subject, scene-to-scene, aspect-to-
aspect, and non-sequitur (serving no narrative purpose) transitions—is
compromised by the fact that the typology does not take the context into
consideration. Furthermore, the categories describe panel relations at
varying levels of organisation and meaning-making. As Bart Beaty has
pointed out, the category of scene-to-scene transition actually describes
a transition between larger units of narrative, that is, scenes of narra
tive, rather than just between two subsequent panels. The other types
of transition reflect more micro-level shifts of temporal frame, perspec
tive, subject matter, or causal connections. In contrast, the non-sequitur
transition involves a non-narrative relation between panels, such as a
challenge to narrative logic (Beaty 1999, 70).12
At the same time, McCloud’s understanding of the function of panel
transitions does not capture the full flexibility, instability, and hetero
geneity of their functions. Still other types include changes of narrative
42 Time in Comics
voice and mood, such as perspectival shifts between characters, or al
tering effects of angle, focus, and objectivity and subjectivity in the
viewpoint.13 One rather common type of transition is the change of
truth-value (modality in the linguistic sense) with regard to the image
content in the panels. The modality-to-modality transition, involving a
transition in the truth-value or credibility of what is seen, for instance,
in a dream, fantasy, hallucination, or memory sequence, is regularly
accompanied by stylistic markers, such as changes in the graphic line,
lettering, and colour, or alterations in verbal narration, layout, and per
spective. For example, in the Doll’s House story arc in The Sandman,
the powers of Rose Walker’s “dream vortex”, which threaten the sta
bility of the fictional world of The Dreaming, are made evident to
the readers by the fact that the panels of the sequence turn sideways.
Later, when Rose’s dreaming becomes even more powerful, the dream
sequence creates a vortex on two double spreads, where the few remain
ing panels give the impression of having been thrown around the space
of the composition.
My intention here, however, is not to expand McCloud’s typology
on its own terms, but to integrate the general notion of panel transi
tion, or panel relation—rather than the gutter—to the narratological
analysis of temporal order, rhythm, and frequency. Many of the short
comings of the closure theory can be mended by Thierry Groensteen’s
much more comprehensive model of ‘arthrology’ that seeks to cover the
entirety of the relations between panels in the spatial organisation of
comics.14 Beyond the pure linearity of subsequent panels, the question
of panel transitions is conceived here in relation to the two-directional
dependence of panels in a sequence (the notion of ‘iconic solidarity’ bet
ween the panels).15 The main degrees of articulating panel relations in
this model are the elementary, linear panel-to-panel organisation (linear
or restricted arthrology) and the principle of braiding (translinear and
distant or general arthrology) that concerns panel relations other than
those between immediately juxtaposed images. The latter can include,
for instance, interrelations between panels in different parts of the page,
such as the panels at the right end of each strip on the opening page
of the Tintin album The Red Sea Sharks,16 a composition that sets the
two pages of a double spread in contrast or harmonious relationship
(the famous central spread of Watchmen), or more distant relations bet
ween different parts of the story. The linear organisation of the panels
is governed by the idea of narrative breakdown, that is, the process of
dividing the narrative into images (in a strip or a larger zone of compo
sition, such as the page), and “most often subordinated to the narrative
ends”, whereas braiding involves a more elaborate integration between
narrative progression and the spatio-topical form of comics (2007, 22).
Another advantage in Groensteen’s multidirectional approach is that it
sheds light on how breakdown and page layout interact and are mutually
Time in Comics 43
informative. Their rapport is an important formal aspect of study, given
the fact that comics can vary greatly in this regard.
The all-inclusiveness of arthrology may become a problem: where
should one draw the line of possibly significant relations between panels
or other units of a comic narrative? However, the problem seems less
serious if we consider that the salience of a given relation must always be
justified, on a case-by-case basis, in text analysis. The point in identifying
any ‘tabular’ or ‘translinear’ panel relations, or relations between larger
segments of the narrative, is precisely to explain how the given relation
is significant, for instance, in terms of temporal information, in contrast
with a multitude of other relations that are not relevant to consider.17
Groensteen’s capacious theoretical model of panel relations can easily
accommodate those forms of temporality in comics that move beyond
linear connections between single panels. These include, for instance,
forms of open-ended temporal logic, or the effect of simultaneity bet
ween different storylines or two acts of telling. Think, for instance, of
the dramatically different pace and tempo between words and images
in the “Paper Doll” passage in Building Stories, where the old land
lady tells about her memories relating to her dislike of dolls, but the
reader simultaneously sees her aging in this passage from a young child
into an old woman. The woman’s descent down the stairs also reflects
the aging process.18 Consider also the effect of the simultaneity in
the panel-within-panel structure in Matti Hagelberg’s imaginary bio
graphy of the Finnish president Urho Kaleva Kekkonen (1900–1986)
(Figure 1.3). The arrangement in this passage allows the reader to either
choose one sequence or alternate between two sequences, the small
panels-within-panels storyline that depicts an interview of the cartoon
ist Hagelberg. The main story portrays Kekkonen, as the President, on a
fishing trip to Iceland where he is swallowed by a great whale and gets
to meet the treacherous Pinocchio in the whale’s belly. In the interview
that progresses in the corners of the main panels, Hagelberg answers
the question about the sports that he would like to practise if he were
elected President of Finland. The two storylines are not only distinct but
also complementary; there are not only two linear sequences in the same
pages but also a thematic connection through space.
Despite the many advantages of this approach, one serious narratolog
ical problem that we encounter in Groensteen’s theory of spatial articula
tion is the argument that mere spatial arrangement of the elements of the
page can somehow in itself create a sense of time. More precisely, Groen
steen’s notion of multiframe (multicadre) refers to the complete spatial
composition in a work of comics, pertaining especially to the arrange
ment of panels on a page, in the sense that this arrangement interacts
with narrativity (or “narrative flux”, 1999, 27) and the sense of time. In
spired by the Belgian philosopher Henri Van Lier, the notion thus presup
poses that spatial structure in comics has an inbuilt temporal meaning.
44 Time in Comics
Figure 1.3 Matti Hagelberg. Kekkonen (2004) © Matti Hagelberg.
We can distinguish between several and not altogether compatible

meanings in the notion of the multiframe. On the one hand, a multiframe
refers to the total spatial arrangement of comics in the sense of a unit
of reading (such as a page, a strip, or any other compositional unit).19
On the other hand, the multiframe refers to an imaginary “contentless
comic” in the meaning of a pure spatial structure of a series of sup
porting frames, or “a comic provisionally reduced to its spatio-topical
parameters” (2007, 24). In this second meaning, therefore, multiframe
is a unit of design, for instance, the gridding of a page into a panel
structure. Van Lier’s perception of the multiframe is similarly flexible,
involving a metaphor for an all-encompassing structural form of comics:
an arrangement of juxtaposed frames that function like empty aircraft
drifting in the white void of the page’s space, calling forth drawn images
and ceaselessly generating their mutations (1988, 5). The multiframe
Time in Comics 45
does not automatically produce a story or lead to a narrative (1999, 124;
2011, 16, 88), 20 but it implies a sense of time and rhythm.
Perceived as a reading protocol, the notion of the multiframe can be
useful in drawing our attention to the way in which the space of the com
position can suggest connections between the panels. Similarly, studies
of layout in comics have often conceived the panel arrangement in terms
of rules for reading and attention.21 Perceived as the totality of the con
tentless frames in a comic, the notion of the multiframe may also allow
us to think of the whole spatial organisation in a given work. However,
beyond the very general notion of the density of the page, which relates to
the number or panels in a given arrangement (Groensteen 2011, 150), it is
not evident how the structural matrix of empty, juxtaposed frames can
represent time through space, or give rhythmic pattern to the narrative.
A series of juxtaposed frames can certainly be used as an “instrument for
converting space into time, into duration” (Groensteen 2013, 138), but
the question remains how that kind of conversion may take place, or how
spatial composition interacts with narrativity. Emphasising the space of
the page as a contentless form, the notion of the multiframe cannot ef
fectively account for the way in which space may convey a sense of time
in picture stories. As our focus lies on narrative comics and narration in
comics, it is necessary to develop our understanding of the narrative con
ventions and means of connectivity in the structure of the composition,
rather than the abstract spatial form of comics.
Layout and Connectivity

Comics can, in many ways, prompt readers to find temporally and nar
ratively meaningful relations between the panels or between a panel,
a strip, and the page. In this respect, Philippe Marion has helpfully
mapped out the general compositional elements and processes that assist
readers in this task (1993, 220–226). These processes include:
• the means of composition and layout

• style
• the narrative content of the images
• the expectation of the reader’s active processing in filling in the gaps
in the story.
Many formal devices of the composition can suggest a sense of conti

nuity between the panels. These include the use of expanding graphic
elements, such as the multiplication or embedding of frames, or chang
ing panel sizes; the contraction, reduction, or suppression of the space
between the panels so that individual panels lose part of their distinc
tion and impermeability; and the overlapping of panels to point out
connections. These techniques thus create the impression of a literally
46 Time in Comics
continuous sequence or ‘stream’ of images in the space of the compo
sition, while they may also undermine the integrity of the panel as a
separate unit of time, space, and reading. Likewise, the page can func
tion as a potential mega-panel that organises the relations of the indivi
dual panels in its space and demands attention to itself as a whole. One
reason why Richard McGuire’s nonlinear story Here (2014) is such a
remarkable achievement is that the work’s composition goes so delibe
rately against these expectations and creates its own logic of spatial con
nectivity. Here, the embedded frames, all identified by a year, depict
one point in space, which is a corner of a room in the twentieth and the
twenty-first century, thus illustrating from this fixed point the passing of
time both towards the future and the past.
The spatial means of connectivity reflect and potentially encourage
the two basic reading conventions that we have discussed in this chap
ter: the reading for the sequence and the global look across the page,
or ‘tabular’ reading. In Benoît Peeters’ much-used and debated typo
logy of mise en page, which describes four common forms of layout in
comics—the conventional, the rhetorical, the productive, and the deco
rative layout style—the most relevant layout types from our narrative
perspective are the so-called conventional and rhetorical forms of organ
isation. Narrative structure dominates in the ‘conventional’ conception
of the regular grid of panels on the page with no variation. In this case,
in order to facilitate storytelling, the spatial form of the page, due to its
regularity, becomes as invisible as possible. In the ‘rhetorical’ use of the
page layout, 22 in turn, the demands of narrative presentation fully dic
tate the dynamic spatial form in the layout. In other words, changes in
the layout reflect the content of the story events. Such changes, typically,
help to distinguish what is narratively salient and important in a passage
or scene. For instance, in Hergé’s Tintin, the panel relations, shapes, and
sizes keep changing according to the evolving story, thus emphasising
narrative action and the situation at hand, or in accordance with the
space and time of the story.
By contrast, the so-called productive and decorative functions of lay
out do not have evident narrative functions. In what Peeters calls ‘pro
ductive’ layout style, as in Winsor McCay’s Little Nemo, the changing
panel sizes and shapes seem to control the story rather than vice versa. 23
Here, the dimensions of the panels and the page have an expressive func
tion, for instance, suggesting an image or an idea in their own right, or
giving shape to the characters. Consequently, the evolving story appears,
at least to some extent, to be an outcome of the spatial arrangement.
We may thus note that the ‘productive’ layout refers to cases where the
relation between segmentivity and narrative, or the zone of composition
and the narrative, are in tension, due to the ostentatious organisation
of the page that seems to dictate the narrative or give rise to narrative
fragments in their own right.
Time in Comics 47
Despite its shortcomings, Peeters’ approach is useful from the nar
ratological point of view as it underscores the way in which the page
layout can support temporal and narrative connections between the
panels. The seemingly least narrative of Peeters’ four general categories
is the ‘decorative’ page layout, in which the page is treated, as much
as possible, as an independent drawing, a design unit, or a concrete
object. Yet, such a decorative panel or page can also have a narrative
function. 24 In his 16-album Philémon series, Fred frequently experi
ments with page layout to undermine the distinction between a single
panel and the temporal progression of the action. One case in point are
the double-spread images that Fred uses to repeat a particular panel,
or a part of a panel, in the ongoing story. Typically, there are three to
four such double spreads per album. These expanded panels, usually
without the verbal content that they may have in the smaller version—
onomatopoeia exempted—can be viewed as painterly re-framings. At
the same time, the expanded panels contribute to the story by giving a
more detailed description of a situation and the world of the story, while
they also expand a particular moment. Moreover, these double spreads
are a means of altering rhythm, creating a sense of momentary stasis in
the story. The original reason for their inclusion, however, seems to have
been purely pragmatic: the need to complete the required f orty-two or
so pages that make a comics album, thus developing the story from the
pages that were first published in the French comics weekly Pilote (the
series started in 1965, the first album came out in 1972). Here again, the
narrative analysis of comics confronts the question of the publication
format, and the requirements of the printing technology, in shaping the
narrative forms.
Beyond layout and the space of the composition, graphic style and the
image content also suggest panel relations and transitions. Typically, the
persistent identity of the graphic trace and style—at least to the extent
that the cartoonist’s style remains consistent—can imply a sense of conti
nuity between the panels. This pertains, for instance, in Fred’s Philémon
series, equally to the coherence of the event and the storyworld. The
change or rupture of style, in turn, can indicate a significant discontinu
ity in this respect. Equally, the repetition or juxtaposition of particular
elements of the story, such as characters and their actions, a continuing
character’s perspective, features of the storyworld or situation, or the
repetition of some other subject matter from panel to panel, can create
narrative connections. Patterns of repetition and juxtaposition, further
more, often rely on the expected relatedness of certain things in our
world. Juxtaposed facial expressions, for instance, can indicate an emo
tional state and sense of involvement; doors suggest that there is an
entrance into another space; falling leaves connote autumn; and so on.
I will discuss the narrative functions of style and continuing characters
and their actions in later chapters.
48 Time in Comics
Marion’s last category of connectivity in comics is, in fact, not a parti
cular device or strategy specific to the medium, but involves the general
expectation that the sense of continuity between panels is established in
the reading process (1993, 220–226). This important point goes almost
without saying: the reader has an active role in negotiating and neutral
ising the inherent fragmentation of the narrative breakdown in comics,
stringing the disparate elements together. Here we, thus, return to the
issue of how to make meaningful connections between different units of
the composition and the story, such as the perception of narrative mean
ing and coherence between the panels. The compositional means of the
page layout also reflect the reader’s choices in reading, for instance, by
emphasising a particular panel, opposition of panels, a sequence, a sense
of depth, the global look, or something else.
Levels of Sequencing
In order to study the logic of connectivity further, we can turn to lit
erary stylistics where narrative sequencing has been conceived in a
more multi-faceted and reader-oriented way. One particularly helpful
point of departure in this respect is Geoffrey Leech and Mick Short’s
approach, in Style in Fiction (2007, 2nd edition), to narrative fiction
that considers three basic principles of narrative organisation on which
writers rely in determining the choice of a particular kind of sequence.
These categories—chronological, psychological, and presentational
sequencing—apply in a general sense, mutatis mutandis, to comics sto
rytelling. The model allows us to consider narrative order, respectively,
from the viewpoint of chronological order, the character’s perspective
and experience, the reader’s response, as well as expectations implied
in the comic, based on the author’s concern for the reader’s interest in
the story. In actual narrative comics, all these dimensions are mutually
inclusive and interactive.
The impression of chronological order follows the presumption that
the passage of time in a narrative imitates chronological real time; that
is, it reflects an order where an event is represented before another
event simply because that event takes place before the other event.
C hronological sequencing, therefore, coincides with the narratological
study of order and anachrony—how the sense of chronology is created,
and how the presentation of the events may digress from it.
Beyond chronological sequencing, we can distinguish two other prin
ciples of sequencing that contribute to a temporal logic between the pan
els. Psychological sequencing can take precedence over the chronological
principle, especially in fiction. In psychological sequencing, the textual
order of presenting the narrative, that is, discourse-time, “reflects the
order in which impressions occur in the mind” (Leech and Short 2007,
190), that is, as they occur in a particular character’s or narrator’s mind
Time in Comics 49
and perception. This involves, perhaps most importantly, the port
rayal of events, and their chronologies and causal connections, viewed
through a sentient agent or an experiencing consciousness. One seminal
technique in this regard is the use of fictional point of view (or inter
nal focalisation) where information about the events and situations of a
story is limited to a particular character’s mind and perception. Another
common method is first-person verbal narration that focusses on the
narrator’s thoughts and emotions, or the use of subjective viewpoints.
In comics, as in literature and cinema, psychological sequencing con
tributes to the order in the arrangement of panels. Sometimes, especially
in stories that focus on a particular character’s experience or psyche, this
principle may become particularly prominent. 25 Such passages can, for
instance, be organised along psychological continuities that the reader
establishes between panels and their groupings and associates with a
certain character. For instance, in Bastien Vivès’s A Taste of Chlorine,
various scenes at the swimming pool are subjectified through the pro
tagonist’s perspective in the images that show him looking at something
or show his field of vision behind his back. By contrast, the various cen
tralised images—masks or kinds of paper dolls—in the double spreads
of Chris Ware’s Building Stories, especially ones including images of the
main characters’ faces with closed eyes, create a psychological frame or
a kind of embedding for the viewing of the surrounding panels by asso
ciation. In Jeffrey Brown’s autobiographical graphic narrative Clumsy
(2002), in turn, the diary-like perspective and style of drawing suggests
that the non-chronological episodic structure of this work, which de
picts a one-year long-distance relationship through a series of significant
situations and scenes, reflects the order in which the impressions may
have occurred in the author’s mind. The emphasis lies on the personal
experience of time, not chronology.
In addition, what is equally significant in terms of psychological sequenc
ing are the potential, but unrealised, alternatives in the evolving events—
things that could happen but are not realised. A character may anticipate
or hope for an event that does not take place, goals may not be attained,
a retrospection of past events may be unreliable, and the narrative may
foreshadow events that do not take place. Such hypothetical, unrealised,
temporal scenarios can have a significant role in the story and its reading.26
Thus, it may be useful, in order to perceive temporality in narrative comics
more fully, to think beyond the question of anachronies and mere sequenc
ing, to the relation between the chronology of the events and the network
of unrealised possibilities that evolves as the narrative progresses.
Finally, the principle of presentational sequencing considers the logic
of narrative sequence from the viewpoint of intelligibility and narrative
tension. We can conceive this principle also as the cartoonist’s or the
storyteller’s concern for the audience’s interest in the projected sequence.
Thus, choices in the sequential organisation of narrative comics are
50 Time in Comics
dictated by the potential of the given organisation to create kinds of audi
ence responses that encourage the reader to keep reading. Leech and Short
formulate the principle as a question: “What is the appropriate order in
which the reader should learn the elements of the fiction?” (2007, 143).27
Thus, presentational sequencing reflects the cartoonist’s choices that are
dictated by the necessity to gradually build a sense of a world, an event,
or a character and, moreover, to do this in a way that maintains the au
dience’s interest in the evolving story. The building of narrative tension
also has an artistic dimension, i.e. the art of holding back information
so that it heightens interest in and curiosity about the story, or creates
effects of surprise and suspense. This is, then, another motivation for the
divergence of story-time and discourse-time: their distance may increase
the reader’s interest in the story. Presentational sequencing is operative
at several levels simultaneously, including the order of the events, their
experiential order, and their psychological significance for the characters,
as well as knowledge about the characters—all perceived through the
author’s and the reader’s shared interest in the evolving story.
Another rationale for distinguishing presentational sequencing from
chronological and psychological levels of sequence is that the structures
of temporal order in narratives cannot be totally cut off from the issue
of knowledge about the fictional world. The manipulation of the chrono
logical order has various effects on our knowledge about the story world,
characterisation, the salience of a particular experience or situation, and
so on. One basic rule of intelligibility is that things that require less back
ground information will come first in the story, that is, that the story
develops “from elements which presuppose the least prior knowledge to
those which presuppose the most” (2007, 143). If this rule is undermined,
as happens in Building Stories, the transgression should serve a particu
lar purpose, since discourse-time is thus foregrounded. The comic may
start in medias res, or important information concerning the story world
is postponed, and this increases narrative tension in the tale.
In actual narrative comics, the three types of sequencing coincide and
create combined effects. The manipulation of chronology can, for in
stance, bring the character’s mental state into better view, or a scene
shift may reflect transformations at all of these levels at once. Finally, it
needs to be stressed that forms of nonlinear temporal order, ambivalent
time frames and unrealistic time are also possible in comics, including
types of circular or contradictory time. 28 The creation of such effects
may require the undermining of all these levels of sequencing and, thus,
the applicability of the narratological model can be noticeably limited.
Duration and Rhythm

In literary narratives, as in comics, textual features or cues cannot
measure the sense of duration, speed, and rhythmic alteration to the
Time in Comics 51
same extent as we can observe anachronies in the order of the narrated
events. This is because impressions of duration and rhythm are depen
dent on the reader’s subjective sense of time in reading. One reader may
spend a lot of time with some scene, while others a lot less. Neverthe
less, we may be able to analyse duration and rhythm in comics in a rela
tive sense by comparing the ratio between the length and complexity of
the representation of an event—in terms of panels, strips, or pages—
and the time span that is covered by that event or situation in the world
of the story. The approximate duration may then be estimated on the
basis of the number of panels (per page, tier, or scene), the number of
pages (per situation, event, or scene), the amount of textual and visual
detail in the images, and the complexity of panel relations and the lay
out in comparison with the length of time passed in the world of the
story, provided that there is sufficient indication about the passage of
time in the story. 29
By ‘duration’ is thus meant the relationship between the time of the
events in the story, i.e. their duration in the time of the story, and the
space given to their representation, and the time of reading that this
space implies. Narrative rhythm or tempo, in turn, can be specified
as the alteration of this ratio, for instance, through changing patterns
(cadence) or speed. Such rhythmic effects are generally more observable
within larger narrative units, such as a scene or between two scenes
that allows us to make observations about significant changes in this
regard. In Ware’s Building Stories, for instance, narrative rhythm is sped
up considerably in a ‘staircase’ sequence of about twenty panels that
shows the protagonist’s landlady age from a young girl to an old woman
about eighty.
In the Philémon series, Fred regularly investigates narrative rhythm
by introducing two simultaneous temporalities and spaces in the zone of
composition so that the panel division on the page reflects, at the same
time, one divided space at one point in time and an event that evolves
in the same space (Figure 1.4). Other techniques for slowing down and
at once complicating the effect of rhythm include the above-mentioned
double-spread wordless images or page-size panels, which draw atten
tion both to the sequence and the composition as a whole. The Philémon
series also frequently features panels that have an ambivalent status as
two-dimensional visual fields, such as panels that flip over or become ob
jects. These instances of ‘metaleptic’30 panel constitute another breach
of the level of reality in the story: the characters relate to the formal ele
ments of comics as objects or as spaces that they can explore at will. In
other words, the spaces of the layout transform into parts of the fictional
world, thus collapsing the border between the space of the composition
and the narrative space.
Duration and rhythm in narrative comics are relational terms also
in the sense that they correlate the relationship between the time of the
52 Time in Comics
Figure 1.4 Fred. Philémon. La mémémoire (1977) © Philémon—tome 11,

La mémémoire DARGAUD by Fred. All rights reserved www.
dargaud.com.
events in the story and the space given to their representation in some
narrative unit, such as a scene, with another unit or the narrative as a
whole. In literary narratives, films, and comics alike, variations of tempo
involve a number of conventional patterns that distinguish, for instance,
a scene from other scenes, or between different modes of narration (dia
logue, thought report, representation of action, and so on).
The manipulation of rhythm is perhaps even more frequent than
anachronies in narratives, regardless of media. In fact, Gérard Genette
has argued in relation to literary narratives that “it is hard to imag
ine the existence of a narrative that would admit of no variation in
speed—and even this banal observation is somewhat important: a nar
rative can do without anachronies, but not without anisochronies, or, if
Time in Comics 53
one prefers (as one probably does), effects of rhythm” (1980, 88, italics
original). Reflecting on the conventions in the history of the modern
novel, Genette then goes on to typify four canonical forms of novelistic
tempo, or what he calls narrative movements. These include the two
extremes of ellipsis (maximal speed) and descriptive pause (maximally
slow rhythm), and their two intermediaries: a scene, most typically in
the form of a conversational scene (dialogue), and which “realizes con
ventionally the equality of time between narrative and story”, and a
summary. Summary, as Genette specifies, typically has a variable and
greatly flexible tempo, which “covers the entire range included between
scene and ellipsis” (1980, 94).
Genette schematises the conventional rhythmic variations, or canon
ical forms of novelistic tempo, with the following formulas, where ST
designates story-time and NT discourse-time (or narrative time), the
sign ∞ meaning either infinitely greater (∞ >) or infinitely less time (< ∞).
0 indicates elision, that is, either that a section of discourse, such as a
descriptive pause, corresponds to no duration in the story, or some part
of story-time is absent from the narrative (ellipsis)31:
pause: NT = n, ST = 0. Thus: NT ∞ > ST

scene: NT = ST
summary: NT < ST
ellipsis: NT = 0, ST = n. Thus, NT < ∞ ST
What is left out from these formulas is the fifth logical option, i.e. that
the effect of rhythm in discourse-time could be slower than in story-time
(NT > ST). Genette admits that this rhythmic formula may be possible,
but only in experimental narration—“the reading of which often seems
to take longer, much longer, than the diegetic time that such scenes are
supposed to be covering” (1980, 95)—or in slow-motion techniques
in film. Later, other narratologists have problematised this exclusion.
B eyond literary experiments with repetition, where the same event or
the statement of that event recurs several times, Seymour Chatman ar
gues that the verbal rendition of a character’s mental events is bound
to be much slower than what has transpired in the character’s mind
(1978, 73). Chatman thus adds the fifth formula, stretch (1978, 72–73),
to the basic variations. For Christine Brooke-Rose, the missing for
mula of NT > ST, or what she calls the ‘slowed down scene’, has an
even broader scope than this. She argues that any narrator’s commen
tary on “gestures or sighs or sinking hearts”, or “thoughts, memories,
emotions and observations” in a dialogue scene, expand the scene to
“a much ‘slowed down scene’” (1981, 315). Therefore, Brooke-Rose
argues, NT = ST can be reserved for pure dialogue only (1981, 315).
Consequently, there should be plenty of room to consider the rhythmic
formula of NT > ST in fiction.
54 Time in Comics
All five basic types of narrative rhythm can be found in comics story
telling. In fact, in visual narratives, such as films and comics, the rhyth
mic formula of stretch seems to be one of the basic rhythmic variations.
In film narratives, a stretch may be realised, beyond slow motion, by a
number of techniques, such as overlapping (or repetitious) editing, types
of camera movement—the camera moves slowly around close-up details
of an image or a scene without action, for instance—and an extended
shot of a static subject. In comics, cartoonists have various graphic and
spatial techniques at their disposal to slow down discourse-time consid
erably. This can be done, for instance, with a descriptive passage that
shows various aspects of the same object or scene in the same instant, a
page layout that breaks down the sense of temporal progress by means of
multidirectional or ambivalent panel relations, or a scene where hardly
anything changes from one panel to the next. One striking example of a
stretch is a page from Scott McCloud’s The Sculptor (2015), where the
protagonist, the sculptor David Smith, is lying on his bed with his lover,
Meg, seen from above, and becomes excruciatingly aware of the passing
of time. Here, a close-up image of David’s face is replaced, in a sequence
of panels, by four images of a clock on the wall in a kind of zoomed-in
effect. The hands of this clock are not moving, but we ‘hear’ it ticking six
times—as shown in the gradually growing and ever bolder letters of an
onomatopoeic ‘tic’ that accompanies the panels, depicting the clock and
again the protagonist’s distressed face. This suggests that the length of
the scene is only a few seconds. In this time, the protagonist, becoming
aware of the clock’s ticking, closes and opens his eyes. The last panel on
this page, a bleed image in the corner of the page, shows David’s horri
fied face with the last ‘tic’ superimposed on his mouth.
To summarise, the techniques for diversifying narrative rhythm involve:
• means of layout:
• changing panel size, shape, or colour
• rhythmic function of the panel frames32
• changing layout style
• creating relations between the panels and the page layout that
undermine the expectation of temporal progress through linear
sequence (many panels form one image, page as a geometrical
shape, stacked panels, etc.)
• rhythm-activating and reinforcing relations between speech and
thought balloons and captions (or words in the images) in and
across the panels
• means of drawing and style:
• change of style
• collapsing the distinction between the spatial composition and
the narrative space and time
• increasing density of visual information
Time in Comics 55
• means of verbal narration:
• changing amount of verbal narration and information
• narrative summary that speeds up narrative time
• distancing verbal narration from visual information
• size of the page and the form of the book.
All narrative comics share the basic expectation that their space can be
perceived in terms of duration. However, depending on the narrative
context, the panel shapes and sizes, sequences, and layout techniques,
or other formal aspects of comics, can suggest a great variety of effects
of temporal duration. To use Thierry Groensteen’s formulation of this
point, there is no automatic correspondence or direct proportionality
between the shape or size of the panel and the duration of the action
presented in the image (1999, 56). Instead of conveying some standard
temporal effect, such as an increase or decrease in speed, the relation
between panel sizes, sequences, and duration is only suggestive of some
temporally organised meaning, or a condition of such meaning. A series
of multiplying panels can create an impression of speed, as happens, for
instance, in the flashback passage in Prosopopus, but the multiplication
of panels may equally well indicate the impact of repetition, or give an
effect of slow motion. Therefore, the nature of the relation between spa
tial form and temporal effect needs to be estimated on a case-by-case
basis. Ultimately, the sense of time in each picture panel, sequence, or
other narrative unit depends on the context, i.e. the local means and
aims of narrative breakdown, the relations between the panels, and the
reader’s active cognitive mapping of these relations.
Duration in a Single Panel

It is another commonplace in comics theory to argue that an individual
panel represents a well-chosen moment in narrative action.33 A sequence
of panels, then, can be said to be composed of a series of such discrete
moments.34 However, the ‘panel is a moment’ thesis is problematic for at
least three reasons. First, not all panels are so much moments as they may
be perceptions, perspectives, thoughts, ideas, or representative scenes of
action. Second, single panels in comics frequently represent duration,
a span of time, instead of a moment. Third, if all moments in comics
are inscribed in a sequence (or other form of grouping), this opens the
question of the interdependence between the supposed moments and,
subsequently, about their lack of distinction as a moment.
Usually, when the ‘panel is a moment’ thesis is made, it is accompa
nied by one exception that is too obvious to dismiss, which is that speech
necessarily introduces a sense of duration in a single panel. In actuality,
however, there are many other exceptions: descriptive panels and pas
sages; sequences that follow a character moving or doing something;
56 Time in Comics
images that describe a character’s emotion, attitude, or thought; or the
first-person narrator’s introspective passages that do not portray charged
moments. Verbal narration in captions can imply a sense of duration for
the visual contents of the image or frame the image in terms of such dura
tion. Captions can specify whether what is seen in the panel has a certain
duration, or they may include insertions, commentary, and description
that clearly lengthen the sense of duration in a scene. Importantly also,
visual information, the organisation of the field of the image, and spe
cific techniques and conventions of drawing can introduce duration in a
single panel. The latter include conventional signs and symbolic tricks,
such as motion lines, and the use of large panels to create a different
sense of duration or, possibly, an (infinitely) extended moment. A panel
can also contain so much visual information and complexity, or so many
distinct instants of action, that this can be comparable to reading a se
quence of panels. Furthermore, the demands of the ‘iconic solidarity’ of
images in a sequence dictate that the unit of time depicted in a panel is
rarely a single instant, but opens to the previous and subsequent panels,
or the surrounding ones, and the temporal frame that they develop.
In her brief application of Genette’s and Brooke-Rose’s model of narra
tive rhythm to single panels, Julia Round has made a thought-provoking
point about duration in individual panels by claiming that comics can
enhance the distinction between a scene and a stretch (NT = ST and
NT > ST) in a single panel. This can happen in two essential ways: by
way of the amount of visual detail in the panel, involving, for instance,
the characters’ posture, bodies, and signs of emotions, or through the
relationship between the image content and verbal narration. More
precisely, Round points out that panel images, when they depict scenes
where story-time and discourse-time seem identical (NT = ST) and also
when they are without any verbal narration (‘pure description’), can in
clude visual and emotive elements that increase narrative time. When
panels that depict a scene feature verbal narration in captions, they must
then be defined as NT > ST. Thus, discourse-time becomes clearly longer
than story-time (2007, 322).
There exist various more elaborate visual conventions and layout tech
niques by which cartoonists can introduce a sense of duration in a single
image, suggest an ambiguous relationship between a moment and du
ration, or undermine the idea of the panel as a distinct unit of time. 35
These techniques comprise, for instance, the use of superimposed images
or parts of the image that show several positions of a person or an ob
ject simultaneously, or the distribution of different phases of the same
movement among several similar characters in the same image. A panel
may also show objects from different temporal strata, superimposed or
collaged in one, for instance, as a reflection of a character’s or a narra
tor’s memory. The repetition of the same figure in the space of one image
is another option. Such multiphase images may, for instance, depict an
Time in Comics 57
evolving action, different phases of a character’s movement, or a whole
event.36 Still another example of multiple perceptions in a single panel
is the use of split panels, including panels such as those in the Philémon
series, where a larger segment of the story, perhaps a whole page, can be
seen both as a sequence of panels and one image that stretches over the
whole page. This technique can to some extent be associated, as Scott
McCloud suggests, with the art history term polyptych, which refers to
a painting divided into sections. With this technique, cartoonists can
deliberately ambiguate the distinction between a moment and duration,
or a single image and a sequence. They may also, vice versa, turn a po
tential sequence into a single panel.
One variation of the complex polyptych panel are ‘splash’ pages and
the use of large panels as spreads. Typically, a splash panel evokes move
ment or a series of actions in static form. Thus, cartoonists have, to
borrow Charles Hatfield’s phrasing, expressed “extended spans of time
in synoptic fashion” (2005, 54). Jack Kirby’s crowded spreads typically
capture explosive moments of action. By contrast, Fred’s page-size panels
or double spreads slow down the narrative rhythm, depicting, thus, an
extended moment within the action or, perhaps rather, action as a kind
of painting that suggests a longer duration. Similarly, the use of ‘bleed’
images, where the image runs off the page, and which are especially
common in manga, can evoke a variety of effects of time, including the
sense of timeless or boundless space, or a much-extended moment.
Not all cartoonists employ these techniques, but they are illustrative of
the inherently dynamic, flexible, and heterogeneous nature of the panel
as an imagined unit of time and duration. The crucial point to make
here is that single panels vary greatly in the way in which they evoke a
sense of duration. At one extreme, panel images suggest no change in the
temporal frame of the depicted event or situation. Other panels, in turn,
depict minute changes from panel to panel, such as single instants of
movement or action. Some panel images may encapsulate dramatically,
narratively, and emotionally charged phases of an event or situation that
point to some before and after of the depicted scene, while other panels
depict long duration or many simultaneities. Ultimately, however, all ef
fects of duration in a single-panel image rely on the reader’s processing
of the given panel’s relation to the other panels around it and their un
derstanding of the evolving narrative action, event, or situation.
The Problem of the Key Frame

One regular effect of rhythmic variation is narrative saliency. By this
I mean that changes in rhythm help to make certain local features of the
narrative, such as a single panel, a strip, a passage, or a scene, contras
tively salient in relation to other units around it. The point has been made
by several narratologists, including David Herman, who argues that
58 Time in Comics
“Functionally speaking, longer or shorter duration can cue readers to
focus on some narrative details as more salient than others” (2002, 215).
A change of rhythm may, therefore, also highlight what narratology calls
the cardinal function of an event, that is, an action or event that is logi
cally essential to the narrative action as a whole.37 Typically, a cardinal
function determines a causal sequence, such as resolving an instability,
uncertainty, or tension in the narrative which have been of interest to the
reader, or opening up new questions, instabilities, and uncertainties. In
addition, rhythmic changes may indicate narrative s alience in relation to
transitional moments and passages. The a cceleration or slowing down
of narrative rhythm can then raise expectations about a significant event
that is approaching.38
The interaction between rhythmic variation and narrative salience
is a highly common feature in comics. Let us think of their relation
in the French-Canadian cartoonist Guy Delisle’s documentary travel
story, Pyongyang: A Journey in North Korea (2003), which describes
the author’s experiences during a 2-month period of overseeing the
production of an animated work at a film studio in the North Korean
capital of Pyongyang. What is of particular interest to us here is how
the notion of key frame, used in filmmaking to refer to drawings of
important frames of a sequence, is treated as part of this parody, and
how the notion of a key frame might be conceived in relation to nar
rative rhythm.
A passage in Pyongyang that highlights for us the relation between
narrative rhythm and salience includes a visual citation taken from
Hugo Pratt’s Corto Maltese story “L’ange à la fenêtre d’orient” (1971),
originally included in a collection entitled Les Celtiques (Celtic Tales,
1980). The image shows the hero, Corto, about to shoot a revolver into
the air. This citation is, in fact, a reframing: Pratt’s original panel is
reframed, and within this new frame, there are two pieces of additional
verbal information, namely the words “Corto Maltese” above the image
and, below it, the reference to comics as the ninth form of art (“Corto
Maltese—Monument of the 9th Art”), a notion that is widely used in
Francophone comics culture and scholarship. Thus, the same panel is a
unit within the sequence of Delisle’s story, a visual citation that invites
us to think of the relation between this travel book and Pratt’s origi
nal story, and particularly in relation to the cartoonists’ task of turning
Pratt’s album into an animation film. Moreover, the citation functions
as a parodic homage, and a kind of monument to the comic book as
an art form.
The narrative context of the embedding draws attention to the way in
which comics heroes, such as Corto Maltese, are adapted in animation
film. The citation is placed within a scene containing conversation, set
in a Pyongyang café, where one of Delisle’s colleagues explains that he
is currently working on a Corto adaptation. In the subsequent panels,
the author-narrator explains that to minimise the production expenses
Time in Comics 59
of the production, animators in Paris drew only the key frames, whereas
assistants in North Korea did the in-medial frames. The idea of a key
frame is further illustrated on the bottom of the page by a series of eight
small images of a simple anthropomorphic figure making hand gestures,
and the accompanying quiz-style question: “Which of these are key
frames?” The answer to the question—the first, the middle, and the last
position—is given on the bottom of the page and can be seen if the page
is turned around (Figure 1.5). At the end of the conversation, Delisle’s
autobiographical character comments disparagingly on the Corto proj
ect, saying “Great, that way kids don’t have to bother reading books.
They’ll just think everything started on the TV, like Tintin”.
What happens is that the double-framing, the caption, and the inter
textual gesture (recalling to the reader’s mind the original Corto Maltese
story), slow down the narrative rhythm. The effect of arresting narra
tive flow is accentuated by the fact that the original Pratt panel depicts
a central moment in the action. To illustrate the point further we can
consider the original panel sequence from which the given picture is
borrowed. In this passage, we see the adventuress Venexiana Stevenson
escaping in an airplane from the hands of Venetian guards and of Corto
himself. The manipulation of a time lapse between the panels, and the
changes in perspective from one panel to another, reveal Corto’s spur-
of-the-moment decision not to shoot at the plane at which he is aiming.
However, this we only learn by reading the whole sequence. What first
appears like a moment of transition turns out to be a decisive moment of
non-action that is essential to the scene as a whole. It is a key situation
without which the narrative would not be the same. Recontextualised in
Delisle’s travel story, however, the Corto panel loses its cardinal function
Figure 1.5 Fred Philémon. La mémémoire (1977) © Philémon - tome 11,

La mémémoire DARGAUD by Fred. All rights reserved www.
dargaud.com
60 Time in Comics
or, perhaps rather, is expanded with additional meanings. The doubly
framed panel stands for a key frame; it is a symbol of key narrative ac
tion in comics, but it also shows how the notion of the key frame is not
similarly relevant in comics as it is in animation film.
In comics storytelling, to show a particular image and to frame it is,
at the same time, to give emphasis to what is shown and call attention to
what is drawn.39 This is one reason why the notion of a key frame be
comes problematic. Every panel and image in comics already represents,
in a sense, a frame that shows and tells the reader narratively salient in
formation (even if there is no actual frame). The panels also do not refer
to a pre-existing extratextual ‘flow’ of visual information and continu
ing event, as in acted films, but create a sense of that flow by referring to
it. This is at least the default expectation.
The question of salience in a single panel is also relative to two func
tions that single panels can have in narrative comics: their painterly func
tion or their narrative function. The distinction, first developed by Pierre
Fresnault-Deruelle and Benoît Peeters, has been given weight by Philippe
Marion, who has argued that the comics panel can be seen as an expres
sive fragment traversed by two contradictory dimensions: the story and
the picture (1993, 212). The picture function encourages the reader to
arrest his or her attention on a single image and isolate it from the narra
tive continuum, while the narrative function prompts the reader to glide
over the image in order to grasp the sense of the continuum.40 The picture
function of the panel, therefore, urges the viewer to spend more time with
the panel. A given panel can, for instance due to the amount of visual de
tail that it contains, the complexity of its contents, the skill of the artwork,
or the sheer size of the image, call attention to itself as a single image, or
a work of visual art that encourages its appreciation as a distinct unit of
design. An extreme case of the picture function would be a maximally
self-sufficient image, or what could be called a ‘memorable panel’, which
may risk jamming narrative responses to the image. Roy Lichtenstein’s fa
mous enlarged and altered comic strip panels from the early 1960s turned
single panels, perfectly cut out from their narrative context, into a work
of art. Thus, the painterly panel draws attention to itself by making the
narrative time of the panel as long as possible, giving the impression, as
much as that is possible, that a single panel can stand on its own.
Frequency
Frequency is the third and last determinant in the relationship between
story-time and discourse-time in Genette’s model of narrative temporal
organisation. Variations in this regard are based, on the one hand, on
how often a particular event of the story is recounted and, on the other
hand, how often a narrative statement concerning an event is repeated
in the story. Thus, Genette distinguishes between four basic types of
Time in Comics 61
relations of frequency that he calls ‘singulative narratives’ (two forms),
repeating narratives, and iterative narratives:
a singulative a: a narrative tells once what happened once;

b singulative b: a narrative tells n times what happened n times;
c repeating: a narrative tells n times what happened once;
d iterative: a narrative tells once (at one time) what happened n times.
The first form of singulative (a) in this scheme involves a ‘natural’ corre
spondence between the narrated events (of the story) and the narrative
statements (of the text). Thereby, something unique and singular happens
once and is told once (1980, 114). By contrast, the more general formula
of singulative (b) is a pattern of repetition where several similar events
are recounted several times (as many times as they occur). In this case,
repetitions of the narrative correspond to the repetitions of the story
(1980, 115) so that Donald Duck is humiliated several times in the same
story, Corto Maltese is repeatedly shown smoking, or Wonder Woman
confronts several villains in a row. Like the first type, singulative (b) also
follows the natural frequency of the events, or the baseline mode of as
sumption, where each event is perceived as unique and only occurs once.
By contrast, repeating and iterative narratives (c and d) comprise at once
more obvious and more elaborate cases of repetition and manipulation
of the relationship between an event and its recounting.
In comics, all cases of singulative, repetitive, and iterative, and their
various combinations, are equally possible. Yet, what is specifically chal
lenging to the analysis of narrative frequency in this medium is that we
need to take into consideration repetition at various levels of represen
tation: the images, the layout, visual style, the words, and their interac
tion. For instance, on one wordless page in Bastien Vivès’s graphic novel
A Taste of Chlorine, the sense of repetition in action is created and rein
forced by at the same time the means of perspective, visual showing, and
the means of layout (Figure 1.6). The nine panels of the page, organised
in three tiers on a regular grid pattern, show the swimming protagonist
turning at the same end of the pool three times. The perspective of the
panels always remains the same, thus emphasising the effect of repeti
tion, i.e. the swimmer’s repeated movements in the corner of the pool.
Visual details of his movement, strength, and speed further suggest that
the swimmer’s technique keeps improving and, further, that time passes
between the three depicted lapses. The layout has an important function
in suggesting the sense of repetition here since similar small panels in a
regular grid are not used elsewhere in the story. In terms of Genette’s
model, the case represents a blend of singulative (b), where the narrative
tells n times what happened n times, and an iterative, where the showing
implies that the action has in fact happened many more times than what
is shown.
62 Time in Comics
Figure 1.6 G uy Delisle. Pyongyang: A Journey in North Korea © 2006

L’Association/Guy Delisle; Courtesy of Drawn & Quarterly.
A caption or dialogue can give the accompanying image an iterative

meaning, indicating that what is seen in the panel has, in fact, happened
many times. In turn, the repetition of a whole panel content, that is,
a series of two or more panels where nothing changes, may generate
various temporal, spatial, and dramatic effects, such as the impression of
dramatically slowed down duration, as happens in some of Régis Franc’s
works, or in the Oupabo experiments with ‘iconic reiteration’.41 Simi
larly, the use of superimposed or partly overlapping images can suggest
the iteration of the recounted event or the act of recounting. Different
effects result again from repeated scenes if they are closely associated
with a particular character’s perception. Thus, repetition in the image
content becomes a product or reflection of the character’s mind.
Time in Comics 63
What is especially useful in this model for the analysis of narrative
comics is that it allows us to think of the various ways in which repeti
tion (cases b–d), can be used to create particular effects in the course of
the story. Repetition may suggest, for instance, a range of temporal im
pressions, such as a sense of circular time, underline the interrelatedness
between different events, situations, or experiences, or undermine the
idea of an authentic single event. Repetition is also a means of emphasis
and narrative salience: it may help to signal the importance of some
thing, be that an idea, emotion, phrase, word, experience, behaviour,
memory, action, or an event. The repetition of the same panel may thus
amplify what is shown—making it more significant from the character’s
or the reader’s perspective—or turn the image into a mental one, such
as a memory image. Furthermore, the recounting of the same event may
provide narrative action with an informative frame or background, or
reveal what is usual (since repeated) in a character’s life, for instance.
Repetition is also another means by which storytellers can create parti
cular effects of rhythm.
Think, for instance, of the structures of repetition in a passage called
“A Feeling” in Building Stories when the main character is shown lying
in her bed in various positions and at different times of her life, but
apparently in the same room, in the space of a page. She is unable to
sleep. The panels of this scene feature the protagonist with two different
cats, a man, a teddy bear, but mostly alone, with different hairdo, body
shape, younger and older face, pregnant and not pregnant, and differ
ent clothes, bed, and furniture. One event thus opens onto a pattern of
repetition of similar moments and situation, some panels describing the
woman at the present moment in the story, while others illustrate her
memories of the same place and situation. At the same time, repetition in
Building Stories creates pronouncedly polychronic effects, in the sense
that David Herman defines polychronic structure in narratives (2002,
212–3), that is, having a fuzzy or indeterminate ordering of events.42
Such polychrony is related to the inexactness in a particular panel’s tem
poral ordering, or their partial and multiple ordering in a repeated struc
ture. For instance, repetition suggests multiple temporal orders when the
protagonist is shown in three different phases in the process of going to
bed in one panel (in a passage entitled “Repetition”) or in four different
positions (and times) in her old room at her mother’s home.
It must be noted, however, that not all effects of repetition in comics
have only a temporal nature. Beyond the manipulation of the distinction
between story-time and discourse-time, visual repetition is a crucial ele
ment in terms of narrative coherence, continuity, and connectivity. Typi
cally, this involves the repetition of characters, scenes, and situations, or
some other image content, such as background, so as to move the story
forward and give a sense of a world.
64 Time in Comics
Conclusion: Time and the Reading Conventions
By way of conclusion, we should identify those features in comics and
their reading that pose the most important and interesting challenges to
the narratological analysis of time in this medium. Perhaps the most ob
vious feature in this regard is the discontinuous spatial form of comics.
A comprehensive take on temporality in comics needs to consider the re
lationship between a single panel, the panel sequence, and the page layout
(or other forms of composition), and how comics exploit the ratio between
all these compositional units to convey a sense of time. Another key fea
ture in this regard, the possibility of seeing the whole strip, page, or a dou
ble spread at once, raises the question of the interplay between sequential
and non-sequential reading. A global look at the whole page has various
potential, but not always unambiguous, consequences for the experience
of time in comics. The possibility of looking at the strip or the page as a
whole, as a composition and a design unit, may contradict, to some extent
at least, the linear following of the story. This is not entirely different from
readers of literary fiction skipping to the end of the page, passage, or the
whole book to know what will happen. Yet, differently from the sequen
tial order of words on a page, gazing at the space of the story can function
as a form of anticipation in itself, giving the reader some notion, especially
in what may be seen in the images, of what is expected to happen.
Still another challenge that we have to come to terms with in apply
ing the story-time and discourse-time distinction to comics is the multi
modal form of expression and communication in comics. This involves
the integration of words and images, and their combined forms, in se
quence and into a whole; the binding of spatial, visual, and linguistic
information to make sense of the gaps of the story; and the translation
of certain conventional signs, symbols, or visuals in comics, such as car
icature, gestural language, or kinds of lettering, as markers of emotion
and thought. The issue of dual input of words and graphics, and their
combined visual–textual forms, is crucial for understanding the repre
sentation of time in most comics.
Thus, to summarise, the conventions of reading comics that pertain to
the representation of time include the following:
• The perception of meaningful relations between panels, in two di

rections before and after a panel, in a sequence, and possibly also
across the space of the composition.
• The processing of the relationship between panels and layout, in
particular when layout style invites a more global look.
• The processing of mutually reinforcing indications of layout, style,
verbal narration, and information, and the narrative content of the
images.
Time in Comics 65
• The processing of the effects of bimodal reading and the dual input
of words and graphics: one modality can help to fill the gaps of
temporal information created by the other modality, but words and
images also expand each other’s meanings, set in a meaningful con
trast, or words can slow down the narration.
• The attention to synchronic reading of, or global look at, the whole
of the composition, typically the page or the double spread in the
longer formats. Readers can always choose to look at the compo
sition as a whole. At the same time, comics vary greatly in their
capacity to invite a global look.
• Expectations concerning the time of reading related to the genre and
the publication format.
The techniques of sequencing and page layout precondition these conven

tions. Therefore, on the one hand, comics evoke a sense of time through
a sequentiality that pertains to the relationships between the panels and
their narrative breakdown. On the other hand, comics can imply forms
of temporality in the overall spatial arrangement of panels on the page,
encouraging a synchronic look at the whole of the composition. Both
aspects of temporality are intricately intertwined in the composition,
but they may sometimes exert independent effects of time. That is a
challenge to the narratological approach, but we can also respond to it.
We can separate these compositional arrangements on a heuristic basis,
keeping in mind that the various forms of the layout, and in particular
the panels and their frames, are potentially relevant both in terms of
space and time, and as forms of organisation (or design) and attention.
The varying possibilities of introducing a sense of duration in a single
panel point to the fact that the tension between linear and synchronic
or tabular reading can take place within a single panel as well; also,
contrasts between two or more panels of varying size, shape, or visual
complexity can create that effect.
At the same time, it must be granted that the idea of global look or
tabular reading may not be that relevant for all kinds of comics, or its
not similarly relevant for all forms of layout, especially in respect to
traditional newspaper strips, comic strip booklets, or comics with a sim
ple grid-like organisation. In these cases, reading is not simply linear
either since regression, i.e. returning to the previous panel, seems always
an important aspect of reading, or it may also happen, for instance, in
Ware’s comic strip booklets in Building Stories, that two parallel strips
create the effect of two simultaneous sequences. Yet, such comics do
not invite, let alone demand, a global look. In these cases, readers can
certainly always choose to look at the whole strip or page at first, but a
distinction can be made between such a glance and a global look that is
anticipated and invited by the layout and, in a sense, built into it.
66 Time in Comics
We should, however, resist a too easy opposition between the linear
and the tabular (or synchronic) reading.43 One good reason for not re
ducing their relation to a simple opposition is that the notion of tabular
reading perhaps covers too much ground. In its current use in comics
theory, the notion has a kind of all-encompassing quality with regard to
various forms of non-sequential composition and reading. One ambigu
ity in the notion is that, on the one hand, it refers to specific techniques
of nonlinear and/or non-sequential panel arrangement on the page
while, on the other hand, it may also indicate a manner of reading com
ics in general, including also those comics that strictly follow a linear
arrangement of panels in a sequence. It needs to be stressed that a ‘tabu
lar’ reading is not a technique of composition in itself, but a convention
of reading comics that the compositional choices in the work, such as a
nonlinear grouping of panels, may encourage. Still another ambiguity in
this notion is that it can refer to the strip, the page, or the double spread,
or other zone of composition, as a unit of reading and a unit of design
(involving a picture, a painting, or a surface) at the same time. This may
not be totally avoided since forms of design, composition, and publica
tion reflect and shape conventions of reading in comics, but narrative
theory and analysis should at least be aware of that duality.
The global look, to which Genette refers in passing in his Narrative
Discourse, may involve an appreciation of the design unit of the page or
other aspect of composition perceived as a simultaneous whole. Thus,
the global look is not only ambivalent in relation to narrative structure
and understanding, but it may describe various kinds of readerly desires,
impulses, or intentions, as well as different types of attention, such as
the desire to contemplate the page layout as a unit of visual design, or
a kind of quick look at the whole of the composition before reading it.
Furthermore, there is a potential ambivalence in this concept between
a global look, which can refer to different aspects of nonlinear looking
and eye movement—a global look at the whole of the strip or the page,
or a circular, freely-moving gaze around the zone of composition—and
a type of reading that seeks to anticipate and, perhaps, to some extent
also understand the narrative meanings of the work, in particular con
cerning the visual content of the composition. This contradiction can be
avoided, however, insofar as we may be able to conceive the two acti
vities of looking and reading as aspects of the same cognitive process in
understanding comics.
Finally, it is important to point out that this chapter has been limited in
its focus in that I have not considered the generic frames and publication
formats that also influence, at least within some broad parameters, the
representation of time in comics. One question that could have been given
room here is how reading conventions, including for instance the expected
time of the reading, are related to a particular genre and publication for
mat. Humorous newspaper comic strips, for instance, are usually meant
Time in Comics 67
to be consumed speedily. They suggest a quick narrative rhythm, based on
three or four panels that include a punchline. Generally speaking, popular
genre fictions, such as mass-produced superhero comics, children’s com
ics, humour strips in album series, such as Titeuf or Le Chat in France,
or manga such as Dragon Ball or Naruto, invite extensive rather than
intensive reading. By contrast, more intensive and in-depth reading is ex
pected of narrative genres, where it is common to use more text, visual
detail, more complicated panel relations, and create more complex char
acters, situations, and plots.44 The reading speed of some types of action
manga may approach that of a flipbook, while the reading of complicat
edly self-reflexive graphic novels, such as Chris Ware’s Building Stories, is
expected to take time. However, the relation between genre and the time
of reading is also quite relative: the genre-related expectations are not
always met, or the readers may challenge them. For instance, some popu
lar comics, such as Edgar P. Jacobs’s Blake and Mortimer series, include
a considerable amount of text and visual detail, and invite the reader to
spend more time in reading. Besides, many popular genres and series have
dedicated fan cultures that encourage rereading and intensive reading.
Thereby, a series that may have been designed for extensive reading can
become something that is frequently reread or read more intensively by
the same reader in another situation or a different kind of publication.
The reproductions of popular strips and magazines in the book format
encourage rereading while the change of the format may in itself invite the
reader to spend more time with the work.
Notes
1 See, for instance, Scott McCloud’s influential claim that “In learning to read
comics we all learned to perceive time spatially, for in the world of comics,
time and space are one and the same” (1993, 100).
2 Ware has emphasised that by allowing readers to shape the order of their
reading, he wanted to explore the way in which “stories and memories are
available from all sides and moments in our memories, and not really part
of a continuum” http://www.tcj.com/i-hoped-that-the-book-would-just-be-
fun-a-brief-interview-with-chris-ware/.
3 This is partly due to the privileged position that the notion of event has
enjoyed as the basic constituent of narratives. See, for instance, Prince
(2003, 58) or Abbott (2008, 15).
4 The distinction between fabula and syuzhet was first conceived in theoret
ical terms by the Russian formalists. See, for instance, Boris Tomashevsky
(1965, 66–67).
5 See also Herman (2002, 214–215).
6 Surprisingly, however, Schüwer does not employ the narratological frame
work of temporal organisation in any systematic way. Surdiacourt contends
that Schüwer’s understanding of the story-discourse distinction is, in fact,
based “on a traditional (read: structuralist) conception in which the story
provides the raw material for a particular narrative representation (or dis
course) and thus logically precedes this representation” (2012).
68 Time in Comics
7 The narrative present is a sense of the present moment of the story events. It
is more or less distant from the time of its telling, that is, the temporal frame
of the narrator’s act of narration.
8 Genette derives his terms from Greek—analepsis (flashback), prolepsis
(anticipation), and achrony—but these formulations are not widely used.
9 By a scene I mean an event or a situation that is unified by space, time, and
the characters that are present. A scene can also be defined as the conven
tional equivalence between story-time and discourse-time in the given nar
rative segment. See Prince (2003, 85–86).
10 I find Neil Cohn’s remarks in this respect to be highly relevant. He points out
that “the gap cannot be filled unless it has already been passed over, mak
ing closure an additive inference that occurs at panels, not between them”
(2010, 135).
11 For Robert C. Harvey, a breakdown is the process of dividing the narra
tive into successive panels in a narratively effective way (1994, 8, 14–15;
2005, 21).
12 Beaty also argues that the six categories could be boiled down to four basic
cases: transitions that involve either a change in the subject of action or
attention (subject-to-subject), a shift time (moment-to-moment), a complete
change of scene (scene-to-scene), and the non-sequitur (1999, 69–71).
13 Dale Jacobs’s notion that the gutter can be employed for “virtually any
rhetorical end” is in line with this argument (2007, 504–505).
14 Groensteen’s ‘arthrology’ is a neologism from the Greek arthon (articulation).
15 Postema has formulated the two-way reading of panels elegantly: “In look
ing at narrative panels in comics it is necessary to read back and forth in
gathering the signification of comics panels. They do not represent stages or
moments, but rather a continuum of possibility that remains fluid even after
one has read the panel. In reading panels one’s eye weaves continuously back
and forth, as the meaning of one panel retroactively resignifies what was
seen in the previous panel” (2013, 75).
16 See Baetens’ convincing reading of this passage (1989, 93).
17 Neil Cohn’s question “Where do the relations stop?” (2014b, 68) in this
regard is relevant. However, the idea of “unrestrained transitions” between
every possible panel in a document, and that would “overload the working
memory of the human mind”, remains mainly theoretical. The semantic rele
vance of all panel relations must be established on a case-by-case basis. See
also Miodrag’s defence of Groensteen (2013, 127–129, 134).
18 See Sattler on the disjunction between text and image, their interpolation
of narrative and episodic memories, and their joint encoding of experiential
memory in this passage (2010, 209–212). I would also like to thank Leena
Romu for her comment on the woman’s downward movement.
19 The multiframe does not have any predetermined limits, but can take vari
ous shapes such as a comic strip, a half-page format (2–3 strips together),
a page, double spread, album, and a whole book (Groensteen 2007, 30–31).
20 Compare with the claim, which is perhaps revelatory of a certain anti-
narrative impulse at the heart of Groensteen’s theory, that the principle of
iconic solidarity does not have an inherent narrative purpose (2011, 17).
21 As in Chavanne’s seminal work (2010, 18–23). See also Jesse Cohn’s claim
that the analysis of page layout can reveal how stories elicit the reader’s de
sire to know (2009, 56).
22 Chavanne emphasises that even if the zone of composition frequently corre
sponds to the page, this is not always the case, and thus the notion of “mise
en page” (the page layout) could be rejected (2010, 13). The point is relevant,
Time in Comics 69
but the emphasis on the strip as a basic unit of design has the unfortunate
result of marginalising the significance of the page.
23 Peeters stresses that these four categories should not be perceived as mutu
ally exclusive or in a historical sense (1991, 36; 1998, 52; 2007, n5; and Jesse
Cohn 2007). Groensteen has complemented this model by the distinctions of
regularity or irregularity, and discreet (discrète) or ostentatious layout style
(1999, 112–118). The English translation mistakes the French ‘discrete’ as
‘discrete’ when it should be ‘discreet’, as in understated (2007, 95–101). See
also Baetens’ and Frey’s discussion of these models (2015, 108–120) on the
basis of the degree of correspondence between the page layout and the panel
content (2015, 130–133) and Bateman’s (2016) further important expan
sion. The relevance of this model for Japanese manga has been questioned
by Rommens (2000).
24 The choice of the term ‘decorative’ is unfortunate on account of its norma
tive connotations, i.e. that some element of the work is only of secondary
value. Jesse Cohn points out, logically, that if the distinction between rheto
rical and decorative layout “is to be of any use, it would seem, there must be
some case in which we can be certain that the visual architecture of the page
is not relevant to the meaning of the narrative, that it is purely ornamental.
If images narrate, however, then no image can be purely innocent of narra
tive meaning” (2007, italics original).
25 Benjamin Widiss has emphasised the meaning of those passages in Ware’s
comics that, instead of (or as much as) engaging the reader fully in a se
quential temporal progress, solicit seeing and “a process of association and
reflection only partially dictated by narrative prompts” (2013, 89).
26 See also Gerald Prince’s notion of the disnarrated (1992, 28–38).
27 Neil Cohn’s cognitive linguistics model perceives sequentiality on this level,
pertaining to the readers’ comprehension of narrative meaning and concep
tual information in a given unit of attention, according to the preference
rules that guide readers. For Cohn, these rules dictate that: (1) grouped areas
are preferred to non-grouped areas, (2) smooth paths are preferred to broken
paths, (3) one should not jump over units, and (4) one should not leave ‘gaps’
in reading (2013, 9; 2014a, 6).
28 See Singer (2012, 57) for some examples of these categories in comics.
29 In McCloud’s theory of time in comics, these two levels are confused. See
also Cortsen (2012, 41–45) and Miodrag (2013, 118). However, it is also
worth keeping in mind, as Genette points out, that diegetic time in in small
segments of narrative fiction “is almost never indicated (or inferable) with
the precision that would be necessary” for a detailed analysis of rhythmic
effects (1980, 88).
30 Metalepsis is understood here in the Genettean sense of a paradoxical trans
gression of the boundaries between narrative levels (1980, 234–237).
31 Genette, in fact, refers to discourse-time as conventional pseudo-time and,
thus, privileges story-time as a kind of real-time (1980, 33–35, 94). The
emphasis is problematic, given the fact that fiction, generally speaking, does
not refer to actual events and situations, but these events and situations are
created by referring to them in the telling.
32 See also Groensteen’s discussion of the rhythmic function of panel frames
(1999, 55–56) and the three basic ways of reinforcing rhythm in comics that
conform to a regular layout: repetition, alternation, and progressivity.
33 See, for instance, Bongco, who argues that “the panel […]graphically and di-
egetically unifies image and text in the comics: it forms a graphic unit which
represents one moment, one instant of action in the narrative” (2000, 58).
70 Time in Comics
34 For Douglas Wolk, for instance, comics are “made up of a series of discrete
moments” (2007, 125).
35 See, for instance, Eisner on the use of a full-page frame as a unit of time
(1985, 63); Baetens and Lefèvre on the superposition of different phases of
movement or the showing of different phases of the same movement by seve
ral similar figures (1993, 51–52); Hatfield on multiple images in a single
panel or parsing simultaneous actions into successive frames (2005, 52–58);
and Postema on the multiphase picture and panels that include visual signs
from different time frames (2013, 19–20).
36 See, for instance, Postema (2007, 498–499) and Cohn (2007, 39–43; 2010,
131–132), on the polymorphic panel.
37 Cardinal functions are distinct from ‘catalyses’ or ‘catalysts’ that are not
essential to the narrative action and the causal-chronological coherence of a
narrative. See Prince (2003, 11).
38 See, for instance, Edward Branigan’s reading of the first sixteen panels in
Nick Fury, Agent of S.H.I.E.L.D. story entitled “Who is Scorpio?” (by Jim
Steranko, June 1968) (1992, 81–82).
39 Groensteen has listed, among the ways in which cartoonists can accentuate
a particular panel or a sequence, both formal features of the composition
and the panel content, such as the panel’s placement on the page, its form
and size, how much of the subject is visible in the image, or a change in
colouring, style, and the quantity of information (verbal or visual) (2011,
170–171). Also, frame types can emphasise narratively salient moments.
40 More precisely, Marion writes, “I would call the narrative function that
which guides the viewer in gliding past the frame, and the panel function
that which, on the contrary, is responsible for eliciting a fixation on the
image by isolating—through various means—a continuum” (Joyce Goggin’s
translation in Mikkonen 2010, 82). See also Hatfield (2005, 48).
41 For instance, Lewis Trondheim and Jean-Christophe Menu’s album Moins
d’un quart de seconde pour vivre (1991), where only eight different images
are used in 100 strips that contain four panels each. See also Groensteen
on repetition as a specific form of emphasis within a regular layout (2011,
160–163).
42 See Herman’s definition (2002, 213): “[polychrony] includes both the more
or less ‘radical’ types of inexactness in coding, as well as both the multiple
and the partial ordering of the events”.
43 See also Baetens on the insufficiency of this division in relation to the totality
of the different operations that comics can employ, including duration in sin
gle panels, or the possibility of tabular or translinear connections between
the panels (1998, 75–76; Baetens and Frey 2015, 106).
4 4 Eric S. Rabkin, for instance, has pointed out how different degrees of infor
mation intensity and representational immediacy in panels affect the speed
with which we read a given frame and how this is often relative to the genre
of comics (2009, 37).
Part II
Graphic Showing and Style

2 Narration as Showing
Narration by visually showing something that evolves from one panel to

another is such a common feature in comics that the importance of this
mode of narration may remain unrecognized, unlike, perhaps, an un
usual layout, impressive spreads, or richly detailed panels that demand
greater attention from the reader. The depiction of characters in actions
and situations in a world that they inhabit easily encourages a narra
tive response and evokes a story-like scenario in the mind of the reader.
For instance, the introduction to Alan Moore and David Lloyd’s V for
Vendetta (1989), involving twelve panels and set in 1997, first shows
four panels portraying a dystopian post-war London and then juxtaposes
two situations with two anonymous characters who are both dressing up
and making up, or masking up, in front of a mirror. E vidently, the two
characters, a masked man and a young woman, are listening to the same
radio broadcast. The characters are shown in alternating panels; first,
the woman preparing herself is presented in two panels, and then a man,
shown from behind, appears in the next panel, after which the woman
appears again in the proceeding panel. This is followed by two smaller
panels of the man’s hand and his mask. Finally, each character is given
one panel each, the last panel including the chapter indication: “Chapter
One: The Villain”. The eight panels of this scene thus follow the pattern
of 2-1-1-2-1-1. The verbal component of the passage does not suggest
a narrative but, rather, gives some information on their world and a
shared temporal frame for the characters’ actions. This verbal ‘track’ is
a radio news broadcast, reporting the weather, police surveillance, and
raids on the homes of supposed members of a terrorist ring, the young
Queen’s public appearance, and other matters. The broadcast also un
derscores the duty of every man “in this country to seize the initiative and
make Britain great again”. The continuation of the overheard radio pro
gramme accentuates the impression of simultaneity created by the two
juxtaposed situations. The images, however, carry the main narrative
load. The two characters, evidently residents of this gloomy world, are
shown to us in similar situations that suggest a s tory-like scenario, thus
encouraging a narrative response. They are both preparing for an event
that evening—perhaps a meeting, a date, a ball, or a masquerade—but
74 Graphic Showing and Style
we do not yet know what it may be or even if it is the same event. What
increases the narrativity in this passage is the mystery of how the two
situations are connected in the world of the story.
Comics usually tell a story in both words and images, but sometimes
they predominantly or exclusively rely on images. The fundamental
point that will be made in this chapter is that images in comics can visu
ally articulate the narrative by showing certain things in certain ways.
By showing I refer here to the use of pictures to communicate narrative
meaning in front of the reader’s eyes. In particular, I will focus on the
presentation of characters, their bodies, behaviours, and actions that
move the story forward, or at least evoke a story-like scenario.
To date, the narrative function of showing has remained implicit or
cursorily articulated in comics theory. One reason for this may be that
much theory on comics, seeking to define the formal essence or the basic
elements of the medium, such as sequentiality, the spatial arrangement
of the panels and the page, or the gutter, has deemed any considerations
of visual material, for instance, relating to characters and their worlds,
a matter of changing content. At the same time, while most comics con
centrate on characters and their actions—showing human-like, but not
necessarily human, characters in particular situations and events—this
is not the only way to tell a story in this medium. Comics can take a nar
rative form merely by means of verbal narration. A further reason for the
lack of theoretical interest in the issue of showing, as a narrative issue,
is that visual showing in comics cannot usually be reduced to a mere
narrative function. When an image in a comic shows a person, thing,
scene, situation, or world, this may be done in terms of describing, pre
senting, or demonstrating instead of moving the story forward. Think,
for instance, of the impressive surfacing of Captain Nemo’s octopus-like
Nautilus submarine in the harbour of Cairo, or the blood-smeared
Mr Edward Hyde bursting into a Parisian flat in the first instalment
of Alan Moore and Kevin O’Neill’s The League of Extraordinary
Gentlemen (1999), and how these page-size panels not only introduce
us to important elements of the storyworld, but also present something
spectacular in their own right. These panels function like a painting,
showing a world, people, and things in terms of description, even if they
are part of a story. Visual description, as in describing a character or
milieu, is an important dimension in most narrative comics, both for
its own sake as an aid and as an elementary part of narration. What
would, for instance, be left of the appeal of Léo’s popular science fiction
saga of planetary romance—the three series of Aldebaran (1994–1998),
Betelgeuse (2000–2005), and Antares (2007–2013)—without the visual
imagination of earth-like but alien environments and strange creatures?
These creatures are at the heart of the mystery of Léo’s intergalactic
adventures, prompting us to ask what these worlds and their habitants
are like, what they contain, and how they function. Many of the larger
Narration as Showing 75
images in these series create a world by describing it, thus presenting a
world and its fascinating creatures for viewing.
Yet another reason for the lack of interest in the category of showing
derives from literary narratology: classical models of narratology per
ceive showing in literary narratives as a problematic notion of m imetic
presentation that equates some techniques of narration with life-likeness.
For this reason, in literary narratology, the distinction between showing
and telling has been frequently rejected, or modified by a gradient scale
of varying modes of presentation and degrees of narrative mediation
with regard to narratorial intrusion. However, in theatre, film, and com
ics, showing can have a more literal sense, with no relation to a narrator,
and may be more easily separated from the presumed effect of verisi
militude. In these contexts, showing can be equated with the evolving
performance, the moving image, or the sense of witnessing an event in
front of one’s eyes. This does not mean that showing can somehow rep
resent the world directly or that the question of mimesis is irrelevant in
this context; rather, mimesis becomes relevant in a different way, in and
through the visual content of the images. In comics, the viewing and
observing of drawn images are never ‘direct’; it is always mediated at
several simultaneous levels, including the level of graphic style. With the
exception of photorealistic images, narrative drawings in comics bear
some traces of the maker, the act of drawing, stylistic choices, and ge
neric features—they do not simply represent something. The viewing of
images in comics is also obviously indirect to the extent that pictures in
teract with the words that surround them, are superimposed onto them,
or are placed within their space, and thus what is seen in the images
becomes filtered and interpreted by words and verbal statements.
Before rehabilitating the concept of showing in the study of narrative
comics, however, we need to specify what we mean by showing as a form
of narration. First, it is important to emphasise that the narrative mean
ing of pictorial content is only an aspect, sometimes more pronounced
and sometimes less, of images in comics, and not the only reason why im
ages and visual material are presented for viewing. Second, the category
of showing provisionally sets aside the important issue of the interaction
between verbal and visual modes of presenting in comics storytelling.
Therefore, the risk of a kind of myopic perspective in this regard is real.
At the same time, it should also be emphasised that there are evident
gains to be made from considering showing as a distinct narrative mode
in comics. For instance, it may allow us to highlight the ways in which
comics can tell a story by visual means alone. Furthermore, it offers us
another means of accommodating the rich history of wordless comics,
sometimes known as silent or pantomime comics, within the narrato
logy of comics.1 In addition, it enables us to consider the narrative signi
ficance of those passages in comics where images lead the narration even
if they also interact with words.
Showing and Telling
From the writings of Henry James and the critic Percy Lubbock until the
present day, an established distinction has existed between ‘showing’ and
‘telling’ in literary criticism even if there has always been much disagree
ment about the meaning of showing in literature. In literary fiction, the
notion may be held to mean the impression that the story event is shown
to the audience as directly as possible, for instance by means of dialogue
or presenting a character’s thoughts to be read word for word. More
precisely, what can be called showing in literary narratives involves the
use of narrative techniques that contribute to the mimetic illusion upon
reading the text: the direct representation of the characters’ speech; the
aspiration of objectivity in scenic presentation; the narrator’s spatial,
temporal, and epistemic distance to the events told; and the avoidance of
techniques of telling, such as summary and condensation. 2 For instance,
a dialogue scene can create a greater illusion of mimesis (imitation of
real speech) than a narrative of events or a reported discourse of a char
acter’s thoughts. In the telling mode, in contrast, the narrative evokes the
impression that the events are told by some agent, such as the narrator
or the storyteller.
The distinction between showing and telling has its roots in the po
etics of Plato and Aristotle, and especially in Plato’s Republic, where
narration can take the two basic forms of diegesis, which Plato calls the
plain narrative, and where the poet (or storyteller) speaks in his own
name, and mimesis, where the narrator imitates the characters’ speech,
which is thus quoted directly. Additionally, storytelling can be based
on a combination of these two forms.3 Contemporary narrative theory
has often employed, modified, or undermined this distinction. For in
stance, in The Rhetoric of Fiction (1961), Wayne C. Booth attacked the
modernist emphasis on showing as the most advanced form of narration
in literary art, arguing convincingly that the use of a particular narra
tive technique does not amount to more life-likeness or higher aesthetic
value. In other words, all forms of narration have a man-made q uality—
they are shamefully artificial (Booth 1983, 20). Gérard Genette’s influ
ential argument in this respect was that all forms of mimesis in literary
narratives, in contrast to dramatic representation, are illusory: no writ
ten narrative “can ‘show’ or ‘imitate’ the story it tells. All this can do
is to tell it in a manner which is detailed, precise, ‘alive’, and in that
way give more or less the illusion of mimesis” (Genette 1980, 163–164).
Language creates, Genette further argued, meaning without imitating
(except, perhaps, for onomatopoeic words) and, consequently, mimesis
in words can only be mimesis of words (1980, 164). Thus, the distinction
between diegesis and mimesis in literary fiction becomes an issue of de-
grees of diegetic presentation, and it may no longer be conceived in terms
of the polar opposites of mimetic showing and diegetic telling.
A benefit of the narratological debate on the value of the category
of showing is that it has illuminated the varying effects of imitation in
different modes of presentation and narration. However, what can be
meant by ‘showing’ in other narrative media and arts, such as comics,
especially if they include performative and visual forms of presentation,
is clearly different. While in written narratives (without illustrations),
showing is metaphorical or involves types of direct discourse, showing
in comics plays a central role both literally—a world is created by show
ing it—and metaphorically, i.e. the drawings stand for a world, charac
ters, and their actions.
This may be stating the obvious, but a major appeal of comics is their
visual style, how they are drawn, and what they show. Visual showing
is a crucial component of the dynamic of the narrative in this art—it
cannot be reduced to issues of panel relations, techniques of sequencing,
the layout, or perspectival choices. Generally speaking, the images in
comics, cartoons, and animated films are drawn in a way that maxi
mizes their narrative meaning and function. What they show, then, is a
form of telling.4 In John Leech’s single-image cartoon “Substance and
Shadow”, published in Punch on 15 July 1843, and often regarded as the
first ‘cartoon’ in the modern sense of the word—i.e. a satirical image in
graphic form—the postures and expressions of the people seen at the art
exhibition depicted in the cartoon create a marked contrast with the af
fluent people and idyllic scenes portrayed in the paintings around the
walls (Figure 2.1). The highly caricatured visitors, the waifs and paupers
in dirty rags, the many infirm and disabled, bemusedly inspect the exhi
bition at Westminster Hall, consisting of ‘cartoons’, in the sense of large-
scale sketches for historical murals, 5 that are competing for selection by
the government to decorate the new Houses of Parliament. Besides the
title of the drawing, which refers to the ironic meanings of substance and
shadow in the situation, Leech realizes much of his satire—a critique of
government, and possibly also of artists who ignore social issues such as
poverty—through graphic showing, by juxtaposing the two worlds of
the visitors and the paintings.
But are images in cartoons and comics more interesting when they
show (or tell by showing) rather than just describe? The question does
not admit a straightforward answer. There is, as has already been under
lined, so much variety in comics, and even inside a single comic in this
respect. Nonetheless, a twofold argument could be made on the basis
of this variety: comics can not only tell by showing, but they can also
show things independently of the narrative function. Typically, the ra
tio between narration and description (or exposition) continually shifts
within one story. It is worth noting that when Groensteen claims that
for narrative drawing, showing and telling are one and the same thing,
he immediately relativizes that claim, conceding that there are moments
in comics when the narrative pressure is, in fact, released (2007, 163).
Figure 2.1 John Leech. “Substance and Shadow” (1843).
Such a release is evident, for instance, in individual panels, sequences, or

larger units of composition that seek to be more pictorial, poetic, intro
spective, or sensual than narrative.
Showing in Comics
The focus on visual showing may allow us to better grasp the narrative
mechanisms at work in comics, and, in particular, the narrative function
that images play in this art form. However, before moving on to a more
systematic discussion of the concept, we should reflect on the concept
of ‘image’, more particularly on the pictorial specificity of images in
comics. Most conventionally, an image in comics is expected to be a
product of graphic design, that is, a drawn image consisting of lines and
strokes that are applied to a two-dimensional surface.6 The comics and
film scholar Philippe Marrion defines the graphic art of comics, or what
he calls graphisme, accordingly, as the “configuring” use of lines, traces,
colours, figures, and signs on a surface, with the intention of expression
and representation (Marion 1993, 3). There are obviously some import
ant exceptions to this expectation, and the idea that comics always in
volve drawn, graphic images is clearly too all encompassing. Comics
may, for instance, incorporate blank colour panels, mere text panels,
collage images or photographs, and they may even be wholly based on
photographs (so-called ‘fumetti’) or computer-generated images. In the
past two decades, computer technology, in particular, has transformed
the making and conception of images in comics, and it can be difficult
or outright impossible to distinguish between what is drawn and what is
computer-designed in today’s comics.
My focus here is on graphic images that invite a narrative response
by showing an event or a situation that involves characters. Whether
they are actually drawn or not is beside the point. For our next example
of showing, let us consider H.M. Bateman’s famous “One Note Man”
(14 December 1921) wordless comic from the artist’s “Man Who” strip
series,7 which was published in the magazines Tatler and Punch in the
1920s and the early 1930s. “One Note Man” is a story four full pages in
length that relates a day in the life of a musician who plays a clarinet-like
wind instrument in a large symphony orchestra. On the first page of the
story, we see the main events in what seems to be an ordinary day for
the musician: he wakes up, takes a bath, has breakfast, smokes his pipe,
practises playing, takes a bus to the concert hall, and chats with some
of his colleagues before the concert. At the concert, his task is to play
a single note, which seems to put him under much strain (Figure 2.2).
We then see the man sneaking out of the rest of the concert, going back
home, having dinner, smoking his pipe again, and getting ready for bed.
The humour of this story has various sources, one of which is the
idea, also conveyed by the title, that the protagonist’s musicianship con
sists of playing just one note. This does not seem like much in com
parison, for instance, to the violinists, whose labours are also shown
in close-up images. Another source of humour is the manipulation of
narrative rhythm during the story, especially in the concert scene when
the man plays his note. The various small events in the man’s life, from
shaving to smoking his pipe, which are shown before and after the con
cert, are given roughly equal space in the story, usually one panel per
event. The panels show us that the man spends most of his waking
hours in non-musical activities that neatly divide his day into a familiar
and steady pattern. During the concert, however, the narrative rhythm
Figure 2.2 H.M. Bateman. “The One-Note Man” (1921).

slows down considerably. The concert scene represents over half the
panels in the story, while the playing of the note, and the immediately
preceding and following moments, take almost one page. Therefore, the
carefully timed moment of terrible intensity is humorously contrasted,
through means of narrative rhythm, with a life that otherwise seems
extremely monotonous.
Perhaps the most important source of humour in this story is none
theless the visual content of the images. First of all, the presentation
of the man’s continuous action from panel to panel, and the pattern
of following that is thus established, allow us to think of these strips
as an evolving story. Second, the depiction of the man’s expressions,
postures, and body language, and the inferences that the reader can
make about the changes therewith, move the story forward. The way
in which the man’s expressions are drawn suggests that he is rather
content with himself and his life. For most of his day, his facial expres
sion changes very little, beyond the need to yawn in the morning and
at night. The man looks unwaveringly content and relaxed, and his
posture is also confident as he walks along the street. The only striking
exception to his demeanour of calm is when he actually plays the note.
This is clearly reflected in his face, posture, and body language, which
are shown in a series of four panels: (1) the man reaches for his instru
ment with a concentrated look on his face; then, (2) bends towards his
music note stand, and very carefully follows the notes on the page;
next, (3) he is shown with both hands placed on the instrument, and
his eyes opened wide; and finally, (4) he is depicted with his eyebrows
acutely raised and his mouth open, ready to play the instrument. After
the note, the man slumps down dramatically, almost collapsing with
his chair, while holding his chest in exhaustion—he has given his all for
that crucial note (Figure 2.3).
To understand the story and its humour, the reader must be able to
attend to changes in what is shown in the image sequence. The title of
the story does not add much to the reader’s understanding; rather, it re
peats what can already be appreciated by following the visual content
of the images.
Figure 2.3 H.M. Bateman. “The One-Note Man” (1921).

How, then, should the narrative showing of visual content be integrated
into comics narratology? At least two approaches are readily available.
On the one hand, to better understand the medium-specific qualities
of showing in comics, we can revisit the theory of showing, or mon-
stration, in film narratology, and modify the concept for our purposes.
On the other hand, we can analyse the synthetic role of the continuing
character as a means of connectivity and coherence. The musician in
“The One-Note Man”, who can be seen in each of the fi fty-eight un
framed panels of this comic strip, builds a sense of continuity and coher
ence between the panels by allowing space, time, and action to continue
throughout the sequence. Next, we will turn our attention to the concept
of showing in film studies. The issue of continuing characters will be
tackled in the next chapter.
The Concept of Monstration

The theory of showing in cinematic narratives is based on André
Gaudreault’s ground-breaking narratological work from the 1980s and
1990s,8 where he redefines Plato’s concept of mimesis as monstration.9
For Gaudreault, a film narrative is the product of a linking, often com
plex and layered, of the two basic modes of narrative communication:
narration and monstration (Gaudreault 2009, 7). In this context, nar
ration does not refer to delegated narrators in film, who tell the story in
voice over—voice-over narrators are, in fact, only a limited source of
narration in films, or a kind of “sub-narration” (Gaudreault and Jost
1999, 53–54); rather, it denotes techniques of editing that permit indivi
dual shots to be articulated together as a narrative (Gaudreault 1987, 31).
Narration, therefore, means sequencing, a process in which individual
frames are turned into narrative units, i.e. shots. Editing also enables
temporal modulation in films, that is, a change of temporal frame allow
ing, for instance, the unfolding events on the screen to be conceived as
occurring in the past. In contrast, Gaudreault defines monstration as a
mode of narrative communication that is limited to a sense of the present
in the evolving events (1987, 31). More precisely, monstration is
a mode of communicating a story, which consists of showing char

acters (in English, monstrance) who act out rather than tell the
vicissitudes to which they are subjected. Monstration could thus be
used to replace the term “representation,” which is too specific, too
compromised, and far too polysemic.
Gaudreault (2009, 69)
Theatrical monstration is therefore the product, at least in part, of the

characters in dramatic action, that is, the performance of characters act
ing out the narrative (Gaudreault 2009, 37, 40). Monstration, moreover,
includes two aspects of communication: mimetic and non-mimetic die
gesis. Mimetic diegesis is a form of staged narrative where the characters
depict the action through their physical actions and behaviour, while
a non-mimetic mode of narrative communication may involve various
forms of talking and quoting, such as conversation—in other words,
“all cases in which words, speech, and verbal discourse are used within
a staged performance to tell a story” (Gaudreault 2009, 71).
The different modes of verbal and visual narration are assigned to
theoretically distinct agents in Gaudreault’s model, which he terms nar
rator and monstrator (show-er of images). These agents are responsible
for verbal and visual narration, respectively, and are set in a hierarchical
model of narrative levels where they are both subordinate to a narratorial
agent at a higher level of cinematic narration, which Gaudreault calls the
‘mega-narrator’, or ‘grand image-maker’. This is the plural agent who is
responsible for modulating all manifestations of “theatrical language”
in film: mise en scène, set design, lighting, and acting (Gaudreault
2009, 72), and coordinating the various sources of expression, including
images, sounds, words, written materials, and music. However, accord
ing to Gaudreault, in collective productions such as theatre or cinema,
the narrator is a more fleeting theoretical construction with less well-
defined borders than is the case in literary narratives.
Monstration in Comics
In comics studies, Philippe Marion has applied and modified Gaudreault’s
concept of monstration in his theory of “graphic enunciation”, published
in the reworked version of his doctoral dissertation as Traces en cases
(1993), which has been one of the key texts in French-language comics
theory since its publication. Sharing Gaudreault’s interest in narrative
agents, Marion uses ‘enunciation’ to refer to the act and circumstances
of producing an utterance in comics—with utterance meaning here, for
instance, a strip, a page of a comic, or a narrative.10 For Marion, the very
heart of enunciation in comics is their graphic material, the graphic trace
(Marion 1993, 9–10). It is this graphic quality of images, so M arion
claims, that always makes the drawings in comics, to some extent,
opaque signs, at least in comparison to cinematic and photographic real
ism. Thus, by letting us see the trace of the graphic act—the signs of the
graphic performance—showing in comics never has the same figurative
transparency or the same transitivity as in cinema (Marion 1993, 36).
Moreover, unlike a photograph, the graphic images in comics constitute
the material that they present.
Generally speaking, enunciation in this sense is distinguishable
whenever readers become aware of any signs of the writer or cartoon
ist’s subjectivity. This may also occur whenever readers find themselves
confronted by comic-book conventions as such, from the observation
that “this is comics”, acknowledged by self-reflexive uses of basic
comic-book devices, such as layout or techniques of sequencing, to “I am
reading a comic book (a strip, a graphic novel)”, such as when a comic
self-consciously reflects on its materiality or the publication format.11
In principle, any noticeable stylistic changes may increase the reader’s
awareness of enunciation. This can be a pronounced graphic line and
trace, unusual colours (choice of colour palette, contrast, tone, shade,
and so on) or chiaroscuro, types of foregrounding and backgrounding
in the image field, or a number of other choices that pertain to the visual
register and graphic design, and which may highlight the subjectivity of
the source of narration.
However, to focus on narration and monstration as instances of
enunciation makes it harder to see the narrative function of showing
in its own right. In fact, what really seems to matter in Gaudreault and
Marion’s notions of showing is the identification of different kinds of
narrative agents, to whom they ascribe responsibility for different as
pects of presentation and who allow them to develop the theory of a
higher-level narrator. The approach is in line with the Genettean model
of narratology, where narrative mediation is equated with the narrator
concept, i.e. all narratives supposedly have narrators, but it distances us
from the question of narration as showing.
We will return to the question of narrative agency in comics in more
detail in a later chapter. Here, it is sufficient to note that the category
of showing, understood in the sense of characters acting out the story
through a comic’s visual content, can be distinguished for analytical pur
poses in comics. More precisely, showing can thus be understood to refer
to certain aspects in the narrative drawings of comics that are presented
for viewing, specifically characters and their actions, behaviours, and
situations in a particular world, and that inspire a narrative response
from the reader.
The way that comics narrate by showing is exemplified in a section in
Bastien Vivès’s graphic novel, A Taste of Chlorine, that relies on visual
showing alone. The lack of words in the story is partly motivated by the
protagonist’s solitary visits to a swimming pool and the depiction of the
act of swimming itself. At the same time, showing through images is also
very effectively used in this work to convey mental and physical states.
One particularly striking wordless scene, which lasts for eighteen pages,
takes place towards the end of the story, when the protagonist decides
to swim the whole length of the swimming pool underwater. As he ap
proaches the end of the pool, he suddenly notices above him a woman
with whom he had become acquainted earlier, a woman whom, to his
great disappointment, he had not seen for a long while. The man seem
ingly tries to grasp the woman’s leg, but either fails to do so or pulls back
since he realizes that she is actually someone else. Whatever the case, at
the same time, he gulps, sinks back, and runs out of air. The man starts
to swim furiously upwards, at times evidently astonished and terrified
by the distance. At the end of the scene, he finally comes to the surface,
gasping for breath.
The various mental states that the protagonist goes through in this
scene are depicted visually, by showing his changing facial expressions
and postures. Thus, following the salient details of how the scene is
acted out, we are able to follow the character’s determination when div
ing, excitement in seeing the woman above him, his struggle in trying
to catch her, surprise at being suddenly out of air and still so far from
the surface, desperation in trying to reach surface, and total exhaustion
when he hangs on the edge of the pool at the end of the scene. There,
resting and looking towards the reader, he then seems to see something,
but what that might be is not shown. Thus, the image of the gaze is used
for a specific rhetorical effect: to stress the importance of the moment,
and the event, and to increase the sense of the story’s open-endedness.
There is obviously some latitude in our verbal interpretation of this
scene and the character’s actions, in particular relating to the man’s
emotional state. The precise meaning of the facial expressions and
bodily postures portrayed in the scene may be hard to pinpoint. Perhaps
the man is not so excited by the woman or the sense of challenge as he
is simply curious and daring. Or, possibly, he is more disappointed with
himself than surprised by not being able to catch her, fearful of what
may happen instead of being desperate. Nevertheless, the exact terms
that we choose to describe his mental state in this scene are to some ex
tent irrelevant. What matters is the necessity of understanding the acting
out of the scene and making narrative sense of the many visual cues for
his overall mental state.
The concept of showing allows us to discuss the tremendous variety in
comics with regard to the degree in which they tell a story by presenting
characters in situations in narrative drawings. The differences in this re
gard relate not only to the ratio between narration by showing and tell
ing by words, but also to the extent to which non-visual objects, such as
propositions and states of mind, may be expressed visually, through such
means as facial expressions. Generally speaking, such differences reflect
both the choices and styles of individual cartoonists and authors and
also generic expectations, particular traditions of the art form, and the
reader’s capacity to detect mental states from external signs. Typically,
action comics and mainstream manga rely more on the image alone to
convey narrative information.
In more text-oriented comics, the use of strategically chosen wordless
panels or sequences is a common technique for achieving particular nar
rative effects. Wordless narrative drawings may, for instance, effectively
alter narrative rhythm (a wordless panel being an indication of eventless
time or the passage of time, for instance), accentuate the visual aspects
of an event or a place, prompt the reader to look for the salient visual
details or create a particular dramatic effect. Such functions are real
ized, for example, in the strategically placed wordless and textless panels
in Raymond Briggs’s graphic story of his parents’ married life, Ethel &
Ernest. A True Story (1998), where dialogue carries the main narrative
load. Among the wordless elements of this work is a page that contains
a picture of Ethel and Ernest’s wedding on which is partly superimposed
another picture of the couple sitting on “the lovers’ seat” in Fairlight
Glen, Hastings in 1930. The juxtaposition of the two images shows that
the couple had got married and travelled, possibly on their honeymoon,
to this site on the south coast of England. Thus, set in relation to the
honeymoon picture, the ‘wedding’ photograph functions as an index
of an event and an easily recognizable scenario that does not have to be
related by words.
While the wedding picture condenses time into one image, two other
wordless panels in this story accentuate the visual and physical aspects
of the represented experience. These include a panel which shows Ethel
struggling with a large white sheet that has been drying on a clothes
line in the garden, and another panel that shows a doodlebug (the V-1
flying bomb) flying over the cartoonist and his father, who have flung
themselves to the ground in a field. In both of these panels, the object
(the sheet, the bomb) that is presented for viewing, as well as the sound
of the bomb, extends over the frames of the panel. The breaking of the
frame heightens the object’s size and speed, and the uncontrollability
of its movement, both from the perspective of the characters and the
reader. On the one hand, the relationship between the moving object
and the characters in the image, and in Ethel’s case also her facial ex
pression and posture, suggests a strong emotional state. Ethel appears
to be as much marvelling at the sheet as she is struggling to keep it in
place—the sheet’s movement is also associated with the white birds fly
ing in the distance—while the speed of the doodlebug, and the shadow
that it casts over the cartoonist and his father on the ground, increases
the sense of fear in this scene. On the other hand, the objects are thus
presented for viewing as something to be marvelled at and, potentially,
as something that allows the reader to share the characters’ sensations
and emotions. Finally, the last wordless scene of the book, comprising
four panels, relates Ernest’s lonely death, contrasted with the couple’s
long life together, in which dialogue had played such a vital role. Here,
the lack of words is a feature of the storyworld and Ernest’s lonely situ
ation, not just an element of dramatic effect or a means of description.
Although the wordless panels in the otherwise dialogue-centred story
involve important events in the narrative, they are not uniform in pur
pose. While they involve scenes in which the main characters act out
“the vicissitudes to which they are subjected” (Gaudreault 2009, 69),
they also serve a descriptive function and invite the reader to share the
character’s visual experience.
Graphiation and Graphic Style
Marion argues that the notion of monstration alone is insufficient to
describe the medium-specific features of graphic showing in comics.
This is because monstration in comics, so he claims, lacks the figurative
transparency or transitivity12 that it has in cinema (1993, 36). Conse
quently, Marion developed the concept of graphiation, to describe the
kind of enunciation that is typical of comics. Graphiation pertains to
the graphic and drawn qualities of comics (lines, traces, and graphic
design, used in both images and lettering) when these elements call at
tention to themselves. The notion of graphiation, therefore, focusses on
those elements in comics that are not simply narrated or shown, but
are only traced, or that have a traced, graphic quality that takes pre
cedence over other qualities, such as the content that they present for
viewing or narrative function. Graphiation may concern both drawn
images and words, in particular when their graphic traces (empreinte
graphique) draw attention to themselves as markers of the cartoonist’s
subjectivity.13 Therefore, by emphasising the drawn and written quality
of graphic lines, the concept creates a unified perspective for viewing
both text and drawing.
The focus of the concept of graphiation, then, lies on graphic style
and, in particular, a specific stylistic effect or use: the autoreferential
function of graphic design, where the graphic trace points to itself and,
thus, functions as a marker of subjectivity, in particular, of the cartoon
ist’s subjective style. This effect may be conceived simultaneously as a
kind of artistic expression and a form of readerly attention where the
graphic material draws attention to itself instead of moving the story
forward. In other words, the notion casts light on the fact that in analys
ing narrative comics, we may be able to differentiate not only between
what is shown in the image and how that something is shown, but also
between the graphic act of showing and drawn traces that do more than
just narrate and show or do not yet (or no longer) narrate and show.
The distinction between graphic showing and graphiation, while use
ful in highlighting the fact that not all visual content in images shows
or tells something, or that graphic styles have different degrees of
autoreferentiality, nevertheless, raises a terminological and conceptual
problem concerning whether the concept of graphiation can do some
thing different or better than the notion of graphic style. While the con
cept of graphiation remains popular in French-language comics theory
(but less in actual analysis and research), it also has been convincingly
argued that the scope of this concept is unclear, in particular regarding
the extent to which it overlaps with the notion of style.14 Here, I choose
not to integrate the concept of graphiation into the narratological ap
proach to comics, the reason being that it is far from clear that we need
a new term, let alone the construction of another narrative agent which
would be responsible for the act of graphiation and would then require
narratological attention of its own. As I hope will become evident in
the course of this book, the notion of graphic style provides us with
sufficient means for making useful distinctions between styles ranging
from the maximally ostentatious, or autoreferential, where subjective
expression may be highlighted, perhaps even at the expense of the ob
jects shown in the images, to styles that aspire to maximal transparency
and objectivity.
Conclusion
By concentrating on the narrative function of the visual content of im
ages, I have demonstrated above why the concept of showing is useful
for the narratology of comics. The concept allows us to better conceive
and emphasise the visual aspects of narration in comics, i.e. the way in
which comics rely on visual means of storytelling by showing charac
ters engaged in a situation, behaviour, or action in some world. If what
Genette calls narrative tense looks at the relationship between story
time and discourse time, voice concerns the question of ‘who speaks?’
(i.e. ‘who is the text’s narrative voice?’) and mood the question of ‘who
sees?’ (or, ‘where is the centre or the focus of perception?’), and then the
category of showing invites a two-sided question: ‘what is shown and
how?’ We can also reformulate this question from the reader’s perspec
tive and in relation to narrative comics in particular as ‘what is shown
in the image that inspires a narrative response?’ and ‘how does what is
shown in narrative drawings make us relate one panel to another?’15
These questions are not just about image content—‘what is this image
about?’—but pertain to the narrative function of that content.
The issue of graphic showing, again, raises the question of medium
specificity. With regard to the visibility and materiality of the graphic
trace, there is perhaps no crucial difference between comics, cartoons,
and animated films, except that in an animated film the trace is per
ceived as moving. In general, comics, cartoons, and animation privilege
maximally narrative drawings that feature expressive physical gestures
and easily identifiable features of physiognomy, action, and situation.
The drawn quality of the image modifies the way in which something,
such as a character’s body, expression, or engagement in situation or
world, can be shown. Nevertheless, showing in comics also has some
medium-specific elements. Such elements include, perhaps most impor
tantly, the narrative function of the sequence and the page layout. Thus,
the frame is both a structural and expressive device—a feature that is
also common in single-image cartoons—which can, in comics, also exert
influence over the narrative process by separating the panels and setting
a pattern for their reading and viewing. Another distinguishing factor in
comics in this respect is the way in which the field of vision relates to the
potential visual field outside the picture frame. While comics regularly
reveal, in retrospect, that something important has been left out of the
previous frames, the off-screen space seems to play a much more import
ant and consistent role in animation and cinema than in comics. This
may be because in comics the image is expected to focus on what is most
salient in the evolving event, situation, or a place, and that any references
to contingent but excluded fields of vision need to be clearly motivated.16
Notes
1 For the history of wordless comics see, for instance, Groensteen (1997,
1998), Gravett (2013, 34–53), and Kunzle (2001).
2 For a more extended discussion of these techniques, see Klauk and Köppe
(2014).
3 See the third book of Republic (3.392c–398b).
4 See also Groensteen, who has defined the laws of narrative drawing in com
ics as those of anthropocentrism (the privileging of the character as an agent
of action), synecdochic simplification (the leaving out of everything that
is not necessary for intelligibility), typification, expressivity (maximal ex
pressivity in the characters’ faces and bodies), and rhetorical convergence,
i.e. the narrative image, by all means of composition, colour, and framing,
obeys the imperative of optimal legibility (1999, 190–191; 2007, 162).
5 Before the popularity of satirical cartoons in the printed press, the term
‘cartoon’, or cartone in Italian, had meant a finished preliminary sketch on
a large piece of cardboard.
6 Baetens and Frey argue that “the fact that the story is less told or shown than
drawn is what defines the difference between comics and graphic novels and
storytelling in other media” (2015, 165). I do not think that such a distinc
tion is useful, just because, as Baetens and Frey also write, “(l)ines display a
story world in which the act of drawing cannot be separated from the drawn
result” (ibid.). Moreover, the distinction between showing and drawing does
not suggest a very sound basis for distinguishing between comics, cartoons,
and animation.
7 It is widely known that the idea of the Albert Hall sequence in Alfred
H itchcock’s original version of The Man Who Knew Too Much/ was derived
from this cartoon.
8 Gaudreault draws not only on the Genettean model, but also on Albert
Laffay’s earlier “protonarratological” narrative theory of cinema outlined
in Logique du cinema (1964).
9 Gaudreault’s (2009) basic argument is that diegesis (the poet or the narrator
speaks in his or her own voice, recounting of the events) and mimesis (theat
rical form of presentation, imitation of an action) are not opposite categories
in Plato and Aristotle since mimesis can be seen as a form of diegesis in Plato
and diegesis as a form of mimesis in Aristotle.
10 In the linguist Emile Benveniste’s definition, enunciation is the act of produc
ing an utterance (énoncé) or mobilizing a language for the speaker’s own use
(1970, 12, “un acte individual d’utilisation”).
11 I borrow Gaudreault and Jost’s phrasing of cinematic enunciation (1999, 49).
12 Transitivity refers here to the way in which the contents of an image refer to
a direct object—in contradistinction to a reflexive relation (in graphiation),
where the material of the image refers to itself or, in other words, the direct
object is the graphic line itself.
13 See also Baetens (2001, 150). Dürrenmatt (2013, 160) compares graphiation
to the poetic function of language since it obliges to view the graphic gesture
in itself, including the visual aspects of the alphabet.
14 See, for instance, Groensteen (2011, 92), Dürrenmatt (2013, 159–167).
Baetens claims that Marion’s distinction between monstration and graphia-
tion is convincing, but is critical of these notions, nevertheless, with regard
to what he sees as Marion’s monolithic understanding and the biographical
trappings of the graphiateur concept (1998, 39–40; 2001, 152–3).
15 Verstraten formulates a similar question about narration through shots in film:
“who or what is being shown and how are they being shown?” (2009, 56).
16 The issue of “hors-champ” (the field of vision outside the panel frame) has
not been much theorised in the field. Baetens and Lefèvre, however, have
coined the term péri-champ and discuss the diverse ways in which a comics
panel can refer to the virtual visual field outside its frame, and activate it
(1993, 26–36). See also Groensteen (2007, 119).
3 Character as a Means
of Narrative Continuity
The continuing presence of a character or a group of characters in a

sequence of images, acting out a situation, or participating in an event,
is possibly the most conventional feature of narrative comics. The abil
ity to follow an easily identifiable character, such as Tintin, Astro Boy,
Tank Girl, or “One Note Man”, in evolving action from panel to panel,
maintains a sense of continuity and coherence in a narrative sequence.
In fact, the continuing character is such a common element in comics
that it has sometimes been perceived as a defining characteristic of the
medium.1 This, however, is taking the argument too far: While charac
ters are undoubtedly central to narration in comics, comics are able to
tell a story without the presence of one entity whose situation, action,
or experience endures for the whole length of the work. Cartoonists,
similar to literary authors and filmmakers, often shift attention away
from one character to another, perhaps even in the course of a single
scene. Sometimes juxtaposed or contiguous images are more relevant
for storytelling than continuing characters, and it is always possible
that a story establishes a scene or describes a milieu instead of a charac
ter in action and, in fact, a narrative can be present in a comic without
a character being shown at all (see chapter on characterisation). For
instance, a focus on the inner workings of someone’s mind, i.e. the
thoughts and emotions of a character or narrator, can make a narra
tive. In more experimental comics, frustrating the reader’s ability to
follow a visible character may also be the desired effect. That said,
however, for a narrative comic to have no continuing entity that is re
peated or which the readers may follow, presents a real challenge to
the comic’s sense of causal continuity and narrative coherence. Here,
rather than being seen as a guarantor of the quality of a narrative or an
attribute of a good narrative, cohesion simply refers to causal cohesion
in a sequence of panels. 2
Consequently, one might say that continuing characters are not a re
quirement for narrative comics; rather, they are a central convention that
increases narrativity in any comic. In terms of the narratology of comics,
the issue of the continuing character brings into play various key con
siderations and concepts. On the one hand, the depiction of a character
Character as a Means of Narrative Continuity 91
in a sequence of images gives the reader access to the story through a
clear unit of attention that can be followed. The continuing character
can thus be conceived as an aspect of focalisation, or more precisely, of
the focalised (the object of the perception). On the other hand, a con
tinuing character allows the reader to gradually construct a person-like
entity engaged in some action or situation and have a sense of the story
content: The narrative is about a particular character or group of char
acters. Narrative events, and the experiences that stories highlight, usu
ally revolve around people, or human-like characters, and their actions,
perceptions, and experiences. Furthermore, the presentation of the men
tal states, thoughts, feelings, and experiences of sentient beings, which
many narratologists regard as key characteristics of narratives regardless
of the medium, suggests the presence of some agent who acts, feels, and
thinks and whose actions or thoughts and perceptions can be followed.
Therefore, characters are simultaneously salient features of narrative
comics at various levels, as person-like agents, as a means of narra
tive transmission, and as units of attention that move the story forward
and allow readers to follow it. In narrative comics, salience in this sense
involves the question of how something is shown, including the means
of narrative drawing, layout, framing, and the juxtaposition of panels,
stylistic choices, and the interplay between visual and verbal narration.
However, what is considered salient cannot be merely reduced to the for
mal features of narrative comics. Narrative salience is also an expectation
that readers have of the way in which stories are constructed and told and
the outcome of their active interpretation. One expectation in this regard
is that characters are allocated different roles and significance in fiction,
i.e. there are main and minor characters. Another significant expecta
tion is that the depiction of person-like agency and goal-oriented action
will enable the reader to become imaginatively engaged with the narra
tive. This point is commonly made in narrative studies of film. The film
scholar Murray Smith, for instance, claims that our ‘entry into’ the nar
rative structure of fiction films is mediated by character (1995, 17–18).
We will return to Smith’s argument at the end of this chapter. Similarly,
many literary narratologists hold that the representation of an experien
tial agent is a minimal requirement of narrativity and our primary access
to the narrative. The mere depiction of some character’s action can say
much about their intentions, thoughts or emotions, and experience.
Here, I consider a character present throughout a continuing situa
tion, event, or action as a basic tool for building narrative continuity and
coherence in comics. The point in thus focusing on and isolating the ques
tion of the synthetic3 role of the character, i.e. their continuity-building
function, from other considerations pertaining to characters, such as
focalisation, characterisation (characters’ person-like qualities), or the
representation of speech and thought, is to better cover the visual and
multimodal means of connectivity employed in comics, and thus further
develop the issues of connectivity and showing discussed in the previous
chapters. This will also raise the issue of the relationship between visual
and narrative saliency,4 which can not only overlap but can also be dis
tinguished from each other in terms of what is central in the image (the
content that is shown) on the one hand, and what is important in the im
age in terms of the narrative as a whole on the other. Verbal narration, in
captions and dialogue, can by itself select for the reader those characters,
entities, or objects that are to be tracked and followed, for instance by
creating a sense of a character’s continuing consciousness or a continu
ing conversation. Verbal narration may also specify how the situation
or the protagonist’s actions and movements should be understood, em
phasise the salience of something, or point out, especially in first-person
narration, that the narrator’s inner experience is more important than
what he or she may be perceiving (as shown in the images). In the next
sections, however, the focus will be on the visual showing of continuing
characters and the reader’s spatial attachment to them.
Match on Action in Comics

The repetition of the same character in a panel sequence creates a vi
sual bridge between the images. This basic convention can be clearly ob
served, for instance, at the beginning of stories. In the first instalment of
Saga (2012), by Brian K. Vaughan and Fiona Staples, the cover image al
ready establishes the two protagonists, Alana and Marko, as important
figures in the story. The first panels of the first scene, which depict the
birth of the couple’s daughter Hazel, then confirm that the story is about
these two characters. Both Alana and Marko are shown in close-up
images that focus on their emotional states and intimate relationship.
The first panel is an extreme close-up of Alana, who is clearly suffering,
while the second panel, an establishing image, shows her lying on a table
with someone between her legs, helping in what is evidently a childbirth
scene. The two characters’ emotional engagement with one another is
then portrayed by an image and reverse image sequence where we first
see the horned man Marko looking tenderly at Alana and commending
the winged woman for her beauty, and then see Alana, suffering labour
pains, looking less fondly back at him and responding with a sarcastic
comment (Figure 3.1).
The visual bridges between the panels of this scene are based on recur
ring characters and the sense of their continuing action. Match on action
is the most common continuity editing technique in filmic narratives and
extremely common in comics as a means of panel transition: to match
different shots with continuing action, the basic purpose of which is
“to allow space, time, and action to continue in a smooth flow over a
series of shots” (Bordwell and Thompson 1986, 231). In this technique,
one shot cuts to another shot portraying the action of the subject in the
Figure 3.1 Brian K. Vaughan and Fiona Staples. Saga © 2014 Brian K. Vaughan &
Fiona Staples.
first shot. Thus, the character’s (or characters’) activity creates a visual
bridge between the gaps—that is, the shots—and conveys a sense of con
tinuity in the scene. The effectiveness of this technique relies on its abil
ity to suggest a simultaneous sense of temporal and spatial coherence.
This is the reason why David Bordwell and Kristin Thompson also argue
that match on action is a particularly powerful continuity editing strat
egy: “If an action carries across the cut, the space and time are assumed
to be continuous from shot to shot” (2008, 250). We could further add
that match on action as it typically involves a character or characters,
provides film narratives and narrative comics alike with a centre of expe
rience that can function as an additional dimension of continuity, that is,
as an experiential frame connecting the narrative units.5 This is because
the character imports a sense of subjective experience to the image, un
like in a match cut (or graphic cut) where bridging between shots, or
images in comics, is based on graphic repetition and spatial matching
between particular objects, shapes, spaces, or other visual aspects of
the composition, suggesting thus an analogy of things seen or, possibly,
a visual metaphor.6
One key difference with regard to match on action in comics and films
is the so-called 180° rule, which can be seen as the very basis of conti
nuity editing in film narration,7 but which plays a more limited role in
comics. This rule dictates that in a film the scene’s movement—a person
walking, people meeting, a car racing along a road, someone swimming
laps in a swimming pool—is assumed to take place along a clear-cut
vector or axis of action. In other words, the axis of action determines a
semi-circle or 180° area where the camera can be placed to present the
action. The rule is not absolute in films—it can be broken for specific
effects without necessarily undermining the coherence of the shot—but
avoiding placing the camera beyond the line of action is a powerful
means of reinforcing the impression of the continuity and coherence of
action. The rule guarantees that the participants in a scene have a con
sistent eyeline direction and the same left to right relationship, and thus
a shared notion of what may be off-screen, and it creates the impression
that their movement has a continuous sense of direction.
In comics storytelling, the axis of action can also function as an import
ant structural principle. Traditional comic strips, where the characters
and their action are systematically seen from a steady angle somewhere
to the side of the action, follows this rule strictly, thus providing the
reader with the experience of stage-like action. However, the lack of
perspectival changes in such strips likens the viewing experience to the
atre rather than modern cinematic fiction. In modern comics that may
alter the angle, focus, and width of the perspective, the logic of direction
in movement also often respects the culturally bound rule of depicting
forward movement from left to right (in a panel as on the page) whereas
movement from right to left typically denotes return.8 For instance, in
Rob Davis’s graphic novel The Motherless Oven (2014), where the story
involves a considerable amount of walking and running—the charac
ters peregrinate in the streets, walk to school, or escape from school,
fleeing from the police, the provost, or their mechanical and sometimes
monstrous parents—movement is invariably depicted from left to right
(from different angles) with very few exceptions. When the characters
appear, unusually, to be moving from right to left, they are forced to run
(anywhere) for their lives or save each other, or seek shelter from a rain
of knives (literally a rain of knives). The changed direction of movement,
thus, stresses the exceptional situation. In turn, panels that show charac
ters moving towards the reader allow us to concentrate on their facial ex
pressions, eyeline, and dialogue. In contrast, the movement of characters
away from the reader tends to reinforce different perspectival effects,
such as establishing an over-the-shoulder perspective, i.e. sharing the
character’s perspective, or emphasising distance, such as the changing
distance between characters.9
In today’s longer comics and graphic novels, such as The Motherless
Oven, which employ alternating perspectives, the logic of narrative
space and the direction of action are seldom challenged by changing
the angle across and around the axis of action. However, at the same
time, the axis of action is a highly flexible and relative notion in this
context.10 The angle can move around the characters and the scene
without undermining the sense of logic in the narrative space, the di
rection of movement, or the relative positions of and distance between
the characters. It is, in fact, a kind of default expectation in much of
the storytelling in contemporary comics that every panel in a scene
changes the angle and field of vision by moving around the characters,
their action, or the whole scene. Such constant shifts and contrasts of
perspective are often employed, for instance, for scene-building in con
versational scenes.
Another medium-specific means of connectivity through the characters
in comics is to place and orient the characters in the panel in such a way
that their gaze points to the next panel, thereby prompting the reader to
look the same way. This may be particularly effective in cases where the or
der of the panels diverges from the most conventional forms of linearity.11
Awareness of a panel connection may thus be inbuilt into the representa
tion of the characters through the direction of their movement and gaze.
In this double spread from the fourth instalment of The Walking
Dead, by Robert Kirkman, Charlie Adlard, and Cliff Rathburn (“The
Heart’s Desire”, 2005), the story is dominated by a large, panoramic
panel which stretches over both pages and ‘bleeds’ to the sides of the
pages and which is surrounded at the top and bottom by smaller panels
(Figure 3.2). One of the top panels also partly extends over two pages,
and the order of reading in the upper strip is reinforced by speech bal
loons that are superimposed onto the gutters. The double spread port
rays the moment when the occupants of the prison, who had just been
engaged in a leadership dispute, realize the approaching attack of the
living dead from another prison block. The living are thus forced to
join forces to protect themselves. Here, the protagonist, Rick Grimes,
is shown from different angles in three corners of the double spread, as
Figure 3.2 Robert Kirkman, Charlie Adlard, and Cliff Rathburn. The Walking
Dead4 (2005) © Robert Kirkman.
well as from above and behind (turned towards the right) in the pano
ramic panel. The constantly changing perspectives and positions do not
undermine continuity in the passage; on the contrary, they add to the
sense of fast-paced action and decision-making in the scene.
The double spread is unusual to the extent that panels or tiers of panels
rarely extend over one page in The Walking Dead series. However, the ar
rangement is not so extraordinary given that in The Walking Dead, page
layouts are dynamic and constantly change according to the action and the
dramatic situation. The panel sequences in this double spread are highly
typical of the series in terms of their perspectives, i.e. how the angle of
vision constantly shifts from panel to panel around the characters of the
scene, with the perspective sometimes focusing on a particular character, or
approximating their perspective, before panning out from them once more.
The flexible, dynamically changing angle of vision around the characters
and their action, irrespective of the 180° rule, does important narrative
work in many genres and works of contemporary comics. Unlike narrative
films, comics are unconstrained by the need to avoid contradictions in the
depiction of space and movement in moving images. On the contrary, they
may exploit their necessarily discontinuous form to enhance the effects of
a moving perspective without causing jarring effects in narrative develop
ment and the sense of space. The characters’ changing position and orienta
tion, furthermore, suggest a sense of continuing action or an ensuing event.
What is extremely similar in narrative comics and film, however, is the
role of recurring characters in creating a sense of narrative coherence, and
the fact that discontinuity in such characters’ action makes scene changes
more obvious. In other words, a character shift,12 which may occur either
by addition, subtraction, or the complete change of major characters across
panels, can indicate that a particular panel transition in fact represents a
transition from one event, scene, or episode to another and from one narra
tively salient element to another.13 Therefore, the change of focus on a con
tinuing character allows the reader to make inferences about transitions
in narrative emphasis and in the larger narrative structure and context.14
Continuing Movement
Repetition of the same character in a sequence of panels can suggest
narrative continuity when the sense of action and event is weak or the
images depict no action. The showing of a character in stasis (that is,
someone who is motionless and unengaged in any obvious activity) can
serve as the description of a character or a mental state as well as the
means of establishing a scene or a situation.15 Moreover, in contempo
rary comics such as in The Walking Dead, it is quite common for the
images to focus predominantly on the presentation of dramatic situa
tions by showing the characters’ faces, positions, and gestures from dif
ferent angles, instead of ‘directly’ depicting their movements or actions.
In traditional superhero comics, action and dialogue are often found
to alternate more frequently, but in The Walking Dead series, despite
the centrality of action, the images rarely portray several phases of an
ongoing movement or action. Typically, while the characters in The
Walking Dead are engaged in a physical confrontation or are escaping
zombies—both very frequent situations in this series—they are also en
gaged in a conversation.16 Although the action scenes in this series show
aspects of the physical action, the focus lies on the characters’ facial ex
pressions and bodies, their gestures, positions, and situation in relation
to one another and the depiction of their engagement in a conversation.
When comics portray movement and phases of action, or when a char
acter is shown moving from panel to panel, the focus on the character’s
spatio-temporal path can serve as a means of continuity in quite specific
ways. This technique is effective for the same reason as match on a ction in
films: the depiction of phases of movement creates simultaneous spatial and
temporal connections. In addition, however, in the showing of a charac
ter’s movement, two levels or articulations of sequentiality also coincide: the
character’s imagined mobility and the sequence of the panels. Both of these
levels of sequentiality help to mask or negate the discontinuous form of nar
ration present in comics. Movement in itself is a way of connecting story ele
ments and adding new elements at various levels: The phases of movement
can create a rudimentary narrative, while moving characters also introduce
the reader to new places and characters, thus giving the reader new rea
sons to follow the story. Furthermore, the illusion of movement provides the
story with a visual logic that helps direct the reader’s attention and the order
of reading. The characters’ forward movement—typically moving from left
to right—reinforces the forward flow of the reading experience from panel
to panel, strip to strip, from top to bottom, or in manga from upper-right
corner to left and from the back of the book to the front. Potentially, a
character’s movement can also contribute to the reader’s sense of curiosity,
surprise, and discovery on the verso page after the page has been turned.
In order to better understand the function of movement in narrative
comics in this regard, we should consider briefly the importance of sto
ries of motion and movement for the development of narrative comics in
the nineteenth century. As much research on nineteenth-century comics
has shown, in the latter part of the century, the narrative art of com
ics developed greater continuity between images through the depiction
of motion and action. For instance, rather than presenting separate
stages of a story or an event, accompanied by captions explaining the
gaps in the action between the illustrated scenes, images in sequence
were increasingly conceived as phases of a continuous action (Gunning
2014, 41). This development was, in part, inspired by advances in mo
tion capture in photography and cinema.17
Let us consider a few examples of the depiction of movement from
the era to illustrate this important historical change. In John Tenniel’s
early comic strip for Punch, “How Mr Peter Piper Enjoyed a Day’s ‘Pig-
sticking’ (Near Burhampoor, Bengal)” (1853), the eight panels of the
story describe the different phases or events of a rather ridiculous and
anti-heroic hunting trip. In these panels, the protagonist, Mr Peter Piper,
first attempts to spear a wild boar and falls off his horse. He then stum
bles over a family of crocodiles in some bushes, meets a tiger in the same
bushes, climbs back on his horse, seemingly frightened by the predators,
catches some piglets on his way back, rides triumphantly away with his
pitiful game, and finally boasts of his achievements to his friends while
having a hearty meal outdoors(Figure 3.3). Even though the panels are
thus connected, beyond the captions, by both the recurring character,
his movements, and the theme of hunting—the protagonist is moving
in each image, and in the first panel we also see motion lines to empha
sise this fact—the sequence lacks a strong sense of continuity between
the various stages of action. Rather, what these panels depict are rep
resentative situations, such as major mishaps or revealing moments of
self-deception, as the day progresses. While many aspects of the story
rely exclusively on visual narration, there is no attempt made to capture
ongoing action or movement from panel to panel.
Figure 3.3 John Tenniel. “How Mr. Peter Piper Enjoyed a Day’s ‘Pig-Sticking’”
(Near Burhampoor, Bengal)’ (1853).
Various means for depicting a sense of continuous action and move
ment nevertheless already existed at this time, and Tenniel was also
aware of the alternatives. Two decades before Tenniel’s stories in Punch,
the Swiss artist Rodolphe Töpffer had developed a sense of continuing
movement and physical action through a lively drawing style and the
juxtaposition of image content, using such devices as character gestures
implying a particular motion or a gradual movement (a chase scene,
for instance) and modifications to the formal features of composition.
The latter included the use of a dynamic, changing panel size (nar
rower panel, shorter duration, for instance), and framing—references
to what is left out of the image can suggest causal links in movement,
for instance.
It is important to note that these historical developments in the port
rayal of movement were neither linear nor unidirectional. In another
comic that Tenniel drew 2 years later, a story entitled “Mr Spoonbill’s
Experiences in the Art of Skating”, which appeared in Punch in three
instalments in February and March 1855, he chose a range of options
for depicting movement. In this comic strip about Mr Spoonbill’s em
barrassing skating experience, we first see how the inexperienced prota
gonist has skates put on his feet in what is called “excruciating torture”,
then goes through a “variety of eccentric movements”, falls down several
times, and then, “blind with enthusiasm”, skates onto dangerously thin
ice and falls through into the water, from where he is fortunately res
cued (Figure 3.4). Finally, he is depicted struggling back home, where
he attempts to warm himself up. While some panels show representative
moments of action, such as when a “vulgar” man puts the skates on his
feet and, in the following panel, conducts him onto the ice, other panel
sequences of the story, especially when the protagonist falls on his back
or when he falls through the ice, show closely linked phases of continu
ous fast movement. Thereby, Tenniel’s story reveals both an interest in
the representation of continuous action and movement and experimen
tation with the visual connections between the panels with regard to the
character’s action.
The situation is no different in the comics of today to the extent that the
depiction of continuing movement is an option in storytelling, but the
impression of a continuing character engaged in an action or situation,
or just being present, is a dominant convention. In some popular genres,
such as superhero action comics, or war comics along the lines of the
British series Commando, scenes of continuing physical action are one
of the main means by which the story moves forward. In contrast, in
many contemporary forms of autobiographical comics and graphic nov
els, physical action and movement play only a limited role as the main
focus of narration is on the author’s experience, thoughts, and emotions.
Nonetheless, in many contemporary comics characterised by their de
piction of physical action, such as The Walking Dead, it is remarkable
Figure 3.4 John Tenniel. “Mr. Spoonbill’s Experiences in the Art of Skating”
(1855).
how movement and action are implied without focusing on the different
phases of movement. The rapidity of action, the quickness of succession
and the complexity of the event are conveyed, perhaps most importantly,
by concentrating on the expressive force of the character’s face and body
language. However, what have changed since Tenniel and the nineteenth
century are the techniques of perspective-taking and layout, which are
far more versatile today, and the various visual signs and symbols of
movement and action, which have become conventionalised. In the past
two centuries, the medium of comics has developed an impressive rep
ertoire of techniques, devices, and visual symbols for portraying move
ment within a single panel or across a sequence or group of panels. These
include motion and speed lines (lines that indicate the direction, form,
or speed of movement), ribbon paths (lines or swaths of a light shade or
colour that show the path along which a character or object has moved),
and impact flashes (symbols that indicate sites where movements are ini
tiated or terminated),18 blurring or streaking effects, the ‘sound effects’
of movement where onomatopoeia shows the direction of motion or the
speed of movement, as well as superimposed images that depict various
positions or parts of the movement.
Visual and Narrative Salience
The effect of the characters’ narrative salience may be reinforced by ver
bal narration, for instance, first-person narration, and also by visual
cues. In Gunther Kress and Theo Van Leeuwen’s visual grammar, sa
lience is one of the basic means of composition in images (the other two
basic factors being information value and framing). In this model, the
salience of particular elements in an image can be emphasised through
various formal features and cues, including the placement of the ele
ments in the foreground or background, or their size, and the use of
colour, tone, focus, perspective, overlap, and repetition (2006, 61, 183,
212–214). With regard to images in comics, Teresa Bridgeman has listed
similar formal features that help the reader create a sense of salience for
particular elements in the composition: depth of field, degree of iconic
ity (iconic or cartoony vs. a realist style), drawing styles, colour, panel
limits/framing (close-up vs. panorama), composition (possible scanning
patterns within the image), composition across page, page layout, and
panel size (Bridgeman 2013). The character’s placement in the fore
ground of the image from panel to panel increases the likelihood, al
though does not guarantee, that readers keep this character primed as an
agent. Page layout, similarly, can be used to distinguish between primary
and secondary information in a scene or ongoing narrative passage. Any
significant variation in layout requires appraisal of what is taking place
in the space of the composition. Importantly, the panel itself, which pres
ents something for viewing, is also a unit of attention,19 and therefore,
what is chosen to be depicted in a panel can be expected to be worth the
reader’s attention. Finally, repetition of visual content alone can draw
attention to a particular figure, action, or behaviour.
Comics can use all these visual and compositional cues to bring the
main character and his or her action into better focus. In this respect,
however, it is important to note that the impression of visual salience in
the image, for instance, due to the angle or focus in the image, or the
placement of someone or something in the centre foreground, may or
may not be reaffirmed as narratively salient by the surrounding panels
or the narrative context. In other words, the visual salience of an element
in a panel is not defined by the reader’s evaluation of its relevance for the
narrative as a whole.
Let us consider the relation between visual and narrative salience
in an example that self-consciously tackles, and at some level also ne
gates, the convention of continuing characters. In Martin Cendreda’s
one-page story “I Want You to Like Me”, which consists of sixteen
panels, 20 the speaking characters and the spaces shown in the images
change constantly, while what is said in the speech balloons, also in
cluding one thought balloon, suggests an ongoing, coherent discourse.
From the outset, the story seems to lack internal consistency at the visual
level: characters, situations, and places keep changing, and we have no
notion of whether these situations have any temporal connection. The
lack of continuing characters creates various unusual effects. One of
them is that the sense of narrative coherence in the sequence relies heav
ily on speech and the coherence of the sentences as a continuing thought,
monologue, or conversation. The continuing speech suggests a shared
idea and emotion for the changing characters, that is, the desire to be
liked, accepted, or adored by someone whom the speaker likes. Another
effect of the evident discrepancy between the characters and what they
say is that by preventing the reader from following a continuing charac
ter, the story thus manages to illuminate the importance of that conven
tion. One effect of the lack of continuing characters is that, facing the
impossibility of following anyone, the reader may be tempted to search
for visual clues for other connecting objects, details, or the milieu, in
order to evaluate whether there might be more implicit forms of con
nectivity in the story beyond the ongoing discourse. In this attempt, a
careful reader is perhaps rewarded by discovering that the first and the
fourteenth panels actually depict the same two characters and the same
scene: an artist painting on an easel and conversing with someone who
is looking at his work. Once this panel-to-panel connection has been es
tablished, the reader may then be encouraged to go further on the same
interpretive path. However, whether there are other translinear visual
connections is much less evident. One option for such a connection is
the potential visual analogy between the artist’s easel, already shown
twice, and the cartoonist’s drawing board, the effect of which may be
further reinforced by various other images or frames within the panels
in this story (a computer screen, graph, blackboard, billboard, mirror).
Furthermore, the angle of the dark-haired man in the last panel perfectly
replicates the cartoonist’s position in the earlier panel. These men also
have quite similar features, thus implying the cartoonist’s self-parodic
reflection on wanting to be liked, through his work, by his audience. Yet
again, in tracing such metaphoric connections the reader makes manifest
the default expectation of being able to follow a continuing character.
Spatial Attachment and the Process of Following

I have discussed above some of the ways in which the recurring, continu
ing character—in action, movement, or stasis—functions as a means of
connectivity between panels, in particular by creating a sense of tempo
ral and spatial continuum through an experiential centre of attention.
Similarly, in the study of narrative film, scholars have formulated models
for analysing the ways in which characters ‘filter’ the story for the view
ers. In this regard, Murray Smith’s theory of spatial attachment in film
and patterns of alignment and Rick Altman’s medium-independent no
tion of following-units and following-patterns have particular potential
for the narratological understanding of characters in comics. Both of
these models stress the importance of the process of following, that
is, the spatio-temporal path that a particular character takes through
out the narrative, as the audience’s main means of access to the story.
For Altman, the process of following, marked by the narrator’s activity
in selecting something or someone to be followed, usually a character
or several characters, is a basic feature of all narratives, regardless of
the medium. This premise seems particularly relevant in the context of
narrative comics and the history of their development. Smith’s approach,
however, has an additional advantage for the narratology of comics in
that his comprehensive account of narrative mediation through charac
ters integrates the questions of following and focalisation into a sugges
tive whole.
One of the key terms for Murray Smith is alignment, which refers to
a level in the spectator’s imaginary engagement with characters, 21 and
which describes “the process by which spectators are placed in relation
to characters in terms of access to their actions, and to what they know
and feel” (Smith 1995, 83). In other words, alignment considers the way
in which the viewer’s attention is restricted to character knowledge, such
as visual and aural knowledge that is more or less congruent with that
available to the characters (Smith 1995, 75). The process of alignment in
volves, more precisely, two interlocking levels: (1) the spatial attachment
to a character in the course of a scene or several scenes—i.e. the process
of following a character—and (2) access to the character’s mind—i.e. the
processes by which the spectator may come to understand how a char
acter perceives a situation, and/or reacts to it emotionally. Smith calls
these two basic structures of alignment spatio-temporal attachment and
subjective access. 22 They are conceived as distinct functions, while, at
the same time, they “interlock to produce a pattern or structure of align
ment” (Smith 1995, 144).
The central insight in the idea of alignment, and one that clearly sets
Smith’s theory apart from literary narratology, is that the filtering effect
of such alignment, jointly produced by these two interlocking functions,
cannot be reduced to the question of perspective (or focalisation) as lit
erary narratology defines it, even if these notions are closely related. In
contrast, the concept of spatial attachment stresses the importance of
characters’ spatio-temporal paths for visual narration. The process of
following a character is important, since it provides the spectator with
an entry into the narrative. Perceptual alignment is only one dimension
of this pattern. Spatial attachment to a character entails many other
features, such as the character’s action and behaviour, engagement in
an event, and way of relating to others and the world. Thus, spatio-
temporal attachment in film narration, similar to narrative comics, es
tablishes a relation between the audience and the character where the
spectator (or the reader of comics) may follow the actions of a character
or a group of characters, witness the actions of other characters when
they are in proximity to the focalised figures, and create a sense of the
evolving event through the character’s spatio-temporal path.
Let us think of an example that makes the significance of the conven
tion of spatial attachment, and the accompanying subjective access to a
character’s mind, evident. In the first instalment of Sandman, entitled
Sleep of the Just (by Neil Gaiman, Sam Keith, Mike Dringenberg, and
Malcolm Jones III, 1989), the beginning of the story suggests a complex,
multi-focus, spatio-temporal attachment to the characters. In the first
two pages of the story, set in 1916, we see Dr John Hathaway arriving at
Roderick Burgess’s mansion in Wych Cross, England and then meeting
Mr Burgess. On the third page of the story, four equally sized smaller
panels are superimposed onto an image of the gate of Roderick’s man
sion, which thus serves as a kind of overall frame for the panels. The im
ages within the frame introduce new characters and places, suggesting
potential story paths that may be followed: Ellie Marsten in Toronto,
Daniel Bustamonte in Kingston, Jamaica, Stefan Wasserman in Verdun,
France, and Unity Kinkaid in London. The four situations are con
nected to the main narrative by the theme of dreams and dreaming, and
through Mr Burgess, whose waking dreams, so we learn, are about the
power and the glory and of death. On the next page, the story refocuses
on events at the mansion, from where Hathaway has left. The alignment
is now with Mr Burgess and a new character, his son Alex, who is shown
with his father. What follows is a nocturnal scene, an unsuccessful oc
cult ritual conducted by Mr Burgess. After that, alignment and focus
shift again. The four characters who were introduced earlier are given
three panels each that reveal more of their respective situations. This is
then followed by another shift in alignment and a simultaneous change
from external to subjective perspective, where we see Mr Burgess and
Alex entering a room from the perspective of someone in the room who
is not shown (later we learn that it is Dream). The subjectivity of the
perspective is affirmed by the eyehole-type framing of the panels, the
changed colouring of the scene, and the first-person narrative voice that
we can read in thought balloons superimposed onto some of the these
panels. Moreover, Mr Burgess evidently talks to and looks at that some
one whose subjective vision we share and whose voice and thoughts we
hear but cannot yet see.
Several more shifts of character, location, and action are to follow in
remainder of the first instalment of Sandman, while some of the spatial
attachments of the beginning have soon run their course. Hathaway’s
suicide, for instance, will shortly close up one potential path to follow.
However, at the time the readers are introduced to the (as yet) unknown
subjective perspective, they may ask themselves where the story is going
and which of the many characters and situations is the main focus of
the story. There are at least seven potential units of following at this
point: Hathaway, Burgess (and Alex), the four characters who are given a
perfectly equal amount of space—Ellie, Daniel, Stefan, and Unity—and
the anonymous subjective viewpoint. Thus, the beginning of Sandman
poses questions about the direction of the story and the reader’s spatial
attachment to the given characters: how are all these characters, situa
tions, milieus, and paths connected? Who or what situation will emerge
as the focus of the story? Here, by steadily multiplying the number of
characters and potential structures of alignment (or following-units), the
beginning of the story exploits the expectation that the reader needs to
be able to follow a particular character or a group for the narrative to
really begin. At this point of the story, it is possible that all these char
acters and their situations, as individuals or as a group, may offer the
reader a potential point of spatial attachment.
From the reader’s perspective, one might say that alignment with
someone in the story involves the recognition that a character or a group
of characters has narrative salience rather than mere visual salience.
This recognition is achieved by the representation of the character’s ac
tions and behaviour, and/or consciousness, and is verified in a larger
narrative unit in relation to other characters who are given less space and
attention. Significantly, alignment with a character is also derived from
the way in which the characters are shown and how the story hinges on
their actions, perspective, and mental life. The continuous showing of a
character plays a crucial role in creating a sense of spatial attachment in
comics even if direct representation (or enactment) is not limited to the
visual channel of the medium. At the beginning of Sandman, the char
acters’ names are mentioned, and what they say or think also necessarily
affects our understanding of who we should follow. Nevertheless, spa
tial attachment is created here by means of visual showing; there is no
competing or accompanying level of continuous verbal narration in the
form of dialogue or narratorial captions that could do this. 23 If the char
acter’s actions are recounted by another character within the diegesis,
either verbally or visually through a flashback sequence, then narration,
as Murray Smith also suggests, can be said to establish an embedded
attachment (ibid.).
From the model of spatial attachment, we can derive a typology of
alignment patterns that also indicate some new possibilities for the de
velopment of a diachronic (historical) narratology of narrative media
tion in comics. We may presume that most narrative comics, as is the
case with narrative films, set up patterns of alignment between the ex
tremes of continuous single attachment, where we follow one protagonist
throughout the whole story, and multiple attachments, where narration
“successively traces the distinct spatio-temporal paths of many differ
ent characters” (1995, 146). Rick Altman’s conception of narratives,
similarly, is in terms of following-units that form following-patterns
(series of following-units) ranging from the basic forms of single-focus
narration to dual-focus and multiple-focus narration and their alterna
tion (Altman 2008, 26–27). The beginning of Sandman keeps all these
three options consciously open. If such formal solutions for continuity
building and narrative filtering were compared in a large body of comics
in a particular genre or a period, this might allow us to detect historical
changes, trends, and conventions in alignment (or following) patterns in
the medium. More specifically, this might allow us to study historical
changes in the association between forms of attachment and access to
the character’s mind, i.e. to identify the degree of subjective access that is
entailed by attachment to a character. Similarly, a narratological history
of storytelling in comics could investigate the emergence and develop
ment of parallel but medium-specific associations, formal patterns and
alignment, and their wider implications for the medium.
Conclusion
What matters from the perspective of narrativity and narrative coher
ence in much storytelling in comics—in first- and third-person narra
tives alike—is that a character or group of characters, and sometimes
other objects or the milieu, such as the landscape or cityscape, reoccur
in the panels. Visual repetition thus builds a sense of continuity and co
herence between images by allowing space, time, and action to continue
over a series of panels. The showing of a continuing character from one
panel to another, either engaged in a situation or action or simply present
in some scene, increases narrativity in any sequence of images. We may
surmise that this ‘synthetic’ quality in continuing characters and their
actions, i.e. their use as a means to connect panels in a narratively mean
ingful and coherent way, functions through the reader’s expectation of
being able to follow an experiential centre of attention. The depiction of
movement, or indicators of the character’s movement and action, such
as changing positions, can further strengthen the sense of connection
between the images. With regard to the characters’ gazes, the means of
connectivity may be derived, for example, from aligning the perspec
tive with a particular character, showing an exchange of looks between
characters, showing a character looking at something and then revealing
what is seen, or by positioning a character close to the picture frame.
In order to ensure that continuing characters provide temporal, spa
tial, and experiential connections between the panels, they are usually
given high visual and narrative salience, or grounding, in comics. At the
same time, it must be emphasised that neither this salience nor the way
in which characters may direct narrative development are based solely
on the formal means of showing the same character from one panel to
another; rather, they are also ‘psychological’ in nature. Characters are
worth the reader’s attention, perhaps predominantly, because they are
sentient beings who have anthropomorphic features. The mere depiction
of a character’s action or movement, for instance, already provides the
reader with much information about the character’s experience, inten
tions, and perception of the world. We will return to the questions of per
spective, and the way in which characters may be constructed like people
(their ‘mimetic’, person-like quality) in detail in later chapters, but before
that we must develop the issues of graphic style and narrative agency.
Notes
1 Comics, Colton Waugh claims, usually have “a continuing character who
becomes the reader’s dear friend, whom he looks forward to meeting day
after day or Sunday after Sunday” (1991, 14). Waugh, thus, refers to the
readers’ emotional engagement with ‘mimetic’ (person-like) characters with
whom they already have become acquainted.
2 See also Henry Morgan’s argument that the character “gives the image se
quence its semantic cohesion and it is around the character that the story
telling is organised” and, further, that “it is the character […] who allows
the reader to understand the sequential logic, both in temporal and spatial
terms” (2009, 35). Teresa Bridgeman argues cogently that the downgrading
of the function of the experiencing centre is a threat to the narrativity of the
text as a whole (2013).
3 See James Phelan’s definition of the synthetic dimension and function of
characters in literary fiction (1989, 2–3, 9, 20–21; 1996, 29–30). The syn
thetic (character as artificial construct or plot device), the thematic (charac
ter as idea), and the mimetic (character as person) dimension can coincide in
varying degrees, or be more or less foregrounded—their relations are deter
mined by the narrative progression where some potential may or may not be
realized.
4 These two may overlap and complement each other. See, for instance, K ruger
(2012) on how visual salience in narrative films, such as the placement of
certain elements in the centre of the image or shot, can be different from
narrative salience, i.e. how something is shown and how certain elements
are significant for the understanding of the narrative as a coherent whole.
5 Other forms of continuity editing include eyeline match, graphic match (or
match cut), cross-cutting, establishing shot, re-establishing shot, narra
tive diegesis, shot-reverse shot, and cut in (Bordwell and Thompson 2008,
235–236; Magliano and Zacks 2011, 1491). Saraceni compares repetition
between panels to a match cut (2001, 171–173).
6 In comics, graphic cuts can connect different scenes, but they seem less com
mon than in films. One reason for this may be that longer ellipses of time
can be easily précised in captions. By contrast, comics lack many cinematic
devices for indicating a longer ellipsis, such as fades, dissolves, and wipes.
7 See Bordwell and Thompson (2008, 231, 234).
8 See, for instance, Guaïtella (2003, 523) on how Hergé’s characters, who
retrace their steps, are always shown to move from right to left.
9 The different connotations of direction in the depiction of movement in
comics should obviously be examined in a larger corpus of examples to be
able to say anything more general.
10 Compare with Kukkonen (2013b, 47–48) who points out that the 180° rule
is often violated in dialogue scenes in comics.
11 Chavanne (2010, 206–207) has shown how the figures’ placement, organi
sation, and orientation in the panels can aid the technique of boustrophedon
(the arrangement of alternate strips of panels in opposite directions) or un
usual movements from right to left.
12 I am borrowing James E. Cutting’s term that he has used in reference to
continuity editing and narrative discontinuity in movies (2014, 70).
13 Some film studies, which draw on behavioural data, also confirm the impor
tance of match on action as a major indicator of a scene and salience and,
consequently, how discontinuity of action is a strong predictor of the view
ers’ perception of event boundaries. See Magliano and Zacks (2011, 1510).
14 Compare with James E. Cutting, who argues that continuity is a hybrid con
cept in films, partly psychological and partly physical—and can be broken
down to the basic parameters of location, character, and time—and that the
recognition of a scene is also psychologically determined, i.e. based on the
psychological impression of what counts as an event (2014, 69–71).
15 See, for instance, Scott Bukatman’s discussion of the significance of stasis,
and the investment in stillness and the sculptural, in Hellboy comics (2014).
16 Baetens and Frey argue cogently that it is “a persistent misunderstanding to
believe that the visual string of a graphic novel shows the successive parts
of an action unfolding in time, as if the graphic novel was offering a selec
tion of shots from a sequence of a virtual movie” and further that “What it
[graphic novel] shows is in the first place a series of variations of the face.
Even if graphic novels do tell stories, their first concern is not infrequently
the portrait of the characters and the multiperspectival representation of
their bodies” (2015, 176).
17 Kunzle argues that framing, already in the 1890s, served the representa
tion of movement, for instance by broken frames (1990, 368–369). See also
Bukatman (2006, 2014).
18 For the definition of ribbon lines, motion lines, and impact flashes, see
Potsch and Williams (2012, 15).
19 See Groensteen (2007, 53–57) and Neil Cohn (2007, 42, 2013, 56).
20 Published in Kramers Ergot, a series of anthology-style books of comic art
edited by Sammy Harkham (Oakland, Buenaventura Press, 2006).
21 The two other levels are recognition (the spectator’s construction and indi
viduation of character based on a set of textual elements) and allegiance (the
moral evaluation of characters on the basis of the values that they embody)
(1995, 82–85).
22 By attachment, Smith means “the way a narration may follow the spatio-
temporal path of a particular character throughout the narrative, or divide its
attention among many characters each tracing distinct spatio-temporal paths”
(1995, 142). By subjective access, he refers “to the way the narration may vary
the degree to which the spectator is given access to the s ubjectivities—the
dispositions and occurrent states—of characters” (1995, 142).
23 Smith points out that the decisive characteristic of what he calls direct repre
sentation in films is the absence of a mediating level of narration in the form
of dialogue or voice-over (1995, 182, n11). He does not seem to consider the
possibility that verbal narration and visual means of spatial attachment can
also be contrasted with each other.
4 Graphic Style, Subjectivity,
and Narration
The question of subjectivity in comics is vast given the multiple senses of

the word. To begin with, we can distinguish between two basic dimen
sions of subjectivity in graphic narratives, that of the author (or author
function) and the character (or a narrator). The cartoonist’s subjectivity
can be detected in the use and combination of stylistic conventions such
as the graphic line, lettering, colour, or the spatial organisation of the
page. Traditionally, graphic style, the way of drawing comics, has been
seen as a kind of signature of the story’s creation, the image bearing the
signs of its making.1 However, style as the maker’s mark is only one
dimension of the various aspects of style in comics. The visual style of
comics can also be a largely genre- and format-related issue, shaped by
a particular culture and production of the comics in question, with few
traces of an individual maker. Think, for instance, of the popular Disney
comics or the Manga industry where the cartoonists are expected to
conform to a highly recognisable style. At the same time, it is not un
usual today that the cartoonists make their mark by changing styles
from one piece of work to another or even within a single work. The di
vision of labour between the writer, the cartoonist, the colourist, the
letterer, and the like introduce further complexities in this regard. The
writer’s style can be distinguished on the basis of linguistic and literary
choices, the cartoonists are known for their visual style, the colourist for
the use of certain kinds of colours, and so on. Thus, graphic style can
also be viewed as the product of artistic cooperation.
Comics have a variety of devices available for presenting a charac
ter’s subjectivity. These include perspectival techniques, narrative voice
(manifest as external/internal, explicit, implicit, in legends and balloons),
the presentation of dialogue and thought (as speech and thought bal
loons), the technique of spatial attachment or following (as sentiments
and thoughts are revealed through action in a sequence of images), and
other means of visual showing such as facial expression, gesture, body
language, gaze, and the character’s position in the image in relation
to other things that are shown. Furthermore, a number of combined
visual and verbal signs, such as metaphoric images and pictograms
(emanata, symbolia) that mark thought, emotion, reaction and attitude,
or onomatopoeia, such as interjections, can offer access to a simulacrum
of the character’s mind. Likewise, various aspects of spatial articulation,
such as framing, sequencing, breakdown, page layout, and tabulation,
can emphasise the attribution of mental functions to particular charac
ters. For example, changes in the frame shape and size can accentuate a
character’s mental and emotional experience, and local changes in the
visual style, such as blurry images, can indicate that a certain passage
is a subjective mental image, such as a fantasy, dream, or memory.2 We
will return to many of these devices in more detail later in relation to
the issues of focalisation and the representation of speech and thought.
The traditional expectation regarding the relation of these two subjec
tivities, that of the author (or the cartoonist and the other creators) and
the character, is that they are distinct, one belonging to the actual world
of the story’s making, and the other to the world of the story. However,
as I will show in this chapter, it is possible that the interaction between
these dimensions becomes significant in the story, especially in cases
where a comic subverts the expectation of a unified style or closely asso
ciates important stylistic features with a particular character’s mind.
The question of the relation between two or more authorial subjectivities
arises especially in cases where the writer and the cartoonist are differ
ent individuals with distinguishable styles and, subsequently, the verbal
and visual narration (and styles) can be attributed to different persons.3
Next, I shall discuss the ways in which graphic style can be understood
in comics. Subsequently, I will move onto the narrative uses of graphic
style by way of stylistic variation and the so-called ‘mind style’.
Graphic Style
In literary stylistics, the concept of style refers to patterns of linguistic
choice and preference that can be attributed to a particular author’s per
sonal style, a period style, a generic style, or a given work of literature.4
In its broadest sense, literary style involves all possible linguistic choices
in the text, whether lexical, grammatical, phonetic, or contextual
choices, figures of speech, or any other. Subsequently, stylistic analysis
focusses on such linguistic elements, patterns, and structures, provided
that they are foregrounded in the text as having stylistic relevance.5
In film studies, David Bordwell has defined a film’s style similarly as
“a system of technical choices instantiated in the total form of the work,
itself grasped in its relation to pertinent and proximate stylistic norms”
(2008, 378). More broadly speaking, style in film is the use that an indi
vidual work of art makes of the medium: “the repeated and salient uses
of film techniques characteristic of a single film, a filmmaker’s work, or
a national movement” (1990, 388). The stylistically important technical
choices can, in Bordwell’s model of functions of style, channel story in
formation (denotative function), convey meanings (thematic function),
Graphic Style, Subjectivity, and Narration 111
signal a feelingful quality (expressive function), and exhibit perceptual
qualities and patterns (decorative function) (2008, 377). All these func
tions of style are observable in comics as well. Our emphasis here, how
ever, will lie on the channelling of story information through stylistic
choices and variation.
In comics, graphic style has various potential functions: it marks the
maker, a period, a genre, a particular work, or a contextual artistic
reference; it connotes the cartoonist’s tone, approach, and perception
of the world and may create specific effects of realism, dream, memory,
humour, suspense, and the like. In comics studies, Robert C. Harvey’s
definition of style as the mark of the maker and as the “visual result of
an individual artist’s use of the entire arsenal of graphic devices avail
able, including the tools of the craft” (1996, 152), represents the tradi
tional conception of the notion. The scope of this definition is relatively
broad as it pertains to all possible devices and formal options available
in the medium, from drawing techniques, the use of the brush and the
pen, narrative breakdowns, and other compositional techniques, to lay
out style and the combined effect of all these devices, patterns of choices,
and preferences. The scope of graphic style thus extends from the indi
viduality of the graphic trace to the structural organisation of mise en
page, that is, the broad functions of narrative organisation, selection,
and arrangement of both words and images within the space of a page.
However, the personal manner of holding the pen and the brush is cen
tral to this definition, and style is clearly understood as something that
belongs to the artist, not the world that is depicted.
A narrower definition of graphic style can serve us to make it a more
analytical concept, and more useful for the narratology of comics. And
yet, the question of style should not be conceived too independently from
the narrative function of channelling story information. For instance,
Hannah Miodrag distinguishes between formal aspects of pictorial
style and narrative functions of the composition. The former comprises
pictorial elements in comics (and their semantic values) in their own
right, comparable with traditional components in the art of painting,
such as line and brushwork, light and shadow (chiaroscuro), texture,
mass, order, proportion, balance, and pattern, as well as figures and
composition (the ordering of the parts of the image into a whole) (2013,
198). By contrast, Miodrag includes in the narrative function of compo
sition those forms of layout that “have no value in themselves” but only
in relation to the narrative content that they organise (2013, 219). The
problem here is not the analysis and appreciation of comics as significant
works of art in their own right, but the distinction made between the
pictorial elements of expressive line work, i.e. the visual form of comics
on the one hand and their narrative content or strategy on the other.6
Beyond layout styles, graphic line and brushwork can also serve narra
tive functions, for instance, by helping to link panels in a sequence by
purely graphic means, much in the same way shots in narrative films are
connected by a graphic cut. Visual style can also contribute significantly
to the reader’s understanding of a character or a storyworld. Think of
the changing styles in Bill Watterson’s Calvin and Hobbes (1985–1995)
that generate from Calvin’s fantasies, dreams, games, and perceptions
(modality-to-modality transition). Such stylistic shifts and transforma
tions, for instance, into cubism, the shadowy world of a hard-boiled
detective or, most often, into the reality of Calvin’s toys and games,
from spaceships and dinosaurs to his favourite tiger, are indicative of
the character’s powerful imagination and capacity to immerse himself
in the world of make-believe. And just consider how important graphic
style is in constructing a fantastic world in Winsor McCay’s Little Nemo
(1905–1914, 1924–1927) or Fred’s Philémon (1965–1986). In these two
classic fantasy comics, stylistic features such as vibrant colours and the
changing panel shapes stress the dreamlike inconsistency of Slumberland
and the unreality of the letter islands of the Atlantic Ocean. McCay’s
and Fred’s stylistic innovation and exploration of the spatial possibilities
of the medium are inseparable from the worlds and the characters that
their series depict. It is quite possible to examine their artwork in terms
of formalist art analysis, but from our narratological viewpoint, we need
to focus on the dynamic relationship between style and narrative mean
ing rather than seek to separate style from narrative content. The point
in this is not to subject the issue of style to narrative sense-making, but
to be better able to account for the narrative functions of style in comics.
In French-language comics theory, graphic style is frequently defined
more narrowly as an instance of graphic showing, a personal graphic ex
pression, or as “individual graphic writing” (écriture graphique singulière)
(Marion 1993, 251; Groensteen 2011, 92). Graphic style, in this context,
is a function of the graphic quality and identity of comics, but it does not
extend to the broad functions of narrative organisation, selection, and
arrangement of words and images—functions that these theorists usu
ally relegate to an implicit and higher level “mega-narrator” (Marion)
or “fundamental narrator” (Groensteen).7 Style, then, is conceived as a
broad phenomenon and graphic style as one of its dimensions, a matter
of pictorial choices that are stylistically relevant. Such relevance can be
measured, generally speaking, by the prominence of certain devices and
techniques of drawing (out of the arsenal that is available in the me
dium), their distinguishable (or foregrounded) qualities, and combined
effects. One possible advantage of the narrower focus on graphic style
is that it allows us to perceive style as a question of enunciation: to
what or whom do we attribute graphic style? This is not simply a prag
matic question of attributing particular graphic features to a particular
author, cartoonist, colourist, or their cooperation, but it involves the
complex issue of the relation between graphic style and meaning, for
instance, the narrative functions of style in terms of the presentation
of the characters’ mental life. Another advantage of a limited focus on
graphic style is that this enables us to highlight the narrative dimen
sion of visual style in comics. Obviously, the difference between layout,
breakdown, and graphic style is not always that clear. Stylistic choices
concerning, for instance, the drawing of frames, the gutter and speech
balloons affect the way layout and breakdown may be conceived, while
pictorial style can also be an efficient means of perspective-building,
attributing perceptual information or types of perception to a particular
character and/or world. For our purposes, however, a distinction bet
ween graphic style and other stylistic features, such as layout style, can
usefully be made.
The aspects that may be salient in terms of graphic style in a given
work of comics need to be evaluated on a case-by-case basis. This may, in
principle, concern any visual elements in comics. However, we may iden
tify certain general graphic features and techniques that are traditional
means of style, and particularly well-suited and readily available for
narrative ends. They include, for instance, graphic lines and line work.
Perhaps the most famous and recognisable style of drawing in comics,
and in particular in the European context, ligne claire (‘clear line’), has
been named after the use of clear strong lines of uniform importance.
The ‘clear line’ style curbs the expressive use of the line, an alternative
stylistic variant in many other comics.8 Another important conventional
dimension of graphic style is the surface of the images and the page.
Hergé’s Tintin albums do not allow us to see the effect of pen work and
brush strokes, whereas much of Joann Sfar’s work, such as Le Chat du
rabbin (The Rabbi’s Cat), creates a bold impression by the use of expres
sive lines and the impression of a more or less spontaneous and intensive
pen work, that the panel images, the writing and the frames alike, are
made by the artist’s hand in a kind of organic whole. Additional vari
ants of graphic style so salient in the ligne claire school are the result
of stylistic constraints that reject some conventionally available stylistic
choices and techniques: the downplaying of contrasts of light and dark
(or shadow), and the avoidance of shading techniques, such as hatch
ing and cross-hatching. The uniform application of strong, flat colours
marks another seminal stylistic feature of ligne claire. Other important
variants of graphic style, but perhaps less prominent with regard to ligne
claire, include the effects of volume (shapes, figures, ink), the impression
of depth (variety of flatness and depth), the use of visual symbols, and
the style of writing and lettering.
Consider, for example, how the three basic colours—blue, yellow, and
red—in Tommi Musturi’s Walking with Samuel (2009) function as a
fundamental element of the protagonist’s world. Throughout this word
less narrative, the three colours of the flag of Romania are featured in a
piece of cloth that the protagonist Samuel carries with him, while invent
ing a number of new uses for this object, such as a towel, flag, armband,
blanket, sail, cloth, and bag. The colours have the synaesthetic quality of
representing various sensory impressions and perceptions such as liquid
(water or alcoholic beverage), smell (flatulence), sound (bird’s song and
Samuel’s willow flute-playing emanate from the same colours), light (co
lours in a prism, fire, and the sun), and smoke (from Samuel’s cigarette).
In what is a potential reference to the cartoonist’s stylistic choices, the
colours are also seen as liquid for washing hands. Later, similar hands
are shown in another double spread with a black background, to hold
fire and point to the reader (Figure 4.1). Clearly situated outside Samuel’s
world, these hands suggest a great ‘maker’ figure, perhaps the artist’s al
ter ego. All in all, the three basic colours simultaneously present the char
acter’s multisensory experience, the metamorphic quality of his world
(all shapes change, but the basic colours remain), and the image-maker’s
stylistic choices in using expressive colours and colour schemes.
There is also a great stylistic range of options available in comics on a
scale between diverse realist styles on the one hand and cartoonish styles
and caricature on the other. This can be highly relevant, for example,
in relation to characterisation and world-building. It is conventional to
mix different degrees of realistic detail and caricature, pertaining, for
instance, to the distinction between characters and their setting, or the
foreground and the background of the image. Furthermore, the combi
nation of different varieties of realism and caricature is a stylistic variant
Figure 4.1 Tommi Musturi. Walking with Samuel (2009) © Tommi Musturi.
in its own right. Few cartoonists seek out-and-out photorealism. For
instance, Tintin and ligne claire are characterised by a specific mixture
of caricature and realism: the combination of cartoonish characters, and
caricatured faces, with realistic and sometimes even photorealistic back
grounds and interiors. In particular, the main characters’ faces are type-
like, simplified, and condensed to a limited number of basic elements
(eyes, eyebrows, mouth, and optional wrinkles) that can be easily ma
nipulated to express recognisable forms of emotion and thought. There
is no styleless comic in this respect. However, graphic style is a relative
concept since the recognition of the salience of some stylistic variant in
a particular work, such as the drawing of faces, requires comparison
with the stylistic features in the rest of the work, or some other com
parable body of work, or a particular tradition. The full stylistic effect
of the graphic elements and techniques can only be measured in their
combined effect with other stylistic choices, such as breakdown, layout,
voice, the ratio of word and image, or perspective, and in relation to the
narrative as a whole.
The Narrative Functions of Stylistic Variation

One particularly fruitful area for investigation in terms of graphic style
is dynamic stylistic phenomena, such as stylistic heterogeneity, rupture,
and shifts within a single work. From our narratological point of view,
these stylistic features concern us for two reasons. First, stylistic vari
ation makes stylistic choices and contrasts particularly conspicuous.
Second, stylistic shifts can serve and highlight certain narrative and the
matic functions, such as indicate a change of modality, perspective, nar
rative situation, temporal frame, or narrative level (between frame and
embedded narrative or between different storylines).
Literary stylistics has shown how variation in style, even just a word
or a banal sentence, may gain impact from the context in which it is
found (Leech and Short 2007, 44). In comics studies, several scholars,
including Philippe Marion, Thierry Groensteen, and Gert Meesters,
have discussed the ways in which stylistic variation can raise questions
about the cartoonist’s identity and the conception of the work as a uni
fied whole.9 A typical effect resulting from the use of heterogeneous
graphic styles in one narrative, as they argue, is that style is no longer
conceived of as a simple mark of the maker. Hence, what may happen
when an artist adopts multiple styles in one work, such as in the French
cartoonist Winshluss’s (pseudonym for Vincent Paronnaud) parodic ad
aptation of Carlo Collodi’s Pinocchio (2008), is that graphic style be
comes an issue in its own right as stylistic variation points to the way in
which the characters and their world are graphically rendered. Another
consequence of the stylistic shift is the stylistic patterning of the work,
such as through contrast or parallelism, which may contribute to the
narrative and thematic meanings of the work. Drawing then demands to
be regarded as a subtle medium that offers an infinite variety of expres
sive possibilities (Groensteen 2011, 125).
Let us examine stylistic shift and contrast in Winshluss’s Pinocchio
in more detail. One major stylistic contrast in this graphic novel is that
between Pinocchio’s tale—about a seemingly mindless robot and an
out-of-control war machine who ends up being adopted as a boy—and
Jiminy Cockroach’s (Jiminy Cafard’s) story. The alternating tales, which
focus on these two characters in their respective settings, are intertwined
through space in that Jiminy Cockroach, who is an alcoholic aspiring
writer and a parody of Collodi’s Talking Cricket and Disney’s Jiminy
Cricket (Pinnochio’s conscience in the 1940 film), nests in Pinnochio’s
empty head. Jiminy Cockroach has found there a room of his own, after
having been dismissed from his work. Pinnochio’s eyes function as win
dows and doors to his apartment.
The stylistic difference between the alternating stories is striking.
Winshluss draws Jiminy Cockroach’s tale in a markedly simpler, sketch
ier, and looser style in black-and-white strips, thus emphasising the im
pulsive and introspective characteristics of the main figure (Figure 4.2).
By contrast, Pinocchio’s story is multicoloured, often includes strong
contrasts of light and dark, and also represents a considerable amount
of internal stylistic variation. Pinocchio’s story comprises, for instance,
some painting-like splash images of varying levels of detail, wordless
passages in an expressionist style, parodic sequences of Disney anima
tion and children’s comics, and panels drawn in the style of British war
comics. In addition, the two storylines utilise different modes of narra
tion, and narrative situation, to the extent that the Jiminy Cockroach
storyline, unlike Pinocchio’s story, features external narrator’s dis
course and, on a few occasions, also showcases a narrator figure in the
images with a highly schematised face who addresses his words directly
to the reader.
The shifts between these two stories are frequently motivated by nar
rative. For instance, when Jiminy messes with the cables in the robot
Pinocchio’s head, this causes a dramatic short circuit that changes the
course of the main tale. The two tales are also sometimes connected by
gaze images, shared events, or the technique of graphic cut. In another
shift between the two stories, the last panel in a sequence of Jiminy’s tale
shows the cockroach looking through Pinocchio’s eye into the country
side. We then see Pinocchio walking, in the following Pinocchio seg
ment, in the same scene, about to approach the house of the seven evil
dwarfs and their captive Snow White (Figure 4.3). Yet another shift
involves a raven pecking at the hung robot’s head in Pinocchio’s tale,
and the sound of this pecking then wakes up Jiminy’s drunken friend
in the subsequent Jiminy passage. A stylistically striking shift occurs
when Pinocchio is thrown into a fiery furnace at a toy factory, and in the
Figure 4.2 Winshluss Pinocchio (2008) © Winshluss.
subsequent passage in Jiminy’s tale, Jiminy feels like he is dying from the
heat in his room. The drawings and the frames of this passage, “Jiminy
Cafard in the heat of the night”, are wavy and sketchy, thus reflecting the
character’s painful situation.
The stylistic shifts, therefore, accompany and amplify the transitions
between the storylines and the focus on one character instead of another,
and reflect the characters’ perceptions and experiences. This is further evi
denced in other shorter and wordless storylines that are interspersed with
Pinnochio’s and Jiminy’s alternating narratives. These include P inocchio’s
friend Candlewick’s story (“Natural Born Loser”), told in flashback
Figure 4.3 Winshluss. Pinocchio (2008) © Winshluss.
(for three and a half pages); the story, told in two instalments, about the
farmer couple who lose their child at birth and who then adopt Pinocchio
as their son; as well as a passage about the drowned Snow White and the
surfer woman who rescues her. The stylistic features of these embedded
stories are again quite distinct. Candlewick’s and the farmer’s tales have
different monochromatic colour schemes, sepia-toned and hues of violet
respectively, whereas the multicoloured four-page passage about the ro
mantic encounter between Snow White and the surfer woman uses im
agery that is reminiscent at once of romance comics and a racy B movie.
Graphic style is a highly dynamic feature in Winshluss’s Pinocchio,
but the result is not one of stylistic clash or unexpected discontinuity in
the work. These shifts in style indicate changes not only in the storyline
but also in the narrative focus and perspective, and they may reflect the
character’s experience, attitude, and emotion or the kind of world in
which they live (dystopian nightmare, parodic Disney world, pastoral
and nostalgic idyll, kitschy romance, and so on). Philippe Marion has ar
gued that Fred’s experiments with stylistic rupture can lead the reader to
deny the mimetic “pseudo-evidence” (1993, 265) of the images, i.e. their
reality effect and, therefore, perhaps better grasp the very consistency of
their graphic traces. Given the dynamism of graphic style in Winshluss’s
parody, and the fantastic fairy tale world of Pinocchio, it remains un
clear whether we can at all say that style establishes some reality effect.
Perhaps rather, the stylistic variation in this work challenges the reader
to think more deeply about the relation between style and narrative,
style and perspective, or the meaning of style in creating a character and
a sense of a world. Here, the varying styles reflect the narrative content,
and contribute to it significantly, or perhaps even create it, at once un
dermining and reinforcing style as a marker of subjectivity.
Mind Styles in Comics

Yet another narrative function of graphic style is to dramatise a parti
cular character’s world view, perception, and habit of thought. In other
words, a narrative can invite the reader to realise that certain choices of
graphic style, such as stylistic rupture or variation, need to be attributed
to an individual consciousness in the storyworld (rather than the author).
Thus, the association between graphic style and a character’s mind also
has the potential to imply that the character’s world view has profoundly
affected the way in which the narrative is told and drawn. Thus, the im
pression that visual style or stylistic variation reflects the character’s situ
ation or state of mind is taken to another level, and in a way literalised, so
that style per se seems to result from the character’s mind and experience.
A useful way of thinking about the functions of graphic style in such
cases is mind style. In literary stylistics, the notion of mind style is de
rived from Roger Fowler, who introduced the term to designate “any
distinctive linguistic representation of an individual mental self” (1977,
103) in his stylistic analysis of prose fiction. More precisely, for Fowler,
mind style is a realisation of a narrative perspective, and it is partic
ularly detectable in clusters of linguistic features or techniques that
give an impression of an author’s or a character’s world view. Geoffrey
Leech and Mick Short elaborate on Fowler’s notion, referring to it as the
way in which prose style creates “a particular cognitive view of things”
(2007, 28) that belongs either to a writer, a narrator, or a character.
When a certain mind style can be attributed to a narrator or a character,
this means that the writer slants the readers towards a particular narra
tor’s or character’s “mental set” (Leech and Short 2007, 151). The basic
premise of mind style is that all systematic linguistic choices or patterns,
such as lexical choices and patterns, figurative language, or conversa
tional behaviour, may reflect style and, subsequently, the workings of
individual minds in narratives.
The concept of mind style has roughly the same meaning as what will
be later defined as cognitive focalisation, or what Alan Palmer calls “as
pectuality”.10 All of these notions—mind style, cognitive focalisation, and
aspectuality—allow us to focus on certain textual and visual markers in
comics as cues of a character’s mental set or world view and, moreover, to
interpret these markers in relation to an evolving frame of consciousness. In
contradistinction to the other notions, however, the concept of mind style
provides us with a focus on the stylistic dimensions of narrative. This is
important in the sense that any systematic investigation of the presentation
of minds in comics needs to incorporate the question of visual mediation
in its diverse forms and must relate this question to the analysis of linguis
tic patterns—such as vocabulary, grammar, transitivity, speech represen
tation, metaphor, conversational behaviour, and deictic choices (Semino
2011, 420)—given that the comic has words. Furthermore, the interaction
between visual and verbal styles and their combined forms can function as
a marker of a mind style in its own right. The study of mind styles in comics
thus refers to those (fictional or authorial) minds to which we can attribute
cognitive functions by way of linguistic, visual, and combined linguistic–
visual patterns, techniques, and other stylistically important cues.
Let us think of some examples of character-bound stylistic choices in
which style contributes to establishing a sense of a character’s mental
state. One remarkable aspect, for instance, about the stylistic hetero
geneity in David Mazzucchelli’s graphic novel Asterios Polyp (2009) is
the function of graphic style as a means of characterisation. The various
visual styles, such as expressionist, realist, or romantic style, abstract or
mimetic style, and the changing colours and hues in the narrative cor
respond intimately to the characters’ personalities and emotional states
(and in some cases to specific events). Therefore, the various graphic
styles and colours are metaphorically attributed to given characters, con
noting their world view, experience, or emotional state.11 Furthermore,
the association between graphic style and the characters’ minds suggests
that the individual minds have affected the way in which the narrative
is visually told and organised. The situation clearly breaks with the con
ventional attribution of graphic style to an author and creates a unique
form of multi-styled and multi-perspective visual narration.
Local stylistic changes or ruptures may also create an illusion of di
rect access to a character’s psyche. For instance, Manu Larcenet’s Blast
(2009–2014), a four-part series about the homeless ex-writer and murder
suspect Polza Mancini, includes various instances of local stylistic rupture
in this sense. These ruptures, accompanied by the surprise effect of colour,
illustrate a complex inner experience in the narrator-protagonist’s mind.
In these moments, which the narrator, Polza Mancini, calls “blasts”, the
black-and-white story incorporates children’s colour drawings (the images
are drawn by Larcenet’s children, Lilie and Lenni).12 The drawings con
stitute a kind of colour explosion as if these colours were emanating from
the narrator’s mind. They first emerge around Polza’s head and then spread
all over the space of the image, sometimes superimposed on Polza’s body.
This suggests that the colour images are something that only Polza sees in
his lonely moments. However, whether Polza literally sees these images re
mains ambiguous. While Polza, during these visions, occasionally seems
to be looking at the drawings around him and even lifts his hands towards
them, the images are also shown around his body and above him when
he has his eyes closed (Figure 4.4). And, the drawings are associated with
other figurations in his mind, such as a hallucination of the Moai statues,
which frequently occur during these experiences. Later, it is revealed that
the visions are stylistically related to the distorted figures in another char
acter’s, the schizophrenic sculptor Roland Oudinot’s, obscene drawings
and collages. The reader sees these too in colour panels (as if we were
looking at Roland’s notebook). While Polza’s coloured visions are much
more childishly drawn, the last part of the series also suggests a close af
finity between these images in Polza’s imagination.
Figure 4.4 © Blast - tome 4, Pourvu que les bouddhistes se trompent DAR
GAUD by Larcenet. All rights reserved www. dargaud.com.
The stylistic contrast the colour drawings establish with the rest of the
narrative dramatises the power of the experience, giving the readers a
glimpse of the protagonist’s inner perception and tumult. The vivid co
lours and the distinctly childish style of the drawings separate the blast
sensations from his everyday experience. The narrator’s verbal descrip
tions specify their meaning. Polza Mancini explains that the blasts in
volve sensations of fullness and of the instant, accompanied by a sudden
clarity of vision as if the whole world appeared to him without morality
or any preconception. The blasts are, as he specifies, an out-of-body
experience of incredible lightness that allows him to hover above the
ground, constituting a kind of rebirth or an ‘intimate apocalypsis’, one
that implies a profound sense of union with nature.13
The verbal narrative track in Blast contributes significantly to a sense
of a continuing-consciousness frame14 in these scenes, thus deepening
the reader’s understanding of the character’s experience. Marion has
pointed out that colour in comics can have an extensive impact on the
development of an array of sensation, evoking a sensation of the real,
while colour may also have a predominantly expressive and poetic
function, presenting a high tenor of the graphic trace (1993, 156–57).
The scenes of colour explosion and stylistic rupture in Larcenet’s Blast
series rely on both of these functions (referential and poetic) at once.
Something similar happens in Marc-Antoine Mathieu’s black-and-white
graphic narrative La Qu… (1991), in which the colour explosion at the
end of the narrative marks the passage between the world of dreams and
the world of reality. Both Larcenet’s and Mathieu’s works employ the
sudden eruption of colour as a means by which to represent a character’s
inner experience.15 In Blast, however, this device is more systematic and
accompanied by a more extensive stylistic rupture.
The concept of mind style provides us with an analytical frame for
studying cases in which graphic style and an individual character’s con
sciousness are intimately associated. Nevertheless, the limitations of this
notion become apparent in wordless comics: precisely how should states
of mind be ascribed to characters?16 For instance, Musturi’s Walking
with Samuel underscores the difficulty of drawing a coherent distinc
tion between a character’s consciousness and the fictional world since
it is challenging, or perhaps impossible, to evaluate how much of what
we see in the images is subjective. Beyond his muteness, the mystery of
Samuel’s cognition is further emphasised by the nearly expressionless
face and lack of gestures. Due to sleeping, physical effort, or (perhaps)
excitement over the course of the narrative, only some minute changes
occur in his pupil-less eyes; the rare gestures he makes when he dances
remain minimal (Figure 4.5). Furthermore, the storyworld, and some
times the protagonist’s body, is metamorphic and destabilised. Samuel
seems at once to be able to live through the erosion of his world, while,
at times, his body is manipulated and literally remoulded by huge hands
Figure 4.5 Tommi Musturi. Walking with Samuel (2009) © Tommi Musturi.
that suddenly appear. The six dictums that are listed at the end of the
book, and that comprise the only words in the narrative (beyond the
title), might represent Samuel’s world view. Equally, they can constitute
the author’s interpretation of his character and the story. The dictums,
such as “Do not be afraid”, are followed by Samuel’s gradual disappear
ance into the whiteness of the page.
Conclusion
The present chapter has focussed on medium-specific features of style
in comics, involving specifically two questions of graphic style: stylistic
variation and mind style, which pose important challenges for the ana
lysis of subjectivity in this medium. The comics that have served as my
main examples show us how graphic style can have many narrative func
tions by channelling story information. Graphic style is another means
of connectivity and coherence between the images, and it helps to create
a sense of a character and that character’s experience, perception, and
world. Stylistic variation and rupture may also be motivated through
perspectival shifts or transitions between various storylines and differ
ent narrative levels. The above examples of stylistic variation, shifts, and
rupture also suggest that the notion of mind style can be applied to com
ics where aspects of style reflect a narrator’s or a character’s world view.
In other words, stylistic changes can be consistent with a given indivi
dual character’s mind, emotional state, or mental state, and they may
prompt us to imagine that elements of graphic style emanate from the
character-narrator. I emphasise here the importance of imagination, and
metaphoric attribution, and do not claim that the narrators of Asterios
Polyp and Blast are or become agents who are responsible for stylistic
choices—graphic style is metaphorically associated with their conscious
ness and experience. What these examples suggest is a kind of thought
experiment that invites the reader to question the distinction between
the character-narrator and the agent responsible for stylistic choices. At
the same time, it is important to realise that the application of the con
cept of mind style is open to challenge in a number of cases. The concept
may lose its usefulness, for instance, when graphic style appears to re
main so uncontrived that it is hard to distinguish it from a generic norm,
when it is difficult to draw a coherent distinction between a character’s
consciousness and the fictional world. With regard to wordless comics,
the challenge is to determine the degree of subjectivity of vision from
images alone.
The functions of graphic style in comics are closely related to the gene
ral problems of perspective on the one hand and enunciation (source of
narration), narrative agency, and narratorial idiom on the other. As we
have seen, graphic style is a potential means of narration, perspective-
taking (from without and within the depicted world), and the creator’s
subjective expression. The importance of first-person narratives in con
temporary comics and graphic novels is perhaps another factor that
might inspire us to rethink the relation between narration, perspec
tive, and graphic style. Yet again, since it may be difficult to agree on
what exactly would be an ‘implicit’ narrator in this narrative medium,
i.e. a narrator who is not a character and does not have a personal voice,
the distinction between these categories in comics must be given some
serious medium-specific attention. We will next turn our attention to the
narrative situation and the issue of narrators in comics.
Notes
1 See Marion (1993, 249–253), Baetens (2001, 147), Bredehoft (2011,
109–114), and Gardner (2011, 54, 66).
2 Gaudreault and Jost (1990, 128–137) refer to “opérateurs de modalisation”
(“modalisation operators”), such as flash images, which mark particular
images out as mental images in cinema. See also Miller (2007, 106, 119,
122–123).
3 In Harvey Pekar’s autobiographical comics, for instance, the question of the
author’s and the cartoonist’s stylistic coherence and narrative control over
the narrative is quite relevant. See Bredehoft (2011).
4 See, for instance, Leech and Short (2007, 10–11).
5 Foregrounding, a term used in formalist and empirical literary research, re
fers to the range of stylistic effects that occur in literature, whether at the
level of phonology (e.g. alliteration, rhyme), grammar (e.g. inversion, ellip
sis), or semantics (e.g. metaphor, irony) (Miall and Kuiken 1994, 390).
6 However, I find Miodrag’s point that words and images, as different semiotic
systems, maintain to a large extent their distinctiveness in comics despite
their various forms and levels of collaboration, quite relevant (2013, 8–11).
7 One option is to equate all visual elements in comics, including page layout
and framing, with graphic style. See Meesters (2010, 217). Jacques Lefèvre,
by contrast, divides style in comics into the components of graphic style,
the composition (mise en scène, framing), and the sequencing of panels
(2011, 15, 31).
8 The term was coined by the Dutch cartoonist Joost Swarte to refer to the
school of work in the tradition of Hergé, to which Swarte himself belongs.
9 See Marion (1993, 262–267), Groensteen (2007, 98–100, 2011, 102, 124–129),
and Meesters (2010, 232–233).
10 For Palmer, “whenever events occur in the storyworld, they are always expe
rienced from within a certain vision” (2004, 51–52). See also Semino (2011,
418–420).
11 Fischer and Hatfield refer to Asterios Polyp by way of Palmer’s notion of as
pectuality to point out how conflicts between the protagonist, Asterios, and
his wife, Hana, “are visually represented as non-compatible ways of seeing
the world” (2011, 77).
12 Larcenet also uses colour in various passages in the series, where colour
indicates a simple flashback or a recurring image in a dream. Another im
portant stylistic rupture in the last part of the series is the coloured comic
strip about a polar bear called Jasper (signed by “Milton Ferri”), apparently
a children’s comic that Polza reads with another character. The main figures
of this comic also become the stuff of Polza’s hallucinations.
13 Larcenet has described his fascination with children’s drawings “without a
code”, comparing them to drawings by mentally disabled persons: “J’ai eu
l’occasion de travailler dans un hôpital psychiatrique et d’observer les des
sins des malades et c’est aussi passionnant que ceux des enfants. Ce sont des
dessins qui n’ont pas de codes” (2011).
14 For Alan Palmer, continuing-consciousness frame is “the ability to take a
reference to a character in the text and attach to it a presumed consciousness
that exists continuously within the storyworld between the various, more or
less intermittent references to that character” (2010, 10).
15 On how colour can function in a perception structure and character narra
tion in films and how the origin of the colour may be metaphorically attri
buted to characters, see Branigan (1984, 94–95).
16 Groensteen poses a similar question about the interpretive challenges in
wordless comics to anchor images to a subjectivity, including the difficulty
in knowing whether what one sees in the images emanates from the reality
or the imagination (2011, 137).
Part III
Narrative Transmission
5 Narrative Agency
(in Jiro Taniguchi’s
A Distant Neighborhood)
The problems of narrative agency and enunciation have received much

less attention in comics studies than they have in film studies. This is
unfortunate because comics have important, medium-specific features
in the way they tell stories. Many of these features we have already
discussed, including the necessarily discontinuous form of the panel
sequence and setup, and the expressive use of graphic style. Sequenc
ing, perspective, layout, and graphic style can also draw the reader’s
attention to a kind of intelligence at work in visual narration, and these
devices do this quite differently from, say, camera-movement in a film.
Moreover, many signs that are typical of narrative discourse in comics,
from the onomatopoeic Wham! or Pow!, to speech and thought bal
loons, cannot easily be placed within the word/image binary opposition
since they are simultaneously visual and verbal, and are not analogous
to the audiovisual means of cinema or theatre.
It is thus highly interesting, but also challenging, to describe how
we go about constructing a sense of the prevailing frame of narration
in comics storytelling. By this I mean a conceptualisation of a kind
of global frame of narration that enables us to estimate the meaning
and importance of the various visual and verbal or visual–verbal ele
ments, their relations, and alternating perspectives of the story. The
notion of ‘narrative agency’ is another name for such a global frame
of narration, and it involves the question of the source of narrative
discourse, that is, the conception of some agency or agent that is re
sponsible for the selection, arranging, and distribution of the story
material.1 The notion is not unambiguous, however, since narrative
agency may be defined alternatively either as an extratextual source
of the discourse, equalling a reader’s construct of a kind of implied
or inferred author who is prompted by the text, or an instrument or
a structural principle within ‘the text itself’ that functions as a kind
of reading instructions. Here, my intention is to combine these two
views. Thus, by narrative agency, I refer to the ways in which a comic
may acknowledge the source of narrative discourse. Ultimately, how
ever, such acknowledgement depends on the reader’s awareness and
construction of that agency.
130 Narrative Transmission
It also needs to be asked to what extent our theoretical notions, such
as narrative agency or a narrator, are informed by background assump
tions about authorship and the practices of creative production in the me
dium. Presuppositions about authorial ‘voice’ or intention, authorship,
and the process of production regularly play into the narrative analysis
of comics whenever the analysis moves into the domain of interpreta
tion. The question about the relation between narrative agency and the
author’s intention becomes relevant, for instance, when the reader tries
to understand what someone intended to convey by writing or drawing
a comic in the way that they did, and not simply to understand what
the story means, what happens in it, or how the comic works struc
turally. Some of the challenges in this respect are medium-specific. As
with films, comics are often collectively authored and produced—most
mainstream mass-market comics and many independent ‘auteur’ comics
are not created by a single author or consciousness—yet, unlike most
films, a single author can also control the whole production. Thus, while
many comics involve a complex set of relations among contributors, in
many other cases we can refer to a single author. Moreover, joint au
thorship can take forms that are quite specific to the art form. This not
only refers to the traditional division of labour in comic book or comic
album productions, such as between the writer, the penciller, the inker,
the colourist, and the letterer, but also the ways of conceiving author
ship in the medium. 2 For instance, some collectively produced comics
are commonly identified with and recognised for the cartoonist’s style
(Jack Kirby), others for the writer’s style and oeuvre (Alan Moore), and
still others for the collaboration between a particular cartoonist and a
writer (Jack Kirby and Stan Lee, Albert Uderzo and René Goscinny,
Jean-Michel Charlier and Jean Giraud, and so on). Authorship is of
ten collective in comics, but since joint authorship and production can
take various medium-specific forms, the crucial question then is how
individual and collective intentions may relate to each other. 3 This is a
particularly relevant question in many autobiographical and first-person
fictional narratives.
Theoretical Possibilities: Narrative Agents in Comics

But how, then, do we conceive of narrative agency in comics story
telling? If narratological notions like ‘narrator’—or other options that
refer to the activities of a more or less personifiable narrative agent such
as ‘teller’, ‘presenter’, or ‘enunciator’—are problematic and potentially
misleading, we are left with the difficult question of to whom to attri
bute the functions of selection, organisation, comment, and distribution.
The problem is partly one of terminology, but it also begs the difficult
questions of the specificity of the medium and the individuality of the
narrative comic (versus what is common to all narratives irrespective
Narrative Agency 131
of their medium). Let us first sketch out a palette of available theoreti
cal options to address the question of narrative agency in comics, and
then elaborate the question in an extended close reading of a first-person
graphic novel.
The Fundamental Narrator

The available theoretical options, due to the composite nature of the me
dium of comics, are in many ways closer to the models proposed in film
studies than they are to narratology that is based on literary examples,
notwithstanding the essentially spatial nature of narrative organisation
in comics and the lack of sound. One option is to hold on to the narrator
concept and redefine it to fit it better to the medium. A strong version of
this approach would be to argue that an implied or underlying narrator,
or a narratorial consciousness, call it for instance a graphic narrator, is
responsible for the whole narrative organisation, including the produc
tion of both the words and the drawings, as well as the showing of each
panel image and scene. This position is one that Thierry Groensteen
adopts in Comics and Narration, following Philippe Marion’s earlier
formulations (1993) and André Gaudreault’s film narratology. It neces
sitates the distinction between the global and implicit ‘graphic narrator’
and any narrator who may be included in the storyworld.
Philippe Marion’s narrative system of enunciation in comics is a hier
archical approach to narrative agency that gives the narrator concept
a global role, as a kind of great graphic image-maker, and envisions
new subcategories of enunciation that are specific to the medium.
Based on Gaudreault’s notion of a higher-level ‘fundamental’ narrator,
a ‘mega-narrator’, or Great Image-Maker, who is responsible for both
monstration (the activities of mise en scène and shooting) and narration
(the editing of the images) (Gaudreault 1989, 88–89, 91–94), Marion’s
fundamental narrator is responsible for communicating the work in
comics as a whole. Therefore, the narrator’s activity in comics involves
narrative breakdown and page layout, and is thus roughly equivalent to
editing and montage in film composition, as distinct from the functions
of the graphic sign and showing, e.g. presenting characters in action
(Marion 1993, 193–194). Again in accordance with Gaudreault’s model,
the agent responsible for graphic showing is the monstrateur or monstra
tor (show-er). Graphiation, by contrast, is reflexive or ‘autoreferential’,
directed to the graphic trace and gesture themselves (Marion 1993, 36).
The two activities, monstration and graphiation, are partly overlapping
due to the fact that, as Marion claims, the graphic trace in comics is
always to some extent self-referential, i.e. marking the artist’s style and
subjectivity.
Groensteen derives his notions of the fundamental narrator (narra-
teur fondamental) and monstrator from these narratological premises.
However, what is important from our perspective is that Groensteen sees
the distinction between monstrator and graphiator as superfluous for
the reason that the idea of graphiation basically indicates the ‘haunting
presence’ of style in all drawn narratives rather than any separate level
of enunciation (2011, 92–93). The question of graphiation can then be
subjected to the issue of graphic style—a position that I also hold here.
At the same time, however, Groensteen distinguishes between je mon-
trant (a graphic shower who is responsible for showing the images), and
the reciter, je récitant, the instance responsible for verbal enunciation (as
in the narrative captions). Clearly, Marion’s model neglects the issue of
verbal narration and narrative instance, which can take both extradi
egetic (narrator/narrative voice does not belong to the world of the story)
or intradiegetic (narrator belongs to the world of the story as a charac
ter) forms. This is an important modifier to Marion’s model: the issue
of verbal narration needs to be given more attention, especially with
regard to many of today’s comics, such as Rutu Modan’s Exit Wounds
(2007), that create complex relations between the first-person voice and
visual ‘track’ of narration. In Exit Wounds, the main narrative burden is
seemingly carried by third-person narration, which focuses on dialogue
and action, but it is interspersed at regular intervals with passages told
by first-person narrative voice, a taxi driver called Koby Franco. These
passages have the effect of subjectifying much of the rest of the story.
Think also of the importance of the main character’s statement of un
reliability in Alan Moore and Brian Bolland’s Batman: The Killing Joke
(1988). The Joker is not a narrator, but his claim that he remembers his
past sometimes in “one way, sometimes another”, and further that “if
I’m going to have a past, I prefer it to be multiple choice!”, affects the re
liability of the many subjective flashback passages in the story. The Joker
can thus be perceived as a potentially unreliable focaliser.
At the same time, while more sustained verbal narration is crucial in
many comics today, comics rarely show narrators in the images in their
role of narrating. There are various significant exceptions, but these re
main mainly local instances. One exception is Jack Cole’s Betsy and
Me comic strip series (1958), where the protagonist Chester B. Tibbit
is shown as the narrator in the first panel of each strip, and Blutch’s
Blotch (1998–2000), where the character-narrator, the despicable com
mercial artist Blotch, is shown in the first panel of each instalment as a
kind of master cartoonist. In Bryan Talbot’s graphic narrative Alice in
Sunderland: An Entertainment (2007), the cartoonist frequently depicts
himself as the narrator who guides the tour of the history of Sunderland.
In Marjane Satrapi’s Persepolis (2000–2003), the autobiographical
narrator-character may occasionally address the reader directly in the
images where she is portrayed—thus the younger experiencing and the
older narrating self temporarily fuse into one. The rarity and perhaps a
sense of awkwardness in portraying the narrator in the act of narrating
may be another indication of the importance of telling by showing. Rela
tively rare are those comics that would suggest that all images of the story
are wholly marked by the narrator’s subjective vision, as well as those,
autobiographical comics exempted, that would imply that the images
are somehow shown to us, or perhaps drawn, by the narrator-character.
A difficult problem that characterises both Marion and Groensteen’s
theories of graphic enunciation is the multiplication of narrative agents
and levels, namely, the positing of various agents responsible for different
functions, such as verbal narration and graphic showing and, further, the
positing of hierarchical relations between these levels. In Groensteen’s
model, there are three levels of narrative enunciation at play. While any
character-narrator, including the writer as the autobiographical charac
ter in their story, is subjected to the graphic shower in the images (mon-
strator), both of these agents, the reciter (or the character-narrator) and
the monstrator, are subjected to and manipulated by the fundamental
narrator (2011, 109–110). The fundamental narrator, furthermore, is
also responsible for the panel arrangement and relations, and page lay
out whenever these elements have a narrative function. What confuses
the hierarchical relations between these levels, however, is that on the
level of graphic showing the monstrator is capable of acting to some
extent independently of the fundamental narrator: “the narrator leaves
to the monstrator the task of representing what the characters are up
to” (2013, 96). The fundamental narrator, nevertheless, ‘intervenes’ in
the monstrator’s act of enunciation whenever that includes speech and
thought. This is because, Groensteen argues, images are powerless to
translate speech, and can only cite it in their space (2011, 106). Accord
ingly, in dialogue scenes speech would be represented as if the narrator
had ‘recorded’ the characters’ speech as it is supposed to have been pro
nounced and then reports that speech to the reader (2011, 106). Thus,
one unfortunate consequence of this approach is that narrative analysis
becomes increasingly interested in identifying the activities of various
narrative instances and evaluating the relations between these modes,
for instance whether one of the reciters or the monstrators is highlighted
or backgrounded with respect to the other instance, instead of looking
at narration as a whole.
The presupposition that dialogue between characters in comics is a
kind of pretended speech act is problematic in at least two senses. First
of all, this presumes that the characters’ speech does not constitute a
level of telling in its own right, but requires us to posit a fundamental
narrator who channels and filters their narrative meaning to the reader.
Certainly dialogue in comics is always mediated, for instance through
layout, style of writing, or speech balloons, perspectival choices (point of
view images), or the narrator’s discourse that may filter and interpret the
speech. Yet, the relevance in positing two different narrative instances
and activities in conversational scenes, those of the monstrator and the
narrator, is not clear, and seems as superfluous as the distinction bet
ween the monstrator and the graphiateur. Second, the idea of pretended
speech acts that are controlled by the fundamental narrator supposes
that verbal narrative information in conversational scenes can be un
derstood on its own, cut off from the way the scene is presented in the
images. However, dialogue scenes in comics are typically based on the
relation between represented speech, the speakers, and the other con
tents of the image. What is shown in the images affects the way in which
the utterances can be understood, while represented speech guides us
to look at the images in certain ways. Briefly, it is not necessary—and it
can even be misleading—to posit two different agents of enunciation in
order to analyse the narrative meaning of conversational scenes in com
ics. It may also be significant to relate speech in dialogue to narrative
captions, or study and appreciate the effect of layout in such scenes.
The notion of the graphic narrator also begs the admittedly difficult
question of how human-like or personal the graphic narrator may be
conceived, in the sense of a human subject or intelligence communicating
to his or her audience. Speaking of film narratives, Seymour Chatman
has offered a clearly depersonalised version of the cinematic narrator
concept, understood as ‘the composite of a large and complex variety of
communicating devices’ (1990, 134). Other film theorists, however, have
preferred more anthropomorphic models, amongst them the strong po
sition that a narrating agency’s consciousness and commentary informs
each scene in a film (see Butte 2008).4 A further problem with concept of
the graphic narrator is that it may be based on false assumptions about
symmetric organisation between narratives in different media, for in
stance, that all narratives would make use of narrators.
The Implied Author

From the idea of some overall intelligence at work in comics, it is only a
short step to the notion of an implied or inferred author, or some other
author construct, that could be seen to be responsible for the choices
and values of a given story. In fact, while emphasising the importance of
distinguishing between author and narrator, Gaudreault perceived his
‘fundamental narrator’ also as an intratextual image of the real and con
crete author (1989, 88–89). The advantage of the implied author theory
is similar to the strong version of the graphic narrator concept in that
it offers us a means by which to account for the possible discrepancies
between the narrative as a whole and what is presented in words or
what we may see in the images, including the role of character-narrators.
In other words, the concept of the implied author is a theoretically
grounded way—and a kind of compromise between text-centred inter
pretation and intentionalism—for talking about the constraints imposed
on the understanding of narratives. These constraints are not created by
textual (or text-image) elements alone, or directly related to the actual
author’s intentions and activities, but require that the reader makes in
ferences about that relation. The implied author approach also enables
us to redefine the concept of the narrator in a weaker sense: the graphic
narrator would then be responsible, for instance, for all narrative ef
fects and conventions that are observable in a particular comic, or that
can be inferred from the text, but distinct from any authorial element
whose presence can be inferred in a work. Some of the uncertainties
and problems embedded in the fundamental narrator and implied author
approaches, however, are similar: is the implied author a textual and
structural element or rather something that the reader constructs?5 How
does one distinguish between those indexes in a narrative comic that
indicate the activity of the narrator and those that may be associated
with the implied author? Is the implied author an intratextual image of
the actual author(s), or not at all an anthropomorphic figure? The use
of any implied author notion requires a definition, and the definition in
itself suggests a particular perspective on the question of enunciation.
Subsequently, our choice is likely to influence the way the narrator can
be conceived.
The two options of an extended mega-narrator or an implied author
may or may not be accompanied by a (re)conceptualisation of the issue
of narrative agency in comics through a new theoretical term. In film
narratology, for instance, Manfred Jahn has coined the phrase ‘filmic
composition device’ (or ‘filmic composer’), which is in many ways com
parable with Chatman’s notion of an impersonal cinematic narrator,
while Jahn refrains from using the concept of the narrator in the broad
sense. This enables Jahn to give the concept of the narrator a strictly
limited role, relegating it to an optional status, as in the cases of a voice-
over or on-screen narrator.6 In comics storytelling, a similar ‘graphic
composition device’ or ‘comics composition device’ could be envisioned.
Concomitantly, a narrator in comics would then only refer to narrators
as characters, or narrative voices when they can be distinguished from
the author(s), i.e. cases when a narrative comic represents in some sense
the act of narration itself, such as shows the narrator telling a story,
while the overall narration is conceived in terms of an impersonal acti
vity (narration) or ‘device’.
Impersonal Narration
Finally, comics narratology can opt for not multiplying narrative agents,
that is, not using any concepts referring to an agent, such as implicit nar
rator, implied author, presenter, enunciator, monstrateur, graphiateur,
Great Image-Maker, or the like. In practice, this would mean defin
ing narrative agency in comics as ‘narration’ or in terms of some other
meaning-making activity, which can embrace the whole complex of
narrative devices and expressive techniques in the medium without re
ferring to an implicit agent of narration and enunciation. One advantage
of this option is that it would allow us to keep better in view the question
of reception and the readers’ active production of meaning. One analogy
in film studies for this position is David Bordwell’s influential theory of
film narration that rejects the idea of external personified agents such as
the cinematic narrator to explain the organisation of a film. Bordwell’s
claim that in watching films “we are seldom aware of being told some
thing by an entity resembling a human being” (1985, 61–62) seems a
logical intuition, even if it is not based on actual empirical evidence con
cerning moviegoers’ perceptions and sensitivities.7 Bordwell’s theory of
narrative agency in films does not adhere to a model of communication
that would look at stories as messages from some sender (author, im
plied author, implicit narrator) to a receiver (reader, viewer, narratee),
but stresses the role of narrative strategies, as means by which films
may have a certain effect on the spectator. In this view, the viewer’s
meaning-making processes constitute the crucial constructive activity,
and the issues of enunciation and authorial intention are to a large extent
left aside. The concept of the narrator is not altogether abandoned in this
approach, but it is relegated to those cases where there are evident traces
of the presence of such an organising instance.8
It is useful, in order to avoid confusion between various narrators at
different levels, to restrict the use of the narrator concept to the diegetic
level. What matters, then, is our capacity as readers to evaluate changing
degrees of subjectivity in visual showing and perspective, and to relate
this information to changing degrees and types of verbal narration, and
less to identify the presupposed overall authority of narration. Besides,
there are two crucial general aspects about the reading and making of
comics that need to be discussed and settled before adopting any strong
notion of a graphic or fundamental narrator. One of them is that it may
not always matter that much to the reader or viewer of comics who is
responsible for the showing or organising of the images, or indeed if
‘anyone’ is showing or seeing at all. The authority behind particular
choices in the images, or their perspective, may remain indeterminate
without blocking our understanding of the story. This is not just because
in fiction, in general, we can overlook inconsistency and paradoxical
implications concerning the source of narrative information,9 but also
that comics, fiction and nonfiction alike, enjoy considerable flexibility in
alternating the subjective quality of the images on a scale from subjec
tive to impersonal images. One option in comics, as in films, are images
(or ‘nobody’s shots’) that present reality “as if it were a question of seiz
ing the essence of the action without underlining who runs through the
shots or shows them” (Gaudreault and Jost 1999, 60). In comics, as in
film, we may also accept, in order to believe in the narration, that despite
the fact that the narrator himself is shown from outside in his story, the
events and the world of the story may be presented in some sense from
‘his’ perspective (as he remembers them, for instance). In autobiographi
cal comics, such as Persepolis, that are both written and drawn by the
same author, the presumption of correspondence between the two visual
perspectives—internal and external—may be particularly strong even if
their distance may be equally important.
Still another point that should be made is that comics vary greatly
in terms of the prominence of the verbal narrator, or narrative voice,
within one work, as well as between genres and traditions. In the classi
cal Franco-Belgian school of comics, for instance, the narrative is typi
cally based on the mastery of the spatial organisation of the images and
page, structured around dialogue and action, and the main character(s)
movements from one panel to another. There is often an extradiegetic
and heterodiegetic10 narrative voice in these stories, but this agent is
transitory and highly restricted in its role to simple indications of tem
poral and spatial changes, or occasional short comments, for example,
in Tintin, Asterix, and Lucky Luke.11 Narration through showing,
breakdown, and page layout style, are essential to this tradition, but it is
not clear why we should associate these activities with a separate agent.
Moreover, in recent autobiographical comics, where the verbal first-
person narration is continuous, and the images are often subjectified in
one way or another, thus creating a sense of a continuing-consciousness
frame, it also seems counterintuitive to separate a graphic narrator, or a
kind of comics mega-narrator, from the author as narrator who speaks in
his or her own name. On the contrary, in fictional autobiographies, the
distinction between the narrator and the author (or authors) is usually
justified, and this is also the case with some autobiographies where the
writer and the cartoonist are different individuals, but again, the choice
of terminology should not be merely based on a wish to respect assumed
symmetry in narrative structures across the media. If the theory does not
aim to create new and better concepts, the choice of terminology should
at least be flexible enough to accommodate genre- and medium-specific
devices and their uses in this regard; narrative theory should, in fact, re
flect such differences and not just infer and analyse similarities between
different narrative media.
The Story
To illustrate and ground my general arguments about narrative agency
in comics, I will draw on examples taken from the Japanese manga
artist Jiro Taniguchi’s 400-page, first-person graphic novel, Harukana
M achi-e (1998), translated and adapted into French as Quartier lointain
by Kaoru Sekizumi and Frédéric Boilet in 2006 and translated into
English as A Distant Neighborhood by Kumar Sivasubramanian in
2009 (adaptation and layout by Sly Wind Tidings). Notice that I am not
only working with a translation of Taniguchi’s novel but with an adap-
tation of his work where the writing, and the order of reading, and to
some extent also the layout,12 have been changed, that is, westernised.
Hopefully, the choice of the example does not limit the more general,
theoretical nature of my arguments.
Taniguchi’s autodiegetic narrative13 is split into two principal diegetic
levels, a frame narrative and an embedded narrative. The frame nar
rative shows us the main character, a 48-year-old man called Hiroshi
Nakahara, at Kyoto train station on the morning of April 9th, 1998. He
is suffering from a headache after a night of heavy drinking. Intending
to take a train back to Tokyo where he lives, Hiroshi, however, finds
himself on a different line that takes him to his hometown of Kurayoshi.
In Kurayoshi, Hiroshi visits his mother’s grave where he kneels down
to pray. The narrative then switches to a different time-frame where the
narrator wakes up by the same grave on April 7th, 1963. In the ensuing
tale, Hiroshi discovers that he is reliving his life as a 14-year-old boy,
trying to prevent his father from leaving his family, while maintaining
his adult consciousness. At the book’s end the narrative shifts back to
the frame narrative where Hiroshi, restored to his adult body, awakens
by the same grave and returns to Tokyo.
An important feature of the narrative is the challenge that it poses
to determining the reality of the embedded story, which comprises the
major part of this graphic novel. The frame narrative lets us understand
that Hiroshi is suffering from memory lapses, perhaps triggered by de
lirium or alcohol poisoning. Yet, the embedded narrative could also be
Hiroshi’s life-like dream, as implied by his waking up by his mother’s
grave. Furthermore, Hiroshi’s prayer at the cemetery, and some of the
symbolism surrounding this scene, associates his transformation with
divine inspiration. Finally, various aspects of the story suggest that we
are not dealing with a shift between ordinary and imaginary perception
but with ontologically unstable worlds. The latter effect is created by
way of the common fictional device of metalepsis, meaning an existen
tial crossing or transgression of different narrative levels in a hierarchi
cal structure (Genette 1980, 234–137; Fludernik 2003, 383). The most
important proof of metaleptic infractions is that Hiroshi is pictured in
the frame narrative receiving a novel from his childhood friend Daisuké
whom he has re-met in the embedded story. Daisuké’s novel has the same
title as Taniguchi’s A Distant Neighborhood, and the dedication in this
novel, ‘to the time traveller’, is only understood if we take Daisuké and
Hiroshi’s meeting in the embedded tale for real (real from the perspec
tive of the frame narrative, that is).
The following investigation comprises three narrative strategies that ma
nipulate the relation between words and images, or verbal narration and
graphic showing, and that can make us aware of narrative agency in this
medium. These strategies include the use of subjective point of view (or the
point-of-view image), the function of the spatially determined perspective in
relation to alternating modes of verbal self-narration, and the role of word
less panels and sequences, seen from a non-character-bound viewpoint.
The Subjective Point of View

To return to my question concerning the global frame of narration, how
does Taniguchi’s first-person novel raise the question of narrative agency?
First of all, the subjective point-of-view panels of the story are care
fully chosen instances that create meaningful contrasts with predomi
nant impersonal or partially subjectified perspectives. The most striking
instance of subjective visual focalising in the narrative is the moment
when Hiroshi recognises his new (old) body as a young boy. In this pas
sage, we see several panels as if through Hiroshi’s eyes as he looks down
at his hands, fingers, and the rest of his body, as well as his shadow cast
on the ground (Figure 5.1). These subjective images create the mimetic
illusion of direct access to the narrator-character’s visual experience. In
nonfiction, this would be equivalent to the witness’s viewpoint that em
phasises the immediacy of the experience. Here, in fiction, the focus on
Hiroshi’s hands, lower body, and shadow invites the viewer to share the
protagonist’s subjective viewpoint.
Notice, however, that the subjective point of view is not at all con
sistent in this passage, but alternates with impersonal perspectives that
show different parts of Hiroshi’s body from the outside. Furthermore,
the position of these impersonal points of view is ambiguous in that,
while they clearly show something that the narrator could not see, they
also refrain from showing Hiroshi’s face or eyes. Due to these effects
of ambivalence in viewpoint, Gérard Genette’s categories of internal,
external, and zero-level focaliser become too rigid for our use (1980,
189–194). In most images included in the passage as a whole, the visual
vantage point is clearly external but also limited by the character’s field
of vision. Simultaneously, on the verbal plane, we have full access to
the character’s thoughts and feelings (internal focalisation). The images
clearly convey a sense of subjective perception, in varying degrees, but
do so mostly—except in these few panels where we are looking with
Hiroshi at his own body—without explicitly internal perspectives.
The effect of this alternating perspective is that the narrator’s pro
cess of self-discovery is extended over several pages of the story, and
the reader-viewer is prompted to participate in this discovery. We fol
low Hiroshi in the town’s streets seen through his eyes, as it were. The
stroll ends in a scene of self-recognition as Hiroshi sees a reflection in a
window. Here, the subjective viewpoint gives way to a kind of graphic
appropriation of shot-reverse shot technique in films. Now we see:
(1) Hiroshi, seen from the side, noticing in a mirror a boy moving beside
him; (2) Hiroshi, seen from the window looking at the window in awe;
Figure 5.1 Graphic Novel Excerpt from PERSEPOLIS 2: THE STORY OF A

RETURN by Marjane Satrapi, translation copyright © 2004 by
A njali Singh. Used by permission of Pantheon Books, an imprint
of the Knopf Doubleday Publishing Group, a division of Penguin
Random House LLC. All rights reserved.
(3) Hiroshi, seen from his back recognising his face in the window; and
(4) Hiroshi horrified at his image, seen from a perspective that is close to
the gaze in the mirror image but not exactly the same (Figure 5.2). The
final close-up lets us see his recognition of himself as a young boy.
We can make some general inferences from this scene. First of all, the
subjective point-of-view panels typically remain localised manifestations.
Comics where subjective focalisation extends beyond a few panels at a
time are not common. If this were the case, the effect would be simi
lar to a film where the camera eye could never leave a character’s view
point and the person whom we follow and/or whose perceptions we share
could never been seen from the outside except through reflections in mir
rors and the like. Second, it is equally rare to find a narrative comic that
would obliterate the authorial focalisation of the impersonal viewpoint
Figure 5.2 Taniguchi, Jiro. A Distant Neighborhood. Vol. 1. (2009) © by Jiro

TANIGUCHI.
throughout the story. In the best-known autobiographical graphic nar

ratives such as Keiji Nakazawa’s Barefoot Gen (1973–1985), David B.’s
L’Ascension du Haut Mal (Epileptic, 1996–2003), Marjane Satrapi’s
Persepolis, Alison Bechdel’s Fun Home (2006), or in the works of Art
Spiegelman and Will Eisner, as well as in much first-person fiction, the
impersonal perspective remains predominant. At the same time, most of
these works also include crucial, carefully chosen instances of subjective
visual perspective. In contemporary documentary graphic narratives, as in
Joe Sacco’s reportages and Guy Delisle’s travel stories, the witnessing sub
jective vantage point is widely used, but still the author is frequently shown
from the outside, for instance, revealing his reaction to what he sees or situ
ating him in particular surroundings and in relation to people around him.
Autobiographical comics drawn by someone other than the writer, such as
Emmanuel Guibert’s Alan’s War or Harvey Pekar’s American Splendor,
pose even more challenging questions about narrative agency in relation to
the divergence of words and images.14 In such cases, it may be relevant to
raise the question of the proliferation of sources of enunciation and the
relation between the writer and the cartoonist as narrative agents.
Still another inference that we may make here is that comics can obfus
cate the divide between external and internal focalisation on the visual
plane, and that the verbal plane of narration can contribute to and even
enhance such ambivalence. We can find another striking example of this
strategy in the embedded narrative of A Distant Neighborhood, when the
narrator sees and hears, in his drunken dream, a conversation between
his wife and children at home. The unusual, furry-edged speech balloons,
which are placed mostly on the frames of the panels, imply that the narra
tor is following the scene, is present in it as an observer, but cannot himself
be seen or heard by the others. He is then disembodied in the scene in
which he participates as an observer (Figure 5.3).15 As for the perspectives
of the panels in this scene, it is possible to interpret all of them as belonging
to his subjective viewpoint, but it is equally possible to maintain that only
some of them approximate his supposed field of vision. The dream con
text, further, allows us to imagine paradoxical viewpoints. We may, for in
stance, suppose that in a dream the narrator could move anywhere without
being seen, or occupy the camera-eye positioned above the scene, viewing
as if from or through the ceiling. All in all, the narrator’s gaze and his voice
are paradoxically located both inside and outside the space of the images.
Hiroshi occupies the position of a kind of frozen narrator who is spatially
cut off from the world in which he thinks he is participating (at least as an
observer). The effect of such partial or one-way metalepsis, which is re
stricted to the narrator’s visual and auditory perception, would be difficult
to achieve in literary narratives, where narrators who do not participate in
the events, but who know the story world thoroughly, are commonplace.

TANIGUCHI.
Alternating Modes of Self-Narration
Another narrative strategy used in Taniguchi’s graphic novel that is re
velatory of narrative agency at work is the combination of intrinsic per
spective of the images with two alternating types of verbal self-narration.
A Distant Neighborhood, like many contemporary first-person graphic
novels and narratives, employs a doubled verbal self-narration. This means
that we have, on the one hand, the narrator’s thoughts, including ‘silent
speech’, as expressed in thought and speech balloons placed within the
images. On the other hand, the story has a continuing narratorial voice,
closer in many aspects to a written discourse, that is visually marked off
the space of the image. For the latter kind of narration, Taniguchi uses
fonts with a drop shadow and placed at the edge of the image frames.
Besides visual marking, the two types of self-narration are distin
guishable by their varying temporal relation to the spaces and perspec
tives shown in the image. The speech and thought balloons represent
the narrator’s thoughts and speech in the present tense, the here and
now of the moment that is shown in the image. The voice given in the
drop-shadow text, which uses mainly the past tense, may include the
narrator’s thoughts at the same moment.16 For the most part, however,
this voice serves the complex functions of thought report, including a
summary of the narrator’s ideas, expression of his emotions, memories,
the meaning of his observations, descriptions of broader stretches of
inner-time, and anticipation of things to come. In general, the second
verbal narrative track is not only spatially distanced from the space of
the image, but also allows more temporal freedom in the narration. At
the same time, it gives the reader much more direct access to the narra
tor’s subjective mental states. Swinging between internal and external
verbal focalisation, the continuing narratorial voice of the past tense
thus connects the two worlds of the adult man and the boy.
Occasionally, the voice in the past tense suggests that certain fields
of vision may in fact belong to the narrator, meaning that something is
perceived by him and not just focalised through the character in the im
ages. One such occasion is the moment when Hiroshi sees his secondary
school building and the narrator explains how the sight moved and al
most thrilled him. Thus, the past tense in the narrator’s voice maintains
a certain distance from the experience, but the subjective perspective of
the panel where the school is shown also closely ties the focaliser (i.e. the
holder of the point of view) to the narrative voice. The fact that the panel
is a point-of-view image is suggested by Hiroshi’s look at the school in
the previous frame. The narration in the thought balloons, in contrast,
requires the narrator’s physical presence in the image. The only excep
tion to this rule—are instances when the narrator is shown looking at
something and then we hear his thoughts in a balloon superimposed on
what he is looking at. These images rely on the narrative context for be
ing understood as subjective. An example of this is when Hiroshi verifies
the date on which he is experiencing his new life from a newspaper. In
other instances where the speaker of the balloon is not present in the
image, as in the paradox of the spatially absent but verbally present nar
rator discussed above, or when we hear a voice from the television, the
frames of the balloons are marked differently, such as with a serrated or
jagged contour, to indicate the absent source of enunciation.
Verbal narration in the past tense, which is tied to the experiencing
character’s perspective, enjoys a significant role in the conversational
scenes when Hiroshi Nakahara first meets his parents—his dead mother
and his absent father—and other family members in the embedded story.
Here, the narrative captions identify the milieu and the people he meets,
and they also explain the narrator’s emotions during the scene—he
did not believe his eyes or was about to faint. This is a highly para
doxical situation: the narrator draws attention to his hindsight as an
adult, whom he continues to be (inwardly) in the embedded story, but
this hindsight—concerning what happened in the past and what conse
quences those events had—is not altogether helpful since he is reliving
his past differently and perhaps in ways that may change his future.
How would it, then, help narrative analysis to think that there is an
extradiegetic narrator, who is responsible for the utterances of the di
alogue, and possibly also for the p erspective-taking and layout, while
the character-narrator is in charge of the narrative captions and a mon-
strator does the showing by images?17 Instead of distinguishing between
these narrative activities and their respective agents, and evaluating their
hierarchies, it seems imperative to understand the relationships between
these activities, and examine carefully the possible gaps between the
diverse forms of narration and planes of enunciation.
All in all, the verbal strategies of split self-narration cue the reader to see
the images in relation to certain temporal frames or changing temporal win
dows, marked by the narrator’s perception and consciousness, even if clear
cases of point-of-view images, belonging either to the experiencing char
acter or the character as a narrator, remain singular instances. In order to
generate a fully developed narratology of comics and narrative mediation in
multimodal documents such as graphic novels, we must be able to account
for such changing relations between various levels of verbal narration and
visual showing, or between voice and the centre and focus of perception.
Wordless Impersonal Perspectives

Another means of manipulating the relation between the vantage point of
the image and verbal narration in A Distant Neighborhood, and indica
tion of narrative agency, is the strategic use of wordless panels with non-
character-bound focalisation. These involve frames that can be attributed
to narration, or the narrative event, and to that alone. The most striking
examples of such panels, which remain localised instances in the story, can
be found in the diegetic transitions between the frame and the embedded
narrative. At these turning points in the narrative, marking the shifts bet
ween the narrative levels, we see scenes of the sky, the moon, or the ceme
tery from above. An important narrative function of this series of images
is to emphasise maximum distance from the narrator-character’s percep
tion and consciousness. The changes in the frame size and shape, break
ing the usual rectangular panel form into uneven frames or to full-page
panels, further stress the sense of distance and transition. Several of these
images portray a butterfly flying over the scene, and also approximate the
butterfly’s position in the air; sometimes the butterfly is also superimposed
on the panel frames and the gutters to underscore the sense of transition
between the spaces and temporal frames (Figure 5.4).

TANIGUCHI.
Symbolically, the butterfly is a likely reference to the many J apanese
stories of the butterfly as the wandering soul and as the soul of the dead.
Many of these stories are of Chinese origin,18 including the C hinese philo
sopher Zhuangzi’s famous butterfly dream: upon waking up Zhuangzi no
longer knew if he was Zhuangzi who had dreamt about being a butterfly,
or a butterfly dreaming he was Zhuangzi. Like this story, we can think
that in the beginning and at the end of the narrative, Hiroshi Nakahara
goes through a miraculous transformation in his dream, like Zhuangzi,
and that he guides us to see the world from this perspective. Yet, how can
we come to this conclusion if the point of view, after all, remains external
and impersonal, and the butterfly is seen from the outside?
One way to answer this thorny question is to argue that the narrative
perspective in the images of transformation supports the theme, namely,
that the ambivalence of the perspective accentuates the double iden
tity between the man and the butterfly. The scene builds on the formal
convention, to which I have already referred, that comics can be clearly
marked by a sense of subjectivity while the subject is shown from the out
side, thus highlighting the question of the relation and distance between
a perceiving entity and a narratorial identity. In comics storytelling, as
in modern picture books, it is almost automatic to combine first-person
verbal narration with third-person images, i.e. images that show the nar
rator from the outside.19 There is thus a potential disparity between what
is told and what is shown. Hiroshi, as we learn in the course of the nar
rative, is uncertain about how to evaluate the reality of the embedded
narrative, to know whether he is inside or outside of the story in which
he participates. The lack of words in this sequence accentuates the uncer
tainty concerning the identity of the butterfly in the scene of transition,
or perhaps, suggests that Hiroshi and the butterfly have merged into one.
At the same time, other visual and verbal elements of the story accentuate
the radical and deeply estranging nature of the transformation: in the em
bedded story, Hiroshi continues to think and speak like an adult resident
of Tokyo, not like a boy in a small village in the countryside.
The function of the transitional panels differs fundamentally from
other wordless panels in the story, where non-character-bound focali
sation establishes scenes, focusing on the central themes and figures of
each chapter. Some of these wordless images located at the beginning
of chapters in fact imitate a photographic image, thus emphasising the
visual or picture function (vs. narrative function) of a single panel image.
Here the images, furthermore, seem to point self-consciously towards an
authority that is capable of such combination and organisation, freely
moving between various positions of telling and showing and occupying
many points in time and space at once. Yet, even the impersonal field
of vision of these images may sometimes encompass a personal one, for
instance, by showing the main character’s shadow in the foreground of
the image.
Impersonal Visual Imagining
Some recent theories in film studies have argued that the default mode
of the spectator’s engagement with cinema is impersonal visual imagin
ing, but that there can always be suitably cued episodes of make-believe
seeing from a subjective viewpoint (see Gaut 2004, 246). The features
of perspective-taking in Taniguchi’s graphic novel point in the same di
rection: at the visual level of the story, the impersonal perspective is the
predominant mode of showing and the (purely) subjective point of view
remains a localised instance of intervention. Impersonal points of view,
however, despite revealing things that the character could not see, are
often more or less related to, or encompass or are synchronised with,
the character’s point of view. They show fields of vision as if from the
character’s back or their side, illustrate the character looking at some
thing, or include their point of view within a broader field of vision in
another way. Purely non-character-bound perspectives remain localised
instances.
The manipulation of the changing relations between the visual and
the verbal register in this story also points to the conclusion that nar
rative agency in comics storytelling is not only a composite of different
semiotic sources or channels, but is necessarily variable and morphic in
its use of these sources. For instance, a continuing use of thought report
is a strong indicator of first-person narrative. Yet, the verbal narrator’s
relation to the vantage point of the image—his presence as the filter and
‘source’ of the images that are shown—can best be evaluated by paying
heed to the changing relations, including contrasts and incongruities,
between words and images. The compositional techniques of the panel,
the frame and the page setup, the balloon shape, or the onomatopoetic
words and signs can also bear traces of perceptual activity, of mental and
emotional processes of selecting, organising, and commenting on the
images. Furthermore, various components of graphic style, for instance,
a recognisable graphic line and trace, colour, or stylistic rupture, can be
attributed to a certain source of enunciation and narrative discourse,
and ultimately the author. The sense of mental processing implicit in
such devices may sometimes be associated with the character-narrator,
even if in general it remains impersonal.
Further, to study the narrative functions of the intrinsic (literal) view
point of the image requires one to revisit the basic narratological dis
tinctions between who sees and who narrates, as well as the difference
between so-called internal, external, and zero-level focalisation that is
important in classical narratology based on the study of literary nar
ratives. Such distinctions are different in comics that both tell a story
in words and images and show a world in images. What counts, then,
are the varying relations between a narrative voice, the verbal focal
iser (i.e. the one who perceives things in verbal narration), the centre of
visual perception (or visual focaliser), and the centre of attention (the
visual focalised). Furthermore, in comics storytelling, these relations
may involve different strategies of congruence and deviation between
the components. We may equally pose the question of the reader’s dispo
sition to appreciate the medium-specific play on the anthropogenic dis
tance between a narrating and an experiencing self, between a self who
speaks, a self who sees, and a self who is seen. A Distant N eighborhood
foregrounds the question of narrative agency by complicating the rela
tion between the narrator’s voice, his physical presence in the image, and
the often photorealistically rendered world that the images depict. The
narrator’s bodily transformation, furthermore, emphasises the neces
sary split in the act of perception, possible in any homodiegetic narrative
comic, between the experiencing and the narrating self, and between the
personal and the impersonal vantage point of the image.
Notes
1 See, for instance, Seymour Chatman (1990, 113–119). I also draw on
Donald Larsson’s formulation of narrative agency: “Is there intradiegetic
or intratextual evidence of extratextual agency? In other words, is there
evidence within the diegesis or the text (aside from title pages, prefaces, ac
knowledgments and so on) of how this text came to be? How, if at all, does
the text acknowledge (or pretend to acknowledge) that agency and why?”
(2000). Larsson’s model, in contrast to Chatman’s notion of the impersonal
“organisational and sending agency” (1990, 127), emphasises the reader’s
role, asking thus “what awareness of ‘acting,’ ‘speaking,’ and ‘creation’ we
bring to fictional texts”, and further, and how texts “encourage, evade, sup
press, or direct that awareness” (2000).
2 Surprisingly little has been written about the theory of intention, author
ship theory, or the history of collective production in comics studies to date.
Some exceptions are Ault (2004) and Uidhir (2014).
3 This question has often been posed in film studies. See, for instance, Sellors
(2007, 268).
4 For an overview of this debate, see, for instance, Gaut (2004), T
homson-Jones
(2007), Bordwell (2008, 121–133). Similarly, in literary narratology there
has been some well-argued resistance to the notion of the narrator, espe
cially in relation to third-person narration. See, for instance, Patron (2009,
135–147). However, many literary narratologists also hold that the tendency
to attribute stylistic features to a hypothetical narrator persona may be a
kind of default expectation. See, for instance, Fludernik (2001, 622).
5 For a thorough discussion of this paradox, see Kindt and Müller (2006).
6 The filmic composition device is a theoretical agency and device that, as
Jahn specifies, “need not be associated with any concrete person or charac
ter, particularly neither the director nor a filmic narrator” (2003).
7 Bordwell also argues that “very few viewers would take, say, a bit of actors’
business or a pattern of lighting as having its source in an intermediary, a
cinematic narrator” (2008, 122–123).
8 Gaudreault and Jost (1999) point out, however, how Bordwell’s abstract
instance of ‘narration’ is occasionally granted clearly human attributes.
For example, Bordwell argues that: “Furthermore, the pensive ending
[to Antonioni’s La Notte] acknowledges the narration as not simply power
ful but humble: the narration knows that life is more complex than art can
ever be” (1985, 209–10).
9 See Thomson-Jones (2007, 84), Walton (1997, 65).
10 Genette distinguishes between two basic types of narrative with regard to the
narrator’s relation to the story and its characters: “one with the narrator ab
sent from the story he tells” (heterodiegetic), and the other with “the narrator
present as a character in the story he tells” (homodiegetic) (1980, 244–45).
11 This is, obviously, a sweeping generalisation. See, for instance, Groensteen’s
insightful analysis of the different modalities of the narrator’s intervention in
Franquin’s Spirou and Fantasio album L’ombre du Z (1962) (2011, 96–98).
12 Various changes occur in the translations: the placement of the narrative
texts, speech balloons, and onomatopoeia in the image field often differ, the
number of balloons per panel may change, onomatopoeias are sometimes
omitted, the manga convention of using colour pages in the beginning is not
respected in the English translation, and so on.
13 In Gérard Genette’s terminology, an autodiegetic narrative is a first-person
narration in which the narrator is the story’s protagonist (1980, 245), and
thus the narrator’s experience and actions are central to the story.
14 See Baetens (2008, 84–85).
15 In contradiction with Jan Baetens and Hugo Frey’s claim that in graphic
novels characters are never disembodied (2015, 185).
16 Since there is continuous interaction between these two levels of first-person
verbal narration, I find it problematic, and potentially misleading, to suggest,
as does Badman (2010, 97), that narrative discourse in the thought balloons is
not part of narration but monstration, i.e. showing and narrating by images,
and an indication of internal focalisation at work in the monstration only.
17 One open question about the category of the monstrator in Gaudreault’s
theory is also whether the monstrator can have more than just a narrative
role, such as a descriptive or painterly function. See also Gunning 2009,
xxii–xxiii.
18 See Hearn (1904, 96–110). I am aware that the symbolic meaning of the
butterfly as the soul, or as the soul of the dead, can be found in other cultures
and traditions. For instance, Psyche of the Greek myths is a butterfly-winged
goddess. I would like to thank Associate Professor Mayako Murai for her
helpful suggestions about the myth of the butterfly and her interest in the
presentation, based on an earlier version of this chapter, which I gave at
Kanagawa University, Yokohama, on 24 August 2010.
19 See Nodelman 1991, 4; Yannicopoulou 2010, 66–67; 73–75.
6 Focalisation in Comics
Perspective is a key aspect of transmission and mediacy in narratives,

regardless of the medium. The choice of the kind of narrator or narrative
voice, for instance, brings with it a narrative perspective from which
the events, the characters, and their world are presented. The reflecting
character through whom the events, other characters, and their world
are perceived is another crucial technique of perspective. Beyond this,
stylistic and compositional choices pertaining to language and visual ex
pression, which cannot be attributed to a narrator or character, may also
be conceived in terms of authorial perspective. For instance, in much of
Chris Ware’s work, such as in Building Stories or in his graphic novel
Jimmy Corrigan: The Smartest Kid on Earth (2000), the graphic quali
ties of the text, and more precisely, the careful attention to the place
ment, shape, and colour of the graphic words and letters, functions as a
form of perspective-taking.
Perspective, which is usually referred to as focalisation, is one of the most
debated and researched questions in narratology. In his ground-breaking
treatise Discours du récit, Gérard Genette undermined the hierarchy of
‘showing’ and ‘telling’, much used in earlier literary studies, by claiming
that ‘showing’ in verbal narrative discourse can be only a way of telling.
Genette then introduced the new term ‘focalisation’ as a replacement
for ‘perspective’ and ‘point of view’ and made the distinction between
narrative mode and voice. In this distinction, focalisation answers the
question Who is the character whose point of view orients the narrative
perspective? whereas voice pertains to the question Who is the narrator?
More simply, Genette argues, this distinction can be formulated with
the questions Who sees? and Who speaks? (1980, 186). The distinction
between focalisation and voice is significant for narrative analysis, in
particular, as it is possible for one person to express the perceptions and
experiences of another. This can happen in real life as well, but it is char
acteristic of narrative fiction where narrators relate characters’ vision
and perceptions within their own discourse. If the distinction between
voice and mode is not made, we may then not be able to sufficiently
comprehend the role of mediacy in narrative fiction. Furthermore, from
this distinction, Genette derived the well-known triadic typology of
Focalisation in Comics 151
the focal possibilities of internal, external, and zero-level focalisation,
based on the degrees of access to characters’ minds in a given narrative
(Genette 1980, 187–189). Internal focalisation, whether fixed, variable,
or multiple in kind, involves a perspective that is limited to some char
acter’s mind. External focalisation, in contrast, is spatially limited to
the role of the witness, without direct access to characters’ psychology.
Finally, zero focalisation, which Genette sometimes also calls ‘omni
scient focalisation’ or ‘non-focalized narrative’, gives the illusion that
the narrative perspective is spatially unlimited.
The concept of ‘focalisation’ has changed a great deal since these for
mulations, and Genette’s basic premises have been brought into ques
tion numerous times. For one thing, Genette himself later redefined the
question of Who sees? as Where is the focus of perception? (1988, 64),
thus moving the question of perspective from some character as a seer
to the problem of the affective, perceptive, and conceptual centre orient
ing the narrative. However, no character needs to embody such a centre
of perception. Mieke Bal’s early critical comments on this theory have
also been highly influential in that she threw out Genette’s category of
external focalisation altogether because it rests on a confusion between
Who sees? and What is seen? The category is really based on clues about
how something is seen from the outside without the mention of inside
views—if, that is, focalisation is consistent throughout the story (Bal
1991, 83–84; 1997, 142, 146). Bal thus redefined focalisation as the rela-
tion between the vision of the agent who sees (i.e. focaliser)—which does
not need to coincide with any character—and what is ‘seen’ or perceived
(i.e. focalised) (Bal 1997, 142, 146). Thus, any analysis of focalisation
in narratives should investigate the relationship between the agent that
perceives and that which is seen. Moreover, we need to add, with regard
to visual narratives, the importance of the relation between what the
characters are presumed to be seeing and the image field as a whole.
These two can coincide in point-of-view images, when we supposedly
see what a character sees, overlap in other cases to a varying degree, or
remain more or less distant.
The re-emphasis of focalisation as a relation between the focaliser,
the focalised, and the image field is significant in the context of comics
storytelling since the person or character who is the focus of attention
always has the potential to import perceptual information and subjec
tive vision into the image. The general point that this brings to view is
that there is a potential focaliser in every focalised person in comics:
images reveal looks, acts of looking, fields of vision, and so on. This
potential is similar to that of literary narratives, but images in comics
also literally reveal acts of perception from the outside, that is, from
an external viewpoint. Furthermore, other images in the sequence may
always modify, through juxtaposition, the sense of subjective perception
in a particular panel.
This chapter will concentrate on a limited number of vital questions
about focalisation in comics, some of them medium-independent but
others more medium-specific, that need to be accounted for in order to
achieve a more comprehensive understanding of narrative perspective in
this medium. These include the problems of:
• the relationship between narrative voice, mode, and showing,

• epistemic access to the point of perception and what is perceived,
concerning the spatially determined point of perception in graphic
images,
• the distinction between perceptual and cognitive focalisation,
• embedded and simultaneous focalisation, i.e. the simultaneity of dif
ferent visual focalisers inside and outside the image frame,
• the complex scale of intermediate (visual) focal points between in
ternal and non-character-bound positions, or between personal and
impersonal viewpoints.
These questions involve visual information about observation, such as

the centre and object of perception, the relation between various visual
focalisers, and the degree of deixis1 and subjectivity in the field of vision.
In addition, I must make one further note about the consciously limi
ted scope of this chapter: in what follows, I will be operating with a no
tion of focalisation that is restricted to the information that the narrative
conveys about the spatial and physical point of observation, the sensory
range of that position, and the spatio-temporal position of the focalised,
that thing which is perceived. This is simply to avoid making the term
too broad to be useful. I admit that it may not be possible to make an ab
solute distinction between an analysis of the spatial point of observation
and, for instance, assumptions about the potential meanings indicated in
what is seen. Judgements about internality, externality, or omniscience
in narrative perspective are frequently accompanied by presuppositions
that concern the limits of the narrator’s or the character’s knowledge and
experience, or the epistemic motivation in perception.
Who Speaks? Who Sees? and What Is Shown?

Recently, there has been a strong trend in so-called postclassical narrato
logy to emphasise perception as a fundamental cognitive frame in under
standing any narrative, while these new theories have again challenged
earlier definitions, typologies, and uses of the focalisation concept. Per
haps most importantly, Monika Fludernik has called our attention to
the idea that focalisation is largely an interpretive category, or at least
“not exclusively a textual category” (1996, 345), i.e. a postulation that is
determined by the processing of certain textual clues, which in literature
are verbal clues, such as deictic and expressive markers. The person who
‘sees’, as she claims, “is the reader, but à travers the linguistic medium, and
not in terms of visual perception” (Fludernik 1996, 345). For this reason,
Fludernik argues that narratology has assigned false rigour to the dis
tinction between ‘who speaks’ and ‘who sees’. Furthermore, Fludernik’s
subordination of perceptional parameters in a n arrative—or the percep
tual metaphor—to the question of the presentation of consciousness may
help us to see how it does not always matter who speaks or sees in the
narrative (1996, 345–346). What may be much more important is how
the reader, or the viewer, gets optimal information about a character’s
consciousness: his or her motivations, thoughts, and perceptions.
In the multimodal narrative environment of comics, the classical nar
ratological distinction into voice and mode (or focalisation) must be
modified for at least three reasons. First, the question of Who sees? or
Where is the focus of perception? is relevant on two different modal
levels simultaneously, both in words and in images. Focalisation is ex
pressed in words in Taniguchi’s A Distant Neighbourhood, for instance,
when the narrator relates his emotions and thoughts as he looks at his
little sister who is sleeping next to him (“I notice the light breathing of
my sister sleeping beside me. It feels very strange”; “I can’t believe that
I’m sitting here like this…looking at my elementary school-aged little
sister”, 63). At the same time, focalisation is realised visually in a series
of perspectival shifts between the narrator’s gaze images and the object
of looking at different angles.
Second, while the question of Who speaks? is significant in relation to
verbal narration in comics, that relationship must also be reconciled with
the visual narration and information. In the multimodal context, the
formal markers of perspectival filtering, therefore, are necessarily both
verbal (metaphorical) and visual (literal) clues that require the reader to
integrate perceptual information from different semiotic channels in a
meaningful way, for instance, in terms of complementary functions or
contrast. Thus, the question of cross-referencing, or interplay, between
visual and verbal information conditions the distinction between focal
isation and narration. Wordless comics make an important exception in
this case, as the question of Who speaks? can become largely irrelevant,
even if, insofar as a wordless comic has a title, the issue of how to relate
images to words remains pertinent.
Third, as comics can also narrate without any words, we need to add
to the questions Who speaks? and Where is the focus of perception? an
other set of questions of narrative mode and mediacy: What is shown?
and How is something shown? These questions underscore the impor
tance of narration by predominantly visual means. Here, we could also
refer to a kind of universal ‘grammar’ of pictorial narration where visual
showing is essential in narration, that is, to the importance of showing
(versus telling about visual perception) across various visual narrative
media, from picture books to films and from comics to photo essays.
In comics, therefore, the processing of narrative information involves
paying attention not just to the distinction between Who perceives? and
Who narrates? but to the interplay between a narrative voice, a verbal
focaliser, a centre of visual perception (the visual focaliser), a centre of
attention (the visual focalized), and the image field seen in the picture
frame. We have to take into consideration the multiple ways in which
the textual element (by which I mean written and drawn language) and
visual focalisation interpenetrate each other and thus allow a multipli
cation of perspectives by way of typography, page and panel setup, and
other means. Comics present a specific challenge to transmedial nar
ratology in this respect, as the medium requires the reader to integrate
perceptual information from different semiotic channels in ways that are
both similar to and different from other forms of visual or multimodal
narration.
Fludernik’s claim about the perception of reality in films, put forward
in An Introduction to Narratology (2009), points to a fundamental
difference between verbal and visual ways of encoding narrative per
spective beyond particular narrative techniques that may or may not be
employed across media. This involves the question of epistemic access to
narrative mediation, including knowledge of both the point of percep
tion and what is perceived. Recent debates over narrative agency in film
narratives offer us additional concrete means by which to conceptualise
such a fundamental difference between verbal and visual focalisation.
The basic idea in this debate is that film images give us a view into a
space from some more or less determined perspective and that films, in
fact, cannot avoid doing so.
Visual and multimodal narratives can employ a wide range of visual
techniques as means of focalisation, including, for instance, techniques
that may be associated with stylistic features in literary narratives
but that do not have exact equivalents in literary discourse. In David
Mazzucchelli’s graphic novel Asterios Polyp, for example, every charac
ter is given an individual visual style, lettering, dominant colour scheme,
and word/thought balloon format, and therefore all these techniques
also become means of focalisation. Another important feature of per
spective in Asterios Polyp, and one that would be difficult to realize in
the moving images of a film, is the partial superimposition of panels on
one another. For instance, the overlapping panel frames of a passage
that relates the meeting of the main character Asterios and his girlfriend
Hana with the composer Kalvin Kohoutek create a sense of a continuing
space, while simultaneously they also divide the space into individual
areas (the varying colours in the panels further sharpen that distinction).
Thus, the ‘polyphony’ of these panels does not concern spatio-temporal
transitions as may be the usual effect in the cinematic technique of the
dissolve. Moreover, the overlapping panels may be taken as an illustra
tion of the themes of simultaneity and polyphonic experience that are
discussed by the characters in the panels. The split screen device in films
may create similar effects of simultaneous events or disruptions of space,
but the stress in Asterios Polyp is on spatial simultaneity and ambiva
lent spatial relations. There is no split screen. The speech bubbles enjoy
spatial freedom from any screen structure, being both juxtaposed on the
images and located outside the panel frames. The spatial organisation of
the panel sequence and the page, as well as the distinct visual styles that
are attributed to individual characters, highlight the importance of the
questions What is shown? and How is something shown? in the space
of the image.
Spatially Determined Forms of Perception in Comics

Narrative theory has to come to terms with the simple fact that images in
comics, as in films, have an intrinsic, explicit point of view from which
they are seen, no matter how personal or impersonal that perspective may
be. In literary narratives, the sense of a physical, spatial vantage point
from which the story is told may, likewise, be indispensable, serving as
a cognitive frame in the understanding of the story (see Bortolussi and
Dixon 2003, 171). Spatial positioning in literature, however, is crucially
different from perspective in comics, in that prose fiction may relatively
freely fuse, in any one sentence, the vision through which the elements
are presented and the narrator’s voice that verbalises that vision.2 More
over, spatial positioning in literature is not an inbuilt feature of the me
dium. Genette has famously stated in Narrative Discourse Revisited
that, unlike the movie director, “the novelist is not compelled to put his
camera somewhere, he has no camera” (1988, 73).3 As the invention of
concepts such as Manfred Jahn’s “zero”, “weak”, and “ambient focali
sation” (1999, 98), and Fludernik’s “neutral narrative situation” (1996,
172–177) also indicate, it is not always possible to locate any exact point
of perception in literary narrative discourse. Point of view can certainly
remain ambivalent in visual storytelling as well, but any image that cre
ates the effect of three-dimensionality presupposes a perspective.
In literary narratives, further, we conceive propositions about focali
sation and voice by determining certain verbal clues, such as deictic and
expressive markers (see Fludernik 2001, 633; Herman 2002, 306–309).
The sense of vision or voice in literature, therefore, is always a mat
ter of verbal interpretation of another verbal representation. In com
ics, the reader may conceive propositions about the presented world
both in terms of verbal clues (or metaphors of visual perception) and
graphically rendered (literal) visual perception. When reading comics,
we often literally see a person or a character in action, such as looking
at something, from an external focalised perspective, or through the
perspective that belongs to that person or is close to a figure who partici
pates in the scene. The attribution of traces of personal or impersonal
intelligence to a continuum of a single individual consciousness in the
images can be challenging. Still, this ambiguity has a different character
from literary narratives.
When film theorists compare techniques of focalisation in film and
literature, they often point out a crucial difference between these media
in regard to the epistemic access available into the world that is seen.
For instance, Katherine Thomson-Jones has argued that the “placement
of the camera in filmmaking makes it impossible not to have an explicit
point of view in cinematic narration, whereas literary events need not
be described from anywhere in particular” (Thomson-Jones 2007, 88).
Other film theorists, similarly, contrast the ‘explicit focalisation’ of films
to metaphorical viewpoints in novels.4 The basic idea put forward in
these approaches is that cinematic images give us a view into a space
from some more or less determined perspective even if the degree of their
deictic value or the degree of subjectivity of vision may be difficult to
designate from the images alone.5 In comics storytelling, as in film, the
point of view may always remain impersonal or ambiguous in terms of
its possible subjectivity, but the image necessarily reveals a spatial point
from which something (the focalised) is perceived. This constraint has
various potential consequences for the understanding of the text-image
interaction. One possible result is that it may not really matter that much
to the reader of comics who is seeing, or if any person is seeing at all,
since we know (culturally and intuitively) that images always show things
rather than tell them, and also that if they represent three-dimensional
space, they conventionally do this from a particular viewpoint.
There are some important exceptions to the general rule of the de
termined perspective, especially graphs and symbolic images, non-
perspectival or multi-perspectival images, or polyphasic figurations that
are used in comics. The non-perspectival images are ‘pure’ surface images
that refuse the window-effect of a built-in perspective and the iconic ex
pectation of ‘pointing to something’. Such images, include, for instance,
pictures that are blank panels or frames that include mere w riting or
speech and thought balloons, symbolic, abstract, or conceptual im
ages, or scientific and technical (‘objective’) pictures that neutralise the
viewer’s spatial perspective. The latter category would comprise, for in
stance, maps, diagrams, charts, and geometric shapes. The case of the
polyphasic image, where one picture juxtaposes or superimposes differ
ent points of movement or time, creating a more or less blurred image
(Kolp 1992, 134–137), does not challenge the rule since it depends on
the norm from which it is a deviation. Typically, in cases of effet Marey,6
the suggested ocular position, as well as the image frame, remains rela
tively stable, whereas the focus in the image undergoes changes in time,
reflecting various stages of a process. By showing that the focus of per
ception is more or less undetermined, such images accentuate, as in the
fight scene between Tintin and Doctor Müller in L’ile noire (The Black
Island), a particularly hectic temporality. Similarly, a layered page or a
double spread, which creates an effect of an ambiguous relationship be
tween the frames and their backdrop, can manipulate the default mode
of viewing from a determined point in space. Typically, the technique
prompts the viewer to imagine different interrelated viewpoints. This is
what happens, for instance, with what Groensteen has called a multi-
couche in shōjo manga (2011, 66–67), a multi-layered page layout typi
cal of this Japanese genre that is especially targeted to girls and women.
The same principle applies to split panels that juxtapose different points
of perception and fields of vision.
While such exceptions are easily incorporated into comics storytelling,
they usually tend to have limited and strategic functions, i.e. they remain
localised instances. Many of them also depend on the general rule of
the determined perspective. Typically, multi-perspective images, such as
blurred or superimposed images, acquire meaning as perspectival distor
tions in relation to the expectation of a fixed viewpoint. Moreover, their
visual form and frame may carry information about a particular (expe
riential, conceptual, stylistic, or other) viewpoint related to the other
images around them.
Focalisation and Ocularisation

When a character in narrative fiction, regardless of the medium, per
ceives something, the object of that perception is regularly given and
understood in the context of that character’s thoughts, emotions, and
experience. Thus, the distinction between perceptual and cognitive fo
calisation can be quite artificial. At the same time, the distinction bet
ween perceptual and cognitive focalisation, or perception and thinking
(or knowing), is commonly made in film narratology, where it has been
argued that in film narratives, the point of perception and the cogni
tive perspective (or attitude) do not have any obligatory correlation. The
argument is not only medium-specific, but it also carries significance
for other forms of visual storytelling. François Jost, drawing attention
to how Genette’s focalisation concept refers simultaneously to the act
of perceiving and the acts of thinking and knowing (1983, 195; 1987,
21–22), has pointed out that this confusion becomes particularly prob
lematic in the analysis of film narratives, where not all uses of the terms
‘perspective’ or ‘focus of perception’ are metaphorical.7
In films and comics, we need to come to terms with the fact that the
image has a certain deictic value in itself (as pointing to something),
and that the point of view is their inherent meaning-making mechanism.
The representation of perception in visual narratives must also be con
ceived differently from those in literary narratives due to the functions
of showing, including the showing of deictic relations in space, the literal
spatial positioning of the focus of perception, and the spatial impact of
the frame. In films, as Jost argues, perceptual and cognitive focalisation
can have clearly separate functions, expressing quite different things
(2004, 78). While in literature, we may need to distinguish between a
narrator (who speaks) and a focaliser (who perceives), films may prompt
us to further distinguish between perceptual (ocular position) and cogni
tive orientation. In films, Jost points out, “it is possible to show some
one or something and at the same time express something completely
different through the voice” (2004, 73). At the same time, it is possible
for the perspective in films to be simultaneously ‘inside’ and ‘outside’ a
character, i.e. to see a character from the outside while the perspective
is also to some extent tied to the character’s perception and experience.
Jost thus distinguishes perceptual focalisation (what he calls “ocular
isation”), meaning the “relation between what the camera shows and
what the characters are presumed to be seeing”, from cognitive focali
sation, or what he calls ‘focalisation’, designating “the cognitive point
of view adopted by the narrative” (2004, 74). Similarly, despite the ab
sence of a camera in comics, the manipulation of the relation between
what the image shows (from some perspective) and what some character
supposedly sees is an essential means through which comics can create
a sense of a subjective vision. Not surprisingly, then, Jost’s distinction
has been increasingly applied with varying degrees of theoretical rigour
and clarity to comics and graphic narratives.8 The same premises about
perspective-taking in cinema apply to comics storytelling, with some ob
vious differences of technique, such as camera movement, or the means
of aural perspective in films.
However, it is important to specify that Jost does not define focalisa
tion (as distinct from ocularisation) purely in cognitive terms, but claims
that the distinction enables us to focus on a relation between perception
and knowledge. Focalisation, he argues, is “a complex product of what
one sees, what the character is presumed to be seeing, what he or she
is presumed to know, what he or she says, and so forth” (2004, 74).
In other words, in Jost’s definition, focalisation concerns the relation
between the ocular position that the image postulates, the visual field in
the image, and the character’s speech, thoughts, and presumed know
ledge, while ocularisation refers to the visual representation of percep
tion in a literal sense. The analysis of focalisation thus addresses the
way in which cinema works in several registers at once, combining the
visual, aural, and verbal narrative tracks, to give the reader an illusion
of an individual mind, while ocularisation characterizes visual informa
tion about the point and focus of perception only (and auricularisation
describes the centre of auditory perception).
The ocular position is a prerequisite for the impression of depth in
the image. Ocularisation, however, does not have to belong to any per
son or instance in the storyworld or in the frame narrative; it can re
main fully impersonal and hypothetical. Thus, Jost’s category of “zero
ocularisation” does not refer to a missing point of perception, but to a
broad range of perspectival options where the focalizer remains imper
sonal. Equally, the distinction between external and internal ocularisa
tion is potentially misleading in comics because purely internal or purely
external points of perception are usually only short instances. The cru
cial distinction, then, is not between external, internal, or zero-focal
positions, but how and to what extent the image, and what is seen in
the image, is subjectified by narrative conventions and context and as
sociated with the perception and consciousness of a character in the
storyworld.
Jost’s categories of narrative perspective in films are useful in the ana
lysis of visual focalisation in comics given that we do not make the dis
tinction between perceptual and cognitive focalisation mechanistically.
I see three main reasons for developing the dimension of perceptual fo
calisation in comics narratology. One reason for this is that the (narrato
logical) functions of the point and focus of perception in comics, and the
visual encoding of narrative perspective in this medium, are only par
tially understood at present. Therefore, we must try to take the question
of the perceptual and, more precisely, optical dimension of focalisation
in comics to its outer limit—to see what can be said about the specificity
of comics in terms of narrative perspective—before a more holistic the
ory of perspective can truly be developed. Second, in a medium that is
much less reliant on narrator figures than literary prose fiction—even in
first-person graphic narratives—it is important to thoroughly investigate
all relevant aspects of visual narrative mediation, including not just the
point and focus of perception, but the functions of graphic showing and
style, to better understand how minds and worlds are created in comics.
A third reason for focussing on ocularisation in its own right is that
it allows us to recognize some of the formal options and narrative de
vices that are available in comics to present the characters’ minds. The
manipulation of the ocular position in comics can create various kinds
of effects that may be important for the understanding and enjoyment
of the story. For instance, this can contribute to an effect of ambiguous
subjectivity and communal perceptual focalisation that would be diffi
cult to imagine in literary narratives in quite the same way. In M atthieu
Bonhomme and Lewis Trondheim’s Omni-visibilis (2010), the whole
storyworld shares the sensory perceptions of the main character-narrator
Hervé, who is an office employee in his thirties. The point of perception
is often the very basis of the parody of continuous webcam presence.
In Omni-visibilis, all people, even from other parts of the world, begin
to see through the protagonist’s eyes when their eyes are closed, hear
what he hears, and feel what he feels. Especially interesting in this light
are the point-of-view images in the story. For instance, when the main
character sees his pursuers through a peephole in the door, this cues the
reader to imagine that the perspective belongs both to the protagonist
and his pursuers. Furthermore, it is implied in these images that the peo
ple whom the readers see with the main character’s eyes can also observe
themselves from the same viewpoint. Thus, the experience of communal
visual perception reaches new heights of ambiguity. The verbal narra
tive track is crucial to knowing what exactly is happening, but without
the skilful manipulation of the ocular position, the whole point of the
parody of webcam presence and surveillance, and paranoid reactions to
webcams, might be lost.
In films, as Jost points out, verbal anchoring can easily transform in
ternal into external ocularisation or vice versa (1983, 196). The same is
true of comics. However, in a wordless comic such as Tommi Musturi’s
Walking with Samuel, shifts in perceptual focalisation can be detected
by visual clues alone. In this story, the reader is suddenly invited to share
a perspective that appears to belong to a bird that pecks the protagonist’s
eye. In order to understand the viewing position, the reader must compare
this image with the surrounding sequence of images and the respective
perspectives they offer. This wordless comic underscores the important
role played by following image-to-image and scene-to-scene transitions
in the representation of perspective, but, equally and efficiently, under
mines the degree of deixis in the image, that is, the distinction between a
character’s consciousness and the rest of the fictional world.
I would like to conclude this section with the idea that Jost’s ocularisa
tion concept itself is not necessary in comics narratology,9 but that a sys
tematic assessment of the distinction between perceptual and cognitive
focalisation can improve our understanding of the multimodal dimen
sions of focalisation in comics and graphic novels. Such an investigation
involves the study of the relation between the point of perception and
what is seen in the image, the interaction between perceptual and cog
nitive focalisation in a sequence of images, and the relationship between
visual and verbal focalisation. In other words, for the theory of narrative
perspective to develop in comics studies, it is imperative that we inves
tigate how visual perspectives are created and manipulated so that they
become associated with a certain subjective vision or cognitive attitude.
Equally, this requires that the criteria that help us determine how and
to what extent something in the image is subjectified should be defined.
Simultaneity of Different Focalisers Inside

and Outside the Picture Frame
All focalisation, be it in a literary or visual narrative, is typically vari
able. Whether in literature, films, or comics, perspectives vary between
more or less subjective and objective viewpoints, or between different
narrative levels, and can shift between different perceiving characters,
or between narrators and characters. While proposing his original typo
logies and distinctions, Genette was fully aware that it is difficult to find
pure examples for his three basic categories of focalisation: external,
internal, or zero focalisation (1988, 73–74).10 Yet, in visual and multi
modal narratives such as comics, the variety of forms of focalisation
seems even greater. We need not only a revision of these categories, but
the creation of more flexible ones.
As can be easily shown in film narratives, several internal and external
focalisers can appear simultaneously at different points inside and out
side the image frame. In his analysis of classical films, Celestino Deleyto
has analysed the tendency to use external perspective to make the inter
nal gaze understandable (1996, 223, 226). Similar practices and effects
are known to us from picture books, where it is a common convention
to combine first-person verbal narrative with uninvolved third-person
visual perspective.
While different internal and external focalisers can appear simultane
ously at different points inside and outside the picture frame, this capa
city can also set up tensions between the character’s perception and the
field of vision on the one hand, and between visual and verbal narration
on the other. Playing with the divergence and convergence between visual
and verbal perspectives is a common practice in autobiographical and
other kinds of nonfiction graphic narratives. Typically, contemporary
first-person graphic novels, such as Taniguchi’s A Distant Neighbour-
hood, create a meaningful contrast between the narrative ‘voiceover’
and the narrator who is seen in the image from the outside. While the
‘voiceover’ narration often offers a clarifying perspective on the contents
of the image, indicating what is important and salient in the things that
are seen, the words may also maintain an ambivalent relation to the
space of the image. This is different from literary narratives, in which
the ability to present characters at the same time from the outside and
from within is usually another indicator of the fictionality of the world,
and the notion of simultaneity in this respect is more metaphorical.
To illustrate these points about multiple visual perspectives in comics
storytelling, I would like to briefly discuss one specific device for multi
plying perspective, which is the use of embedded photographs in the panel
sequence. In the following three examples, the photograph, whether a real
photograph incorporated in the story or a graphic rendition of an imag
inary photograph, draws attention to the choice of perspective, and the
use of juxtaposed perspectives, by multiplying points of view. In thus re
mediating photographs, narrative comics further pose the question of the
specificity of their art form for representing visual experience. Here, com
ics, so to say, think with their own medium by way of the other medium.
The Australian artist Shaun Tan’s graphic novel about immigration,
the prize-winning The Arrival (2006), is a story with wordless images.
What I would like to highlight in this work are some of the functions
of the graphically drawn family photo portrait that we see at various
moments of the story. The portrait is pictured at the beginning of the
narrative and at its end, in addition to when the father of the family is
shown looking at it after being separated from his wife and daughter
in a strange new land. First of all, it is significant that this image draws
our attention to the characters, who gaze directly at us, and furthermore
that this exchange of gazes is seen through the father’s eyes. The sub
jective aspect of the viewpoint is revealed by the father’s hand, which
we see taking the picture from the shelf and packing it in his luggage.
Besides the subjectivity of the gaze, the introduction of the portrait re
veals the emotional intensity that accompanies all the later viewings of
the image. Another important aspect of the viewing is that it emphasises
the impression of a photographic frame, that is, the sense in which all the
panel images in Tan’s book more or less resemble old black-and-white
photographs. The effect is even more prominent in the father’s passport
photo and the worn-out panel frames in some of the embedded stories,
which recall an old photo album.
Jarmo Mäkilä’s fictional portrait of an artist, Taxi van Goghin k orvaan
(A Taxi into Van Gogh’s Ear, 2008), the second part in a trilogy by this
Finnish painter, is a complex mixture of childhood memories, hallucina
tions, and a Dantesque journey into the world of the dead. The main char
acter is split into different personalities: a lonely boy, a clown, and an adult
man, the artist called Itikka (meaning ‘bug’ or ‘mosquito’), who all seem
to live in two different realities at the same time, the world of the living
and the world of the dead. At one moment in the narrative, the artist is lost
somewhere in a forest where he finds a photograph in his pocket depicting
himself as a young boy with his parents and a girl he loved ( Figure 6.1).
The incorporation of a graphic version of this photograph again multiplies
perspectives by introducing subjectivity into the sequence, and underlines
the interpenetration of different levels of reality and memory. The effect
continues in the next panel, spread over a whole page. Here, we are sud
denly taken, as if momentarily sharing the main character’s gaze, into
another world where we meet the artist drifting in space amidst toys on
a plastic rowing boat. The multiplication of perspectives leads us to other
levels of experience, contrasted with the verbal focalisation of the story,
which gradually moves from first- to third-person narration.
Finally, Emmanuel Guibert and Frédéric Lemercier’s travel book
trilogy Le photographe (2003–2006, translated as The Photographer:
Into War-Torn Afghanistan with Doctors Without Borders) raises quite
explicit questions about media hybridity and the realism of the photo
graph. The Photographer is based on a true story written by the French
photographer Didier Lefèvre about a Doctors Without Borders mission
in Afghanistan in the late 1980s. Almost every page and spread of this
story includes Lefèvre’s actual photographs from the dangerous journey,
and sometimes a whole page or a double spread is composed just from
them. The graphic image and the photograph have pretty much equal
weight throughout the book (Figure 6.2).
Figure 6.1 Jarmo Mäkilä. Taxi van Goghin korvaan (2008) © Jarmo Mäkilä.
Figure 6.2 Emmanuel Guibert, Frédéric Lemercier, and Didier Lefèvre. The
Photographer © 2009 by First Second.
What interests me in this last case is not so much the remediation
of the photograph as part of comics storytelling—a process by which
comics, as in the two previous examples, refashion photographs, break
them into a story and thus, in a sense, try to improve upon them—but
a more balanced dialogue between the media, where photographs have
an impact on the graphic images, and vice versa. On the one hand, the
various instant moments that are captured by the camera and included
in Le photographe are given narrative shape and meaning by the graphic
images and the first person verbal narration that accompany and sur
round them. The photographer, his medium, and his equipment are
literally made part of the story. On the other hand, the integration of
photographs has an obvious, many-sided impact on the graphic images
and their sense of veracity. This is not so much that the photographs give
an air of authenticity to the graphic panels even if this might happen as
well, but that they make the drawn images look more like photographs,
like an instant shot.
The effect of this intermedial composition may perhaps appear awk
ward to eyes unaccustomed to it, but the alternation between the media
is itself also worth attention. Neither medium serves as the other’s relay
or amplifier, but both move the story forward. Sometimes one of them
dominates a single page, a double spread, or an episode. The photo
graphic images are interwoven with the graphic panels in the sequen
tial order of the narrative, while they also preserve something of their
status as individual images illuminating an instant. The photos enjoy a
certain distance from the verbal ‘voice-over’ in the narrative boxes. The
same is not afforded to the drawn images. While verbal narration often
accompanies the photographs, the photographic images never include
dialogue, and the verbal narration next to them remains carefully sepa
rated, not part of the frame.
The question of veracity is further complicated by the graphic nature
of the images involving real objects and people. The reality of the point
of observation and the field of vision in comics storytelling, even if it is
nonfiction, is necessarily make-believe. No one ever saw the world as
drawn, or as caricature, no matter how detailed or instantaneous the
graphic line may be. Naturally, no one sees the world as photographs
either, but there is a necessary graphic distance from the world that
is represented. As happens in Le photographe, the distance between
different kinds of image and the perspectives they offer can provide
nonfiction with particular self-critical potential, pointing out that the
reality in pictures is always mediated. In Le photographe, the hybrid
ity and the multiplication of visual points of view highlight the mean
ing of the perspective both for narrative organisation and for the sense
of reality.
Scale of (Visual) Focal Points between Internal
and Non-Character-Bound Positions
In visual storytelling, in general, impersonal focalisation is a much more
prominent and broader phenomenon than in literary narratives.11 Visual
narration across media, from films to photo stories and picture books,
can deploy a broad scale of techniques between more or less personal or
impersonal perspectives, from panoramic views that no person could see
to partly sharing subjective vision and to a point-of-view image. Com
ics storytelling, when it comes to the visual perspective of the narra
tive, uses an extremely complex scale of potential intermediate positions
between subjective or internal focalisation at one end, and clearly non-
character-bound perspective or external focalisation at the other. This
scale, however, is not organised in a strict external/internal, or reflec
tor character/narrator binary, as may be the case in literary narratives.
Rather, it reflects varying degrees of congruence and divergence between
a character’s point of view and the reference world of the narrative, as
well as the fact that comics can use internal and external viewpoints at
the same time. In a great number of externally focalised images, no sup
position of an individualised perception can be made. For these panels,
we may postulate some anonymous agent, a non-character-bound ex
ternal focaliser. What makes subjectivity particularly flexible in comics
is also the capacity to alter the perspective from one panel to another.
However, contextual information, based on our evaluation of the nar
rative sequence of panels, the layout, or the relationship between words
and images, becomes important especially when the degree of subjectiv
ity changes from one panel to another. Only contextual information of
a larger narrative sequence may allow us to determine whether a certain
perspective moves towards or away from subjective perception.
The techniques for subjectifying narrative perspective in comics in
clude different formal choices with regard to the position, angle, field
(or scope), distance from the regarded object(s), depth, and focus of
vision—the latter involving, for instance, foreground-background rela
tions, level of specificity, and detail.12 The most common techniques of
subjective focus of perception, in comics as in film, comprise the various
ways in which the character’s positioning in a given image or sequence—
in relation to the frame and what is shown in the images—suggests a
subjective narrative perspective. Such techniques comprise, for instance,
the point-of-view (POV) image (the impression that the reader shares the
field of vision with a particular character), the positioning of the view
point on the eye level of the characters (without implying an actual POV
image), the over-the-shoulder image, the gaze image (showing a charac
ter looking at something), the eyeline image/match cut (a combination of
a gaze image that is preceded or followed by a point-of-view image), and
the reaction image (a character reacting to what he has just seen).
Equally, subjective vision can be suggested by the exaggeration of the
foreground of the image, by particular optical effects, such as an out-of-
focus image that is related to someone’s mental condition, and that can
be used to create a sense of a perception image,13 as well as by visual
effects or signs of the observer’s physical presence, such as a shadow, a
speech balloon, or parts of the body, positioned in the foreground and
close to the image frame. All these effects can be accentuated by the con
text of other panels and the page layout, where a field of vision may be
connected to someone looking or speaking. All “objective” or descrip
tive images may also be subjectified or have this effect when juxtaposed
with a gaze image, a reaction image, or other clear indications of gaze
and subjectivity in the panel sequence.
To better describe this multitude of positions classifiable as ‘vision
with’ or ‘vision from behind’, we can first point out that comics story
telling has at its disposal most of the cinematic techniques for getting
close to a specific subjective point of view, from suggesting a subjec
tive perspective to adopting and wholly assuming it. We can use as our
starting point Mandfred Jahn’s (2003) list of the five most important
subjectif ying filmic devices. The typology, however, requires us to
change the filmic term ‘shot’ into ‘panel’ or ‘image’ and make some fur
ther medium-specific adjustments. It needs to be stressed that the equa
tion between a shot and a graphic image in comics is not unproblematic
since a shot is a sequence of frames14 and a panel is (mainly) a single im
age. Moreover, the perspectival function of a single image in comics may
often be specified only in the context of other images. The categories,
however, are illustrative of the general options in this regard:
1 the point-of-view image (or sequence) (POV)

2 the gaze image
3 the eyeline image/match cut
4 the over-the-shoulder image
5 the reaction image.
First, the point-of-view image (or sequence) (POV) is the most internal,
direct, and subjective perspective. It assumes the viewer’s position; the
image frame functions as the representation of someone’s gaze and a
field of vision. Yet we must also add that the presumed subjectivity of a
POV image always involves an interpretive move and potential ambigu
ity: how do we know that a certain perspective in a panel or an image
sequence belongs to someone? One clear cue for this is the positioning of
the viewpoint on the eye level of the characters even if it is insufficient in
itself to imply a POV image. In comics, more context can be given in a
sequence or other group of panels, including for instance a gaze image or
indications about the viewer’s position and relation to the field of vision.
The technique, however, also cannot be used extensively for obvious
reasons, as studies of ‘subjective camera’ in film or the picture book have
shown. The main difficulty is that the character whose perception we
share could never appear in any picture except in a mirror or a reflection.
What film studies calls the perception shot is also always possible in
comics. This is a specific type of POV shot that, as Branigan defines it
(1984, 81), reveals the mental condition of someone looking at something.
A perception image in comics most often shows difficulty in seeing or the
viewer’s heightened attention, typically a blurred scene that presents the
vision of a drunkard or someone who is fatigued. Another option in this
respect is the highlighting of something in the image by graphic means as
the focus of a character’s attention. This is realised, for instance, in a se
quence of two pages in Scott McCloud’s The Sculptor, where the protago
nist, David Smith’s focus of attention on a young woman is indicated by
two visual means that operate together: the POV image, where the readers
share David’s perspective, and the colour contrast of the female figure’s
bold black colour against the faint blue-toned grey background scene.
Moreover, the way that something like a city scene or a landscape is
visually rendered, by using conventional implications of colours, vary
ing intensities of lines and shading, or patterns of shapes, can suggest
a subjective viewpoint. In this respect, there may be great differences
between different visual media in terms of technique even if the basic
function of the perception image, sequence, or shot is similar. The tone
of a film narrative, for instance, is the result of a wide variety of stylistic
choices encompassing lighting, cinematography, mise en scène, and the
editing of both the image and sound tracks, while cartoonists can use
the panel setup and page layout, the frame shape and size, colouring,
and the graphic line for similar purposes.
Second, the gaze image is a picture of a character looking at something.
More precisely, this is an image with external perspective that shows a
character looking at something that cannot be seen in the same image, thus
drawing our attention to perception. Generally speaking, in viewing visual
stories we continually make inferences from people’s looks, gazes, glances,
and facial expressions so as to have access to their subjective states even if
these states may remain fairly indeterminate. Another important category
of looking and gazing in comics is the direct gaze, meaning participants
in pictures who look directly at the viewer. We could recall here the theo
retical division into ‘demand’ and ‘offer’ images, as defined by Kress and
van Leeuwen (2006, 126–127). Realising a visual ‘you’ in this way, the
‘demand’ picture suggests a particular, often genre-specific type of interac
tion between the picture and the viewer. The technique can also increase
narrativity by suggesting a heightened level of involvement for the reader.
Third, the eyeline image/match cut is a combination of a gaze shot
that is followed by a POV image. Thus, the gaze image cues the audience
into interpreting the preceding or the following image as a POV image.
The use of image/reverse image technique is also common, as we will see
later in this section of the book, in dialogue scenes. Another form of this
technique is the impression of a collective vision in which the perspective
is not mediated through any one character, but rather falls within the
perceptual range of several characters whose gazes and objects of vision
alternate in the given scene.15
Fourth, the over-the-shoulder image is a less direct and less internal
means of subjectification than the point-of-view image. In a film, this
means that the camera gets close to, but not fully into the viewing po
sition, or that a camera follows the movements of some character. In
comics, the image may be shown, for instance, from behind a character’s
back, close to the character’s viewing position, or in conformity to the
character’s direction of looking. A sequence of images may also follow
a character’s action or movement so closely as to suggest that the story
follows not just the character’s action and movement, but also their per
spective. Another similar device is the placement of the character in the
image in such a way that his or her position can subjectify the perspec
tive, for instance, when the character’s back or side profile is placed be
side the edge of the image frame so as to heighten the reader’s association
with the character’s perspective. Likewise, the depiction of the viewer’s
hands or lower body beside the image frame, or his or her image in a
mirror, also points out that we share his or her perspective. The drawing
of a character’s fingers beside the panel frame, as in the example above
from The Arrival, is another pictorial convention that reveals the subjec
tive perspective of the image.
Finally, the fifth subjectifying device is the reaction image, which
shows a character reacting to what s/he has just seen. Similar to the gaze
image, the reaction image draws our attention to perception, but does so
retrospectively, after the act of viewing or perception.
All these techniques and devices amount to an extremely complex
scale of intermediate positions between clearly subjective and clearly
non-character-bound perspectives. One important conclusion to draw
from this is the central role of ambivalent focalisation in the medium.
This could also be characterised as the predominance of ‘free indirect
perception’. The category of ‘free indirect perception’ originally refers to
a kind of ‘narrated perception’ in literary discourse, which occurs when
the narrative clearly describes or implies the perception of a character.
A character’s mind is thus implied as the perceptual angle of some tex
tual passage, but his or her perceptions are never directly introduced by
perception verbs or other linguistic means.16
In films and comics alike, similar effects are commonplace. Charles
Forceville, for instance, has suggested that studying techniques like
character-bound camera movement in terms of free indirect discourse,
specifically when such techniques create ambiguity between the external
narrative instance and the character, could contribute to a transmedial
narratology (2002, 133). The free indirect instances in films, which take
place when audiovisual information is shared between the character and
the overall narrative frame, also involve passages when certain things are
distilled through a character’s perception by means of colour filters, visual
distortions, and other such techniques. In comics, as we have already seen
above, there are a number of ways by which the narrative can focus on
the character’s field of vision and make us systematically look either with
him or her or look from behind him or her. These techniques may allow
the reader to look deeply into the character’s field of vision, even limiting
the view to the range of perceptions available to some character, while
at the same time retaining the sense of a hypothetical viewing position
that does not belong to any character. In comics, the spatially determined
viewpoint obviously cannot be associated, as it is in narrative film, with
the movements of a camera, but this does not make it more personal as
such. We watch along with a character, from the character’s back, or gain
insight into his or her point of view through what we see in other ways,
but focalisation is usually not entirely left to the person, or any person.
To briefly return to Shaun Tan’s The Arrival, we can detect here a great
number of images that imply a strong subjective angle even if there are rel
atively few clear cases of a POV image or sequence. Consider the end of the
episode where the main character, who at this point is still living alone in the
foreign country, is invited to have dinner at the home of another immigrant
family. In one large image that spreads over a page, we see a scene around
the dinner table (Figure 6.3). The following twelve close-up images consist
of a series of eyeline matches, leading us from one subjectified field of vision
to another, motivated by the direction of the character’s gaze in the previ
ous panel, and at the same time restricted to what these characters may see
from their respective positions. This creates the effect of intertwined looks
around the table that reveal to the viewer things happening from different
angles, as if we could alternate between each person’s eyes. Yet, while the
lonely father’s subjective v ision is at times strongly implied, especially at the
end of the scene when the couple looks directly at him and us, we also see
him from the outside. We also observe the other people from angles that
could belong to any of the characters present or to no one in particular.
Figure 6.3 E mmanuel Guibert, Frédéric Lemercier, Didier Lefèvre, The Pho
tographer, 2003. © Emmanuel Guibert, Frédéric Lemercier, Didier
Lefèvre.
It does not seem necessary to postulate one particular subjective view
point in order to understand this scene. Instead, what matters, and what,
to some extent, subjectifies all the views at the dinner, is the restriction
of visual information to the immediate views of people around the table.
This limitation of the perspective, combining a possible gaze image with
a potential POV image in a sequence, emphasises the joyful atmosphere
and the feeling of sympathy among the members of the party. Instead
of any fully subjective, personalised vision, the readers have access to
many subjective views. Besides, we may also have access to what literary
narratology sometimes calls the empty deictic centre: a position that is
clearly on the scene with the characters, but without the possibility or
even the need to identify with any of them.
One explanation for this vast scale of varied and combinable options in
focalisation is, to follow François Jost (1983, 195), the difficulty in desig
nating the degree of deixis in a single image, even when we are dealing with
cases of subjective perspective. The non-character-bound perspective, gen
erally speaking, is coded for transparency: while showing and framing a
field of vision, it does not necessarily presuppose a human narrator or a re
porter. Depending on the narrative context, however, the ‘objective’ image
or sequence can also be marked by a character’s or a narrator’s subjective
field of vision and perception, or encompass it. The default mode for visual
focalisation in general seems to be an external viewpoint subjectivised to
varying degrees by various visual, stylistic, and compositional techniques.
Another conclusion that we can draw from the importance of the vast
range between impersonal and subjective forms of visual focalisation is
the way that the intermediate positions may heighten the play of diver
gence and convergence between words and images. These techniques
enable comics storytelling to fully exploit the distance between a self
who speaks, a self who sees, and a self who is seen, or the split between
a narrating and an experiencing self.
Exempting wordless comics, the effects of all visual techniques of sub
jectification are accentuated, complemented, and sometimes contrasted by
verbal narration. Thus, to understand the degree of subjectivity in focal
isation in comics, it is necessary for the reader to process the interaction
between ‘focalisation markers’ at these two levels of narration, the verbal
and the visual dimension of narrative representation. Verbal narration may
also complicate the establishment of perceptual focalisation. Particularly
in cases of split verbal focalisation, in which the same person (or character
of fiction) speaks and narrates simultaneously outside and inside the im
age, the relationship between the visual and the verbal narrative perspec
tives can become quite complex and dynamic. This is usual in first-person
literary narratives, but what is specific to comics in this regard is that the
two modes of telling need to be reconciled with two modes of showing:
the narrator shown in action and the narrator shown as a narrator, or as
someone who is aware of his or her drawn self and role as a narrator.17
Embedded Focalisation
The distinction between the source of focalisation and the focalised en
tity is often relaxed, or remains ambiguous, in the technique of embed
ded focalisation, which takes similar yet ultimately different forms in
comics (and visual storytelling) compared with literary narratives. What
Mieke Bal calls an embedding of focalisations involves transference bet
ween subject and object positions along different diegetic levels in a nar
rative, for instance when some narrative object (the object of narrating,
focalising, or acting) becomes the subject (the narrator, the focaliser,
or the principal actor) of the following level (Bal 1981, 45). Another
possibility of the same phenomenon is that the external focaliser in a
narrative watches “along with a person” but without leaving focalisation
entirely to this character (Bal 1997, 159).18 In visual terms, this may be
likened to the many techniques that subjectify the field of vision without
turning the image into a pure subjective image, as discussed above.
It is difficult to find any literary equivalent for the techniques of per
spective filtering in Calvin and Hobbes, where the mere presence of some
figures in the scene may indicate a change in perspective. Most impor
tantly, when only Calvin and Hobbes are present in the image, Hobbes
is not drawn as a stuffed animal but as a living being, an anthropomor
phised tiger. This is one of several sophisticated means of focalisation in
this series, and not only different in degree from the techniques used in
literary narratives. The effect is based, however, not so much on infer
ring how graphic style reveals a narratorial or authorial perspective that
might embed other perspectives, but on visual information about the
supposed focaliser in the image. The child centre of attention (Calvin)
is simultaneously often shown from the outside. Hobbes is depicted in
a particular way in particular images—either alive or not alive—since
we have come to learn that certain people, whom we also see present in
the image (Calvin or his parents or someone else), perceive Hobbes in
this way. What Calvin and Hobbes shows us so well is that, in visual
narratives such as comics, internal and external focalisers can appear
simultaneously, embedded in a literal sense, at different points inside and
outside the image frame.
Conclusion
In order to develop a narratological understanding of perspective in com
ics, it is important not only to find common ground between different
narrative media with regard to their techniques of perspective-taking,
but also to develop ways to explain the fundamental differences between
different narrative media in these techniques. Emphasising the issue of
perceptual focalisation, I have posed a number of medium-specific ques
tions that should be accounted for in order for us to conceive a more
comprehensive narratology of perspective and narrative mediation in
comics. These include, in particular, the problems of epistemic access to
the point of perception and of what is perceived, pertaining to the spa
tially determined point of perspective and the evaluation of the degree
of deixis and subjectivity in graphic images, as well as the difficulty of
distinguishing between the focaliser and the focalised in some cases, and
the implications of that difficulty for the theory of focalisation.
These are not the only important questions to ask about the possibili
ties of analysing focalisation in narrative comics, but they are ones that
are of great value for narrative theory that seeks to be relevant in this
field. In the narratology of comics, problems will arise if we assume that
notions of narrative perspective, transmission, and mediacy can be
transferred from one medium to another without due modification. Yet,
we may also confront problems if we are unwilling to see some of the
crucial similarities that exist across narrative media. The latter include,
for instance, the importance of a situated focus, spatial representation
and perceptual information, the relation between perception and cogni
tion in the representation of fictional characters’ minds, and the variable
quality of types of focalisation in narratives across the media.
Notes
1 Indications of the viewer’s identity, location, and time of viewing within the
given image or sequence. In linguistics, deictics are most commonly defined
as linguistic expressions, such as personal and demonstrative pronouns,
whose referent has to be found in the situation relative to their act of utter
ance. See, for instance, Lyons (1977, 636–690).
2 See Bal (1997, 143).
3 Genette’s conclusion, however, disagrees with those views that emphasise the
necessity of spatio-temporal positioning in all narratives, such as Bortolussi
and Dixon’s psychological viewpoint (2003, 166–178). For Genette, focali
sation is merely a device that may or may not be employed in literary narra
tives, and the question of the literal location of the point of view is often of
negligible importance (1988, 76).
4 See, for instance, Deleyto (1996, 222) and Gaut (2004, 247–248), on what
he calls the intrinsic perspective of the film image.
5 See also Jost (1983, 194–195) on this question in film narration.
6 The Marey effect is named after the French physicist Étienne-Jules Marey’s
(1830–1904) chronophotography. Marey photographed the movements of
men and animals several times in a second with his chronophotographic
gun, thus allowing the decomposition of movement in elementary phases
(recorded on the same image or in several frames of print).
7 Genette’s original formulation does not distinguish between focalisation as
the “focus of perception” or focalisation as the “selection of narrative
information” (1988, 64, 74). For a discussion of this fundamental contradic
tion, see Jesch and Stein (2009). In comics studies, Badman (2010) follows
Shlomith Rimmon-Kenan’s distinction between perceptual, psychological,
and ideological orientation; Horstkotte and Pedri’s definition is similar
(2011, 331).
8 See Miller (2007, 106, 109), Badman (2010), Groensteen (2011), and M ikkonen
(2013).
9 Groensteen prefers the notion foyer perceptif to ocularisation so as to avoid
a too-close association with cinematic devices such as the viewfinder, the
objective, and the camera eye (2011, 90–91). However, ‘focus of perception’
or ‘perceptual focus’, which are possible translations of ‘foyer perceptif’, are
too limiting for my purposes. Bal’s definition of focalisation as “the relation
between the vision and that which is ‘seen,’ perceived” (1997, 142, emphasis
added), is more encompassing.
10 As William Nelles points out (1990, 371), Genette has explicitly recognized
this difficulty in regard to the category of zero (free) focalisation. He writes,
“the right formula would be: zero focalisation = variable, and sometimes
zero, focalisation. Here as elsewhere, the choice is purely operational”
(1988, 74).
11 Much of the critique of anthropomorphism in focalisation theory, meaning
the assumption that the focaliser would always be a person who perceives,
has been based on the insight that we need to pay more attention to tech
niques of impersonal perspectives.
12 Corresponding in large part to what cognitive linguists call perspective-
related parameters in narrative contexts and that include: the location of
a perspective point with a “referent scene”, the distance of a perspective
point from the regarded scene, the perspectival mode, and the direction of
viewing. See, for instance, Talmy (2000, 311–344), and also discussed in
Herman (2009b, 128–132).
13 Branigan distinguishes the perception shot from the point-of-view image,
suggesting that the former includes an indication of a character’s mental
condition, “a signifier of mental condition has been added to an optical
POV” (1984, 80).
14 A shot “is a sequence of frames filmed in a continuous (uninterrupted) take
of a camera” (Jahn 2003).
15 See also my discussion of a passage in Daniel Clowes’s Ghost World from
this perspective in Mikkonen (2011b, 647–650).
16 See Chatman (1978, 204) and Mikkonen (2008, 301–321).
17 The insufficiency of the distinction between narration and focalisation in
film narratology is also reflected in the need to add a third component to the
model. Branigan has argued that there are three distinct types of narration
in films—narration, action, and focalisation—that describe “how know
ledge may be stated, or obtained” (1992, 105).
18 However, Genette rejects the notion of embedded focalisation in literature,
claiming that the focus of the narrative cannot be at two points simultane
ously, even if a narrative mentions that a glance “perceives another glance”
(1988, 77). What Bal defines as embedded focalisation, Genette would call
displacement of focus (1988, 76). See also my critique of O’Neill’s trans
medial version of the notion of embedded focalisation (Mikkonen 2011b).
7 Characterisation in Comics
How I want thee, humorous Hogarth!

Thou, I hear, a pleasant Rogue art.
Were but you and I acquainted,
Every Monster should be painted:
You should try your graving Tools
On this odious Group of Fools;
Draw the Beasts as I describe them:
Form their Features while I gibe ‘em;
Draw them like; for I assure you,
You will need no Car’catura;
Draw them so that we may trace
All the Soul in every Face.
—Jonathan Swift, “A Character, Panegyric, and Description
of the Legion Club” (1736/1998, 229–230)
One indication of the centrality of the character in Western fictional

imagination is the naming of major works according to the protago
nist, from Odysseus and Aeneas to Don Quixote, Tristram Shandy and
Anna Karenina, and from Little Nemo and Adamson to Corto Maltese,
Little Archie, Bamse, and Gaston Lagaffe. These classics are known for
their characters, and the protagonist’s name serves as a kind of read
ing instruction: their behaviour, fate, or experience is the focus of the
story. Another indication of the importance of characters in fiction,
regardless of media, is their semi-independence: we can refer to them
and their unique features independently of the stories where they have
been created. Little Red Riding Hood, Robinson Crusoe, Dracula, Peter
Pan, Sherlock Holmes, Pippi Longstocking, James Bond, Mary Poppins,
Tintin, Asterix, Batman and Superman, and Mickey Mouse and Donald
Duck are not only characters in well-known fairy tales, novels, films,
and comics, but widely known figures in our culture.
Characters are a persistent feature in comics, but are there comics
without characters? In some non-person comics, or it-narratives,1 such
as Intérieurs by Régis Franc (1979), The Short History of America by
Robert Crumb (1979), and The Cage by Martin Vaughn-James (1975), 2
Characterisation in Comics 175
the only focus and ‘engine’ of the story seems to be the changing décor
and the world of places and things held in the field of vision. Yet, the re
jection of characters in these works is not as absolute as may first seem.
The human figures who can be seen as part of the images in Crumb’s
short American history3 are not characters in the story, but indicative of
the signs of human imprint and influence: how people shape the environ
ment over a long period of time. At the same time, a human voice plays a
seminal role in the two other examples: Franc’s and Vaugh-James’s com
ics suggest a character’s presence by verbal narration in the first person.
In fact, in Franc’s Intérieurs, the first-person narration creates a strong
sense of a continuing consciousness frame, accompanied by a subjective
focus of perception, for the entire story. The narrator, who has lived in
the rooms of a small apartment that is the visual focus in the story, gives
meaning to the interiors and objects of the flat through his contempla
tion, memories, and experience. Even if he is a mere voice, he clearly is a
character in his own story.
In it-narratives in comics, there is considerable variety in the way in
which the narrating voice or the perspective can be personified, that
is, attributed personal properties. In Woodrow Phoenix’s Rumble Strip
(2008), where the only human-like characters in the images are the fig
ures in traffic lights and signs, the narrating consciousness may be asso
ciated with the actual author. At the same time, the dynamic narrative
situation of this work complicates our evaluation of the identity of the
narrating voice. Most of the book is told in you-narration, where the
pronoun you hovers between the reference to the actual reader, an im
plied audience and a second-person character. At the end of the book,
by contrast, the narrator shifts from the impersonal mode of argumenta
tion, concerning wasteful road accidents, injuries, and deaths caused by
reckless drivers, to first-person narration by telling a story of how he was
nearly killed in his car on a motorway from London to Brighton. What
challenges the concept of character in this case, however, is not only that
no characters are seen in the images, but also that much of the book is
not framed as a fiction or even as a narrative, despite its story-like situ
ations, scenarios, and case histories of road accidents—Rumble Strip is
rather an argumentation, or an essay, about car culture and road deaths
in the author’s own name. The category of the character tends to con
note fictiveness: characters are an element of a composition; they do not
exist as persons even if they are often person-like.
What is important in all of these four it-narratives (Crumb, Franc,
Vaughn-James, and Phoenix) from the viewpoint of characterisation is
that the spaces that are the focus of the images are strongly marked by
human experience, memory, and involvement. The lyrical captions in
Vaughn-James’s The Cage do not directly refer to any speaking or nar
rating subject even if they clearly suggest that some human consciousness
is filtering the images. The lack of characters functions differently here
from the other examples, in that we can ask if The Cage is a narrative at
all. Might it not be a visual poem, or a lyric comic featuring a series of
tableaux that involve various abandoned spaces only loosely connected
with each other?4 The lyrical captions in the work are strongly indicative
of a human consciousness (or several characters’ minds), but they do not
suggest a continuing consciousness frame that could easily be conceived
in terms of narrative cohesion.
In other comics, buildings, objects, or streets have been conceived as
characters. For instance, the apartment building where the protagonist
of Chris Ware’s Building Stories lives, has thoughts and emotions, and
Danny the Street in Grant Morrison and Brendan McCarthy’s “Doom
Patrol”, an actual street, is a sentient character with superpowers. Sim
ilarly, in Jeanne Puchol’s it-narrative Dessous Troublants (1986), the
furniture shown in a room has a subjective voice that refers to personal
memories and experiences. Also in Wally Wood’s adaptation of Ray
Bradbury’s dystopian short story “There Will Come Soft Rains” (1950),
which describes a series of events that take place in the rooms of an
automated house during a day in 2026, the story does not feature any
humanlike characters. In this case, the narrator’s continuous reporting
voice, which is external to the story world, is remarkably impersonal:
the captions merely describe the evolving events, or indicate the present
time and place of the given panel. Nevertheless, the focus on the auto
mated house that has several humanlike qualities—it has a voice in the
automated walls and some of its appliances, it has a sense of time, it dis
likes dirt, fears fire, and so on—allows the reader to imagine the house
as a potential character who is at least to some degree, no matter how
illusorily, a sentient and anthropomorphic being.
Drawing on these examples of it-narratives, we can identify three
important aspects about characterisation in comics. First of all, it is
difficult to tell a story without characters. A mere voice, a subjective
perspective in the image, a caption, or a place that is followed, may be
perceived as an impression of consciousness and a mental state, which
can then be attributed to a potential character. Second, the function of
voice and perspective as indicators of subjectivity and characterhood
open the question of how to define a character in comics in the first
place. It is important to note how comics vary greatly in this respect.
On the one hand, personal traits can be expressed predominantly by
visual means of external description, or through dialogue, and the rep
resentation of mental states can remain implicit or less relevant. On the
other hand, mental states and verbal narration can also be the main
focus of the story. Third, the limits of the character category in com
ics have, in some cases, less to do with the characters’ visibility and
continuity in the images, or the ambiguity of their consciousness or
physical shape, than the genre in question. Nonfiction comics, as well
as many non-narrative forms of comics, such as lyrical, pedagogical,
and abstract comics, do not necessarily presuppose the category of the
character, despite their possible focus on mental states, consciousness,
and subjective experience.
The Concept of Character

From a narratological perspective, the concept of character is truly
multidimensional: characters in fiction are at once entities of the sto
ryworld, elements of a narrative composition, and a means for telling
the story. The multidimensionality of the concept has posed significant
challenges to narratological theoretisation in the past, reflected in dis
agreements about the meaning and relevance of the category. One reason
for the debate is that the conception of characters in fiction is depen
dent not only on artistic devices of characterisation but also on genre
and medium-specific conventions of storytelling, as well as models of
real-world people and their typical situations. 5 It may then be difficult
to distinguish character as a textual component from person as a kind
of cognitive model.
Earlier in this book, I have discussed characters mainly in terms of
their function in narrative mediation and composition, as important
means of connectivity and units of attention in narrative comics. What
was left out from this discussion was the question of characterisation,
that is, the creation of person-like qualities, and the understanding of
the personality traits of fictional characters. Fotis Jannidis defines char
acterisation as the process of “ascribing information to an agent in the
text so as to provide a character in the storyworld with a certain prop
erty or properties, a process often referred to as ascribing a property to
a character” (2015). In other words, the process of characterisation pre
supposes that characters are agents that have properties. In this capacity,
characters are person-like, i.e. they resemble a person in the real world,
or in previous fictions, at least to some extent.
Not all narratologists have been convinced, however, that the ques
tion of characterisation belongs to narratology at all. Some have limi
ted the interest in the category of the character to the sense of actants,
that is, the study of characters as structural roles in the narrative com
position, such as heroes and villains that fulfil a certain plot function.
Gérard Genette, in turn, has conceived the issue of character as a matter
of story content: what the story is about instead of how it is told. More
precisely, Genette defines characterisation as “the technique of constitut
ing characters with narrative texts” (1988, 136), and that is constituted
of various narrative devices that are not specific to it, such as denomi
nation, description, focalisation, and speech and thought representation
(ibid.). Therefore, for Genette, characterisation is a particular semantic
effect that should not be privileged over other effects. For instance, if the
reader’s knowledge of a particular character’s personal traits, intention,
experience, perspective, or worldview is revealed in part through the
techniques of focalisation, description and speech, and thought repre
sentation, we should privilege the study of those narrative techniques
instead of the study of the character effect.6
Yet, one rationale for arguing that the question of characterisation
indeed belongs to narratology, and that the category of the character
should not be reduced to narrative techniques or effects of the story con
tent, is that characterisation cannot always be that easily decomposed
into narrative devices without losing sight of a crucially important ele
ment of narrative fiction. The study of characterisation allows us to fo
cus on a vital aspect of narrativity, which is the capacity of fictional
characters to inspire a narrative response.7
In what follows, I will ask how comics may prompt their readers to
ascribe properties to characters. In this investigation, I will consider
the main parameters of characterisation, especially with regard to the
characters’ mimetic (realist) function, that is, how they can be perceived
as possible anthropomorphic persons. Subsequently, I will discuss the
way in which a character in comics may achieve a sense of psychologi
cal complexity (to be defined later). One of the most used and useful
approaches in this respect has been the rhetorical narrative model that
addresses three aspects of characterisation. This model, developed by
James Phelan on the basis of earlier approaches and taxonomies in lit
erary studies, distinguishes between the character’s mimetic (character
as person, as images of possible people), thematic (character as idea),
and synthetic (character as artificial construct) dimension. The premise
in Phelan’s theory is that all these elements are present at some level
in all characters in fiction, but they may be more or less developed or
underscored.8 For instance, even the most realistic character in comics is
an artificial construct at some level, made of drawn images and words,
but that artificiality can be more or less covert. The meaning of the
distinction is that narrative texts make different applications of these di
mensions and can change their relation as the story progresses, increas
ing and decreasing the reader’s interest in the reality of the character, a
particular theme or some aspect of the composition.
Before moving forward, however, it needs to be emphasised that to
focus on characters in narrative comics is always a matter of taking a
particular perspective on certain kinds of stories. The idea of characters
as agents that have properties is dependent, on the one hand, on the
kind of narrative that we are reading and, on the other hand, on how
we focus our attention as readers, such as how much we pay attention to
characters instead of plot, theme, idea, historical context, or something
else. In other words, the focus on characterisation necessarily isolates
the character as a separate category.
Most narrative theory agrees that character and action (or event)
are interdependent elements in fiction and that for that reason their
distinction always remains to some extent arbitrary.9 The characters’
actions create events and move the plot forward, and the meaning of
an event is typically filtered through what we know about the partici
pants and how they perceive the event. The events that are related in the
story, in turn, can reveal and illustrate the agents’ properties. Likewise,
it may be difficult to think of characters as distinct entities in relation to
first-person narrative fiction where the narrator-character tells his or her
story. When the whole narrative is indicative of the narrator-character’s
properties, focussed on the narrator’s speech and thought, it may indeed
be more meaningful to investigate the narrative techniques of voice,
style, perspective, or the relationship between the time of narration and
the time of the events, than to study the narrator as a character. The
relevance of the concept of character in first-person narration will ulti
mately depend on the individual features of the given story and the kinds
of questions we seek to ask from it.
In the following sections, I will focus on characterisation in comics,
specifically in the mimetic sense: how do comics prompt the readers to
ascribe mimetic properties, comparable to some extent with real-life per
sons? Moreover, how can comics prompt us to ascribe such properties
to entities with qualities that are strikingly dissimilar from humanlike
persons? The characters’ thematic function—how do comics associate
characters with certain thematic properties?—will only be discussed
briefly. It is, nevertheless, important to emphasise that while characters
in comics can represent a theme, an allegory, or an ideology perhaps
similarly to the way in which real people may symbolise an idea, an
ideology, or a world view, types and caricature play a noticeably more
significant role in this medium than in many other forms of representing
person-like characters or real people.
Identifying Characters in Comics

The establishing of a person-like character, or a character as a possible
person, involves a set of basic expectations about what characters are,
regardless of the narrative medium. This is, to a large extent, dependent
on the reader’s knowledge about real people, their minds and behaviour.
Fotis Jannidis has referred to these expectations by using the concept of
basis type. This means the presupposition that a character has an inside
and outside, or more precisely that a character has, on the one hand,
an invisible “inside”, which is the source of all cognition, intentions,
wishes, emotions, and, on the other hand, a visible “outside”, which
can be perceived (Jannidis 2015). Therefore, what distinguishes char
acters from things and other entities is that the reader can attribute to
characters anthropomorphic and anthropoid qualities, such as thoughts
and emotions. Subsequently, the recognition of mental states in fictional
entities invites the making of the fundamental distinction between
human-like persons, or anthropomorphic beings, on the one hand, and
things, or non-personal entities on the other. Jannidis specifies, however,
that all aspects of the basis type can be negated for a specific character
in a particular narrative, but either this is done explicitly or it results
from generic conventions that enable the treatment of characters in this
way (ibid.).
But what makes some entity a character in comics, and how can char
acters be given traits of personality? Moreover, is there something that
sets characters and characterisation in comics apart—especially in terms
of their mimetic dimension—from characters in other forms of narrative
fiction such as prose fiction, picture books, or films?
We may be able shed some light on these questions by comparing the
basic forms of characterisation in two adaptations of Dante Alighieri’s
Inferno, the first part of his famous epic poem The Divine Comedy:
Guido Martina and Angelo Bioletto’s Mickey’s Inferno (L’Inferno di
Topolino, 1949), an important milestone among the Italian adaptations
of literary classics in the popular series, Topolino, and the American
artist Gary Panter’s Jimbo’s Inferno, a prequel to Jimbo in Purgatory
(2004).10 The graphic styles, narrative techniques, readership, publica
tion format, and distribution of these two adaptations are radically dis
similar. However, at the same time, there is sufficient similarity in their
characterisation, in particular their use of typified and caricature-like
characters, for the comparison to be illuminating about certain basic
strategies in comics in this regard.
Gary Panter’s Jimbo’s Inferno features as its protagonist the graphic
artist’s pug-nosed character called Jimbo, who is often referred to, and
perhaps with good reason, as a punk-rock character.11 Jimbo’s sidekick
and guide, and a stand-in for Virgil, the Roman poet, is a character called
Valise. Valise is a rectangle box that speaks and hovers above the ground,
reminiscent of a large suitcase, a portable stereo system, or possibly a
miniature storehouse that also functions as Jimbo’s means of transpor
tation. Valise is also referred to as a “self appointed parole robot”. The
Hell in this version is a Los Angeles shopping mall named Focky Bocky,
and Beatrice, Dante’s dead loved one, is represented by a group of “cute
girls” called Soulpinx Girls. By contrast, Martina and Bioletto’s Mickey’s
Inferno features Mickey Mouse as a Dantesque character—although an
actual Dante is also portrayed at the story’s end—and Goofy as Virgil.
The first step in the process of understanding characters, regardless of
the narrative medium, is the recognition of some entity in the story as a
character. For many readers of these adaptations, the identification of the
protagonists probably takes place already before the reading. This is due
to various extratextual reasons. First, the title and the cover image tell us
who the protagonists are. Second, the reference to Dante’s Inferno may
provide the reader, at least those readers who know something about
Dante’s work, with other expectations concerning the characters, their
situations, and world. Gary Panter, moreover, specifies, in a description
of “Focky Bocky” that is placed underneath the Table of Contents, what
kind of adaptation the reader is about to read:
Don’t try to pass a pop quiz on Dante’s hell based on a reading of

this comic: it won’t work. Even though this comic is engorged with
Dante’s hell and though Jimbo mouths a super-condensed version
of what happens in the infierno [sic], canto by canto, characters are
fused, action inverted, parodied, subject to mutation by my odd
memories and obsessions and my odd whims, sentences are clipped.
Third, the reader may be familiar with the characters, Mickey, Goofy,
and Jimbo, from elsewhere, such as from other comics, and can thus
expect them to be the focus of the adventure.
The protagonist Jimbo’s name is made perfectly clear by the paratexts
of the book, including the title and the two subtitles. The subtitle on
the cover reads “wherein, Jimbo, led by Valise—city appointed parole
robot, enters the vast gloom rock mallscape, Focky Pocky, in pursuit of
the Soulpinx”, and on the name page: “A Ridiculous Mis-Recounting
Of Dante Alighieri’s Immortal Inferno In Which Jimbo, Led By Valise,
In Pursuit of The Soulpinx, Enters Focky Pocky, Vast Gloomrock
M allscape”. Upon reading, then, a global look at the first pages of both
stories can verify who the protagonists are: the named figures who are
continuously portrayed in the images (Figure 7.1). In Jimbo’s Inferno,
the protagonist’s name is confirmed in the first lines uttered by Valise
in response to the question of the whereabouts of Focky Pocky: “Just
look over your left shoulder Jimbo” (7). Similarly, Valise’s name is given
in the dialogue as Jimbo reveals that “Valise, I have some fear of enter
ing Focky Pocky” (8). Furthermore, a few pages later at the beginning
of Canto VIII, Valise identifies himself and Jimbo, upon talking to the
amphibious vehicle that is to take them across the marsh of Styx, as
Jimbo and “his parole Valise” (12).
Figure 7.1 Gary Panter. Jimbo’s Inferno © 2006 by Gary Panter.

In the frame narrative of Mickey’s Inferno, where Mickey and Goofy
are acting in a theatre production of Dante’s Divine Comedy, and are
then hypnotically transformed into Dante and Virgil and transported
into the Underworld, the protagonists’ familiar names are given in the
first panel. As Mickey and Goofy enter the Hell of the Divine Com-
edy, they don Dante’s and Virgil’s clothes and Mickey is also ascribed
other aspects of Dante’s identity. Mickey wonders to himself: “I saw
this creepy forest, as I knew it/From my first canto—I was Dante, right?
Or was I? Hmmm! I still felt like a poet… But I’m not Italian, am I?”
(Figure 7.2). Mickey’s uncertainty about his identity does not last long,
however, since a skull falls on his head and he comes back to his senses.
Goofy, by contrast, has not lost his old self in the transition and, fur
thermore, is aware that they have been transported to another world.
The situation thus provides the two characters a kind of double identity
with their familiar personal traits—parts of their ordinary clothing can
also be seen underneath the new cloaks—and with some new features
borrowed from Dante’s characters, Dante and Virgil.
Figure 7.2 G uido Martina & Angelo Bioletto. Mickey’s Inferno (1949/2006)
© Disney.
The technique of transposition between the fictional worlds thus cre
ates a specific effect of parody, where Disney characters are transported
to a well-known literary milieu with some limited changes to their
identity—the technique is then re-employed in many later Italian Disney
adaptations of literary classics. The presence of various other charac
ters from the Donald Duck universe, such as Pegleg Pete, Zeke Wolf,
Pinocchio, and Donald Duck himself further enhances the effect of two
superimposed worlds and casts of characters. As Dante, Mickey Mouse
maintains his basic characteristics: he continues to be as imaginative,
resourceful, persistent, brave, and valiant as always. Yet, his figure also
has some added Dantean attributes: Mickey refers to himself and Goofy
as poets, and the two protagonists are dressed like Dante and Virgil in
the style of Gustave Doré’s illustrations.
Characters’ names and personal pronouns, regardless of n arrative
media, are important points of reference in fiction that unite all r eferences
and descriptions pertaining to that character. Nevertheless, in written
fiction, a name, a personal noun, a name-like description, or other lin
guistic marker of a character12 may play a more seminal role in terms
of keeping track of a certain character than in visual narratives. In liter
ature, a proper name provides not only the process of characterisation
with a rudimentary but also necessary means of consistency, that is, a
point of reference to which properties can be ascribed. Both the implicit
means of characterisation, such as the character’s actions that suggest
traits of personality, and explicit descriptions by the narrator or other
characters, can be united in the character’s name. In comics, the recog
nition of the same figure can already fulfil this function to a large ex
tent. The character’s continuing presence in the panels serves as a point
of reference that helps the reader to construe an entity as a person-like
character.
Not much is required of a thing to become a character in comics. Just
an abstract shape is sufficient, provided that it behaves to some extent
like a person, can speak or think, or engages in goal-oriented action.
Think, also, how square shapes, cubes, and boxes have been characters
in comics from Ernest Riebe’s Mr. Block from the 1910s to Squarehead
who was featured in the late 1950s American comic book Cosmo the
Merry Martian. In Jimbo’s Inferno, Valise is a rectangle robot-looking
box, unchanging in outlook, with no other signs of inner life other than
his speech—Valise’s movement can also be perceived to be mechanical.
Valise’s comments are often mere instructions to Jimbo, and he may
refer to his own thoughts and imagination as “computations”. At the
same time, Valise can be quite spontaneous, opinionated, ironic, and
metaphorical in his responses, and obviously he also speaks and under
stands speech—all indications of human-like mental states. Valise is also
much more reflective and analytical than Jimbo concerning the potential
outcomes of his actions. Jimbo, on the other hand, is a man of action
and simple pleasures who appears to be a kind of blank canvas with little
interiority and is mentally clearly more limited than his robot-like coun
terpart. Despite his person-like appearance, Jimbo is without a sense
of a past or much sense of the future, and seems to have a very limited
notion about the meaning of his experiences. His simple-mindedness,
straightforwardness, and innocence contrast with Valise’s knowledge,
experience, and occasional complex thoughts. Both of these characters,
thus, in their varying ways, challenge the division into a character’s ex
terior and interior qualities.
But let us stop here for a moment: how was I able to claim that these
particular traits can be ascribed to Jimbo’s and Valise’s characters? Be
yond the recognition of a basis type, or the identification of a character’s
bare existence, there are a number of other conditions that affect the
recognition of some entity as a character, regardless of the narrative me
dium. The basic principles of characterisation that narratology has iden
tified in literary narratives are effective here, with the exception that we
also need to pay attention to visual cues of characterisation and the in
teraction between words and images. Mieke Bal, for instance, lists four
principles of characterisation, which work together to help the reader to
construct an image of a character: repetition, accumulation, relations to
other characters (or a character’s relation to itself in an earlier phase),
and transformations (2009, 127). Rimmon-Kenan, in turn, argues that
the four main principles of cohesion that help the reader to create a sense
of a character’s traits in literary fiction are repetition, similarity (or ana
logy), contrast, and implication (2003, 39). It is worth noting that these
two models share three principles: repetition, character’s relation with
other characters, and the question of consistency.13
The repetition of any aspect of the character’s external feature, be
haviour, speech, or thought has the potential to be an indication of
personality. This can involve a word, phrase, way of speaking, facial ex
pression, gesture, behavioural pattern, clothing, association with a cer
tain environment, or the like. Also forms of graphic style, such as colour
and the graphic line, can describe a particular character and his or her
situation. The repetition of Jimbo’s spontaneous reactions to things and
people—“Ugh!”, “Back off mister dim ass!”, or “Hey!”—and his con
stant questions about where they are going and what is happening are
indications of a certain innocence and empty canvas-like quality. These
short comments are contrasted with Valise’s know-how, advice, or ironic
and metaphoric comments, such as “If you don’t bury your eggs in the
sand, you are only a consummate ape of nature” (XXIX canto). Unlike
Valise, however, Jimbo is attributed with explicit signs of emotion that
can be seen on and around his face, gestures, and posture, such as the
character’s sense of surprise, shock, fear, and anger that are depicted
by lines and beads of sweat around his face. Jimbo’s emotional involve
ment in action is a steady feature in the story, and this may be taken as
an indication of a personal trait, such as spontaneity or, again, certain
simplicity. He never seems to learn, for instance, that unexpected things
are bound to happen in Focky Bocky.
The principles of similarity and transformation can be perceived as
aspects of the same rule, pertaining to the evaluation of a character’s
consistency. In this regard, Uri Margolin has helpfully discussed the
minimal constitutive conditions of characters under which they can be
introduced and sustained, including the principle of consistency. Such
conditions include the possibility of assigning at least one property to an
individual whenever that character occurs in the text (Margolin 1995,
376–377). Furthermore, in most narratives, this also requires that the
character is distinguishable from other individuals, has a coherent set
of features (Margolin 2005, 53), and some persisting sense of identity in
temporal continuity despite all the changes that the figure may undergo
in the course of the narrative. All these conditions can be easily under
mined in comics, but such problematisation is likely to be motivated and
thematically foregrounded.
Jimbo and Valise are types, unchanging characters who maintain their
visual and personal consistency throughout their journey. An unusual in
dication that Jimbo might have a more conscious intention behind his
actions emerges when he states in canto XXVIII that “I come here gyre
by gyre to gain experience of the way” (33). The story, however, does not
provide us with any proof that Jimbo’s mental state would go through
changes over the course of the adventure. In Mickey’s Inferno, the situ
ation is more complex in that while Mickey and Goofy clearly maintain
their basic, unchanging personal traits in the Dantesque Underworld, they
also go through a partial transformation under hypnotist Abdul’s influ
ence. Mickey’s first-person narration, in rhymed prose, also creates a con
tinuous consciousness frame for the narrative that allows us to have some
sense of his mental state. At the same time, Mickey’s rhyming narration is
another indication that he has, indeed, become a poet. However, Goofy
also identifies himself with poets as he refers to Dante as a “fellow poet”.
The character Valise’s box-like shape manifests to us that a charac
ter in comics can be invented on the basis of a few recognizable lines
and their bodies can be freely shaped for expressive uses.14 Moreover,
the convention for placing utterances, typically in speech and thought
balloons, next to the source of speech has the potential to match some
entity directly with a mental state. With regard to anthropomorphic ani
mal figures, such as Mickey and Goofy, the expectation that characters
have an “inside” is usually held, and typically made evident by means of
dialogue, although that inside may not contain much depth. What is un
usual with regard to Mickey’s Inferno is that the continuous conscious
ness frame of this narration reveals aspects of Mickey’s mental state,
for instance concerning his conviction, but also sometimes uncertainty,
that he may be dreaming the underworld adventure and not actually
experiencing it.
Complexity in Comics Characters
Characters in comics, in general, are sometimes taken to be flatter than
in literary fiction. Such a generalisation is unfair, at least from today’s
perspective. Yet, the claim also holds true for many kinds of comics.
From a historical perspective, it can be claimed that, especially before
the development of the graphic novel and autobiographical comics in the
1960s, the psychology of characters in comics had not attracted much
attention, and the medium as a whole had strongly relied, as it does still,
on caricature and type-like characters. Types in comics are not without
psychological appeal, since caricature can effectively illustrate forms of
behaviour, individual traits, or social roles, but, as Hergé’s Tintin also
points out to us, despite the realism of the milieu, the development of the
protagonist’s inner life is often not that important.15 One reason for this
is simply that so many comics have been targeted at children.
By contrast, it is a generally held notion that today’s character-centred
graphic novels and narratives, which have an adult audience, can create
complex personalities who have psychological depth. We must then ask:
How do we come by information in comics that suggests that a given
character is an individual with complex mental states?
Before trying to answer this question, it is useful to think of the way
in which character’s complexity has been defined in narrative theory. For
the British writer E. M. Forster, flat characters are constructed around
a single idea or quality in their purest form; they can be summed up
in a single phrase. By contrast, Forster defined, in his Aspects of the
Novel (1927), the round character as someone who is capable of surpris
ing and changing, and thus having “the incalculability of life about it”
(1953, 75).16 However, for Forster, flatness does not correlate in a simple
way with the character’s lifelikeness despite his emphasis that round char
acters are a major achievement of literary modernism. On the contrary,
flatness can suggest a kind of lifelikeness of its own. This is exemplified,
for instance, in Charles Dickens’s protagonists who can be summed up in
a sentence, but still have a “wonderful feeling of human depth” in them
(Forster 1953, 68). Moreover, flat characters are useful in providing the
story with their own atmosphere, and their collision with each other or
with round characters may create specific effects of lifelikeness (Forster
1953, 66). Forster further claims that flatness and roundness can occur
in the same characters in different parts of the narrative.17
Both Mickey and Goofy are clearly flat characters, easily recognised
and remembered. It does not make sense to expect them to be lifelike
persons. Goofy’s dominant trait is that he is clumsy or, perhaps more
precisely, that he is foolish in his clumsy behaviour. The character’s goof
iness is underscored both by his name and appearance: his long legs,
ears, and muzzle, big shoes and tall green, or sometimes blue, hat. The
same quality can be observed in his behaviour as Goofy regularly gets
into trouble due to his rashness and lack of reflection as much as his
physical shape. Perhaps this is also something to do with the perceived
real-world characteristics of large hounds that, with their sometimes
amusingly doleful expressions and clumsy antics, might be perceived as
somewhat comic and affectionately soppy.18 His dominant trait is fur
ther associated with another qualification that is equally unchanging: his
goofiness is always lovable; his clumsiness and lack of reflection is funny
in a sympathetic way. If Goofy stories include exceptions to these basic
qualities of the type, they are thematically emphasised and motivated.
For instance, while it is not unusual for Goofy to make the right decision
and get himself out of trouble, his cleverness is likely to be presented as
luck or a happy coincidence rather than the result of conscious reflection.
These recognisable properties are simultaneously inner and outer qual
ities. The characters’ relationships are similarly invariable. Mickey and
Goofy’s personal traits are sharpened by their contrast: the witty and reflec
tive versus the unreflective character, one imaginative and the other simple,
one more heroic and one foolish, one short and the other tall, and so on.
The contrast between Mickey’s intellect and poet-like qualities and Goofy’s
simplicity is reflected in the English translation in the difference between
Mickey’s standard English and Goofy’s colloquial accent. Goofy responds
to Captain Charon, the ferryman of Hades, when the latter does not rec
ognise Mickey Mouse: “How ignorant can yuh be, cap’n? This is Mickey
Mouse! Don’t yuh read comic books?” (Figure 7.3). Exceptions to expecta
tions are again possible—Goofy, for instance, unexpectedly takes a good
deal of initiative in getting through the Underworld in Mickey’s Inferno;
he is determined to find the exit—even though this can be explained by the
exceptional circumstances. The source text of Inferno provides the story
with an unusual setting, and the characters with new pseudo-identities,
which create some effects of defamiliarisation, that is, limited alterations
to their dominant characteristics (compare with Elseworlds stories of DC
superheroes), without the need to explain these changes.
Figure 7.3 Guido Martina and Angelo Bioletto. Mickey’s Inferno (1949/2006)
© Disney.
Forster’s categories of flat and round character do not do justice to
the medium of comics, where caricature can take so many different
forms and have such a variety of effects. To better understand a char
acter’s type-likeness in comics, we can turn to David Fishelov’s deve
lopment of Forster’s model, where Fishelov distinguishes the characters’
flatness and roundness on the two levels of textuality and the reader’s
construction. On the one hand, flatness on the textual level refers to
the amount of space that is apportioned to a particular character, and
the kind of literary or linguistic attention that a character is given in the
text.19 The character’s textual flatness typically amounts to limited and
one-dimensional attention so that a character is portrayed only from
one perspective, always saying the same things, repeating some pattern
of behaviour, or is associated with only one trait. Examples abound in
comics: Goofy’s goofiness, the extreme luck of Gladstone Gander, the
mayhem-seeking Dennis the Menace, the dishonest and lazy Roger the
Dodger, and the unbearably malodorous Stinky in the Moomins. In con
trast, in ‘round attention’, a character is extensively represented and re
ferred to in the text, and characterisation may take multidimensional
forms. This can involve, for instance, the presentation of consciousness
and inner life, varied points of view on the character’s action, behaviour,
and perception—through narratorial strategies, focalisation techniques,
speech and thought representation, for instance—and dramatisation in
action in different situations and circumstances.
In Posy Simmonds’s graphic novel Tamara Drewe (2008), which in
cludes sustained verbal narration by various characters, two first-person
narrators, Glen Larson and Beth Hardiman, are given extended and
varied attention throughout the work. In their verbal narration, these
characters relate the events at a writer’s retreat in the English country
side through diary-style exposition. Their narration complements and
contrasts with each other and with the perspectives and voices of other
characters; the narrators may also quote each other or other characters.
Beyond the verbal narrative track, Glen and Beth are shown engaged in
action and dialogue scenes and their mental states are also related to us
by means of narrative drawing. All this amounts to ‘round attention’ on
the textual level.
On the other hand, the distinction between flatness and rotundity on
the conceptual level reflects the way in which characters may be per
ceived in the reader’s imagination as person-like entities in a fictional
world. Thereby, a conceptually flat character represents some single
(and sometimes simple) category, such as some moral, social, or aes
thetic category. A constructionally round character is a character who
cannot be portrayed in this way. Such an effect may be achieved by using
multilayered modes of representation, including the presentation of in
ner thoughts, or complex sensory and mental processes. The crucial cri
terion here is that conceptual roundness requires the reader’s perception
of the character’s complexity as an individual rather than as an element
or the narrative structure or a theme.
Jimbo, Valise, Mickey, and Goofy are conceptually flat characters in
their varying ways. The contrast between the two characters of each
pair further associates them with the literary device of characterisation
through contrasting types, such as Don Quixote and Sancho Panza, or
indeed Dante and Virgil, where the opposites highlight each other’s per
sonality as a kind of foil. The reader’s background knowledge of Jimbo’s
other stories, or about Mickey and Goofy’s personal traits in the Disney
universe, may also affect our evaluation of their conceptual flatness.
Mickey’s first-person narration affects the conception of this character
by allowing the reader to perceive him as a more conceptually rounded
character than the usual. His narration, however, does not reveal much
about the narrator himself; Mickey rarely mentions his own thoughts
and emotions. Instead, his narration focusses on what is happening
around him and Goofy, what has just happened and, frequently, he sim
ply quotes dialogue.
Schematic and type-like qualities can have an allegorical function.
This potential is much more prominent in Jimbo’s Inferno than in
Mickey’s Inferno. In Dante’s Divine Comedy, allegory affects the whole
composition, from the story events to characterisation. The narrator,
Dante, is not only a poet, but he also plays the role of Everyman, or
a pilgrim. His journey to Hell, Purgatory, and Heaven, during which
he encounters many people whom he has known in real life, is at once
a personal experience and an allegorical journey that leads him to the
spiritual state of love and faith, represented by his lost love, Beatrice.
At the same time, the journey reflects Everyman’s fate that pertains to
all people who, “Midway upon the journey of our life”, start to ponder
the possibility of their death and what may happen thereafter. Over the
course of his travel, Dante learns the spaces that the soul may inhabit
after death, but these places are also states of the soul, not real spaces.
His first guide, Virgil, symbolises human reason and poetry, whereas
Beatrice, his second guide, stands for love, faith, and heavenly grace.
In Guido Martina and Angelo Bioletto’s Mickey’s Inferno, the journey
in the Underworld is represented as a kind of hypnotic dream. The alle
gorical level is thin or non-existent. At the end of this adaptation, how
ever, the comic is compared to the allegorical qualities of Dante’s epic.
Here, it is revealed that the writer Martina and the cartoonist Bioletto’s
alter egos, who are tortured by Dante, who thinks that the artists have
betrayed him with their adaptation, have used Mickey and Goofy’s char
acters to gain new readers for the epic poem. It is thus suggested, by
analogy, that the comic gives access to Dante’s classic, perhaps similar
to the way in which an allegory can illustrate complex ideas to make
them more comprehensible to readers. The staging of Dante’s approval
of the adaptation does not make the characters of Mickey’s Inferno any
more allegorical or less parodic, however. By contrast, in Gary Panter’s
Jimbo’s Inferno, the allegorical dimension of the story is quite explicit.
Here, Hell is conceptualised as a vast shopping mall called Focky Pocky.
Jimbo’s joy about being once again able to look upon the stars at the end
of the journey can be conceived as an allegory of Everyman liberating
himself from the Hell of suburban shopping and infotainment. Jimbo
may be a punk rock character by appearance and association, but he is
also an allegorical figure in his own right, journeying through a consum
erist Hell on earth.
One clear advantage of Fishelov’s dual perspective on complexity in
characters is that it allows different strategies of characterisation to re
sult in varying effects of flatness and rotundity. Flatness on both the
textual and conceptual levels creates a pure type while roundness on
both levels can produce a ‘pure’ individual, but other combinations are
equally possible. Textually one-dimensional description may also sug
gest conceptual roundness (a type-like individual). For instance, the
trait of eccentricity in a minor character, such as Uncle Toby and Walter
Shandy’s obsessions in Tristram Shandy, may not be tied to a simple
effect, but require complex evaluation from the reader. In these cases,
an individual’s single trait, such as an eccentric obsession, presupposes
some psychological depth, “an element of which a mere type is deprived”
(Fishelov 1990, 430). By contrast, in Glyn Dillon’s graphic novel The Nao
of Brown (2012), changes in colouring and visual ‘flatness’ add another
dimension to the general round attention to the protagonist. Here, the
narrator-character called Nao suffers from severe obsessive-compulsive
disorder, involving especially preoccupation with violent thoughts, and
often the symptoms of her mental state are also indicated by means of
stylistic change. Thus, during some of her violent visions of hurting other
people, the colouring of the panels may change into all red or grey, and
at one point Nao Brown is also portrayed in black and white against the
colour background (Figure 7.4). The uncontrollable symptoms of Nao
Brown’s condition are further dramatised by the embedding of her vio
lent visions, at various instances, as parts of the unfolding events of the
narrative. For this reason, the reality of her visions can sometimes be
evaluated only by reading the story forward in order to detect, retrospec
tively, whether the violence is something that has actually occurred in the
reality of her world or only consists of her inner visions.
Type-like individuals cannot be easily identified in our two Dante ad
aptations, but the possibility that textual flatness can have a great variety
of conceptual effects is significant for characterisation in both. We might
want to claim that Valise, who is given an exceptionally flat visual treat
ment in Jimbo’s Inferno, nevertheless has some roundness in the concep
tual sense as he keeps surprising the reader by the spiritual profundity of
his statements. These words of wisdom include, for instance, his advice
to Jimbo in canto XIX, “Don’t agonize over God’s divine equity”, and
Figure 7.4 Glyn Dillon. The Nao of Brown © 2012 Glyn Dillon.
the insight that I quoted above concerning the burying of eggs in the
sand. Whether these statements presuppose psychological depth and a
sense of individuality is another matter, however. A significant challenge
in ascribing mental states or psychological depth to Valise is his bluntly
mechanical form. With the exception of his capacity to think and speak,
and the circles on his box that could be his eyes, Valise is a radically
non-anthropomorphic figure. By contrast, the appearance of Mickey
and Goofy is a combination of an animal, mouse and dog-like shapes,
and much that is human; their shapes serve as a kind of envelope for a
flat, anthropomorphic self, but one that can display a rich repertoire of
caricatured facial expressions, gestures, and postures. Mickey’s charac
ter shifts towards more individuality through first-person narration, but
not to the extent of becoming a type-like individual—he remains a type
with some new personal features.
Individual-like types that have a round and multi-perspective repre
sentation, combined with typification on a conceptual level, are usual
in comics. An individual may resemble a literary type, such as Sancho
Panza; a roundly described character may also be typified thematically
or socially (for instance, as a representative of ethnicity, gender, age,
or any other group). In Tamara Drewe, Nicholas Hardiman and the
protagonist, Tamara Drewe, are portrayed in a relatively round fashion.
They are not just shown in action or dialogue in the images, for instance,
but the readers also occasionally have access to their thoughts, recollec
tions, and perceptions. They also have type-like qualities in their roles as
a womanising artist, or the adulterous husband, and the femme fatale.
One structurally important contrast in characterisation in this graphic
novel is based on differences in terms of textual attention: the contrast
between characters who are allowed to narrate at length in first-person
narration (Glen Larson, Beth Hardiman, and Jody Long), and those
who do not narrate, but whose thoughts and emotions are presented
to us in direct discourse by other means, such as through quotations of
their writings or speech and thought balloons (Tamara Drewe, N icholas
Hardiman, and Casie Shaw). The lack of continuous narrative voice
does not necessarily make the latter more conceptually flat, however.
On the contrary, the alternation between implicit and explicit means of
characterisation may increase the interest in the characters’ psychology
and thus suggest new levels of complexity.
We run into problems, however, if we insist on a strict distinction
between intermediate categories such as a ‘type-like individual’ or an
‘individual-like type’. The crucial point here is that characters in comics,
both within one story and with regard to one character, may create dy
namic relations and combinations between type-like and individual-like
qualities. Fishelov’s model suggests that there is a long continuum bet
ween flatness and rotundity, or ‘pure’ type and ‘pure’ individual. The
focus on such a continuum, and the variety of relations between type
and individual, is especially important from the perspective of narrative
comics, where types and caricature have traditionally enjoyed a central
position. 20 The multimodal nature of narratives in comics also often
requires that we pay heed to the varying indications of flatness and ro
tundity, or type and individual, in words and images. Fishelov’s estimate
that “the modernist novel strongly tends to dispense with the ‘pure’ type
character as well as with the individual-like type” (1990, 432) certainly
does not hold with regard to the history of comics, even if emphasis on
the complex individual may be a key element in the development of the
graphic novel in recent decades.
One advantage in Fishelov’s model is that it pays due attention to
the issue of the reader’s active construction of characters as person-like
agents. In this perspective, characters are not only products of formal
devices and features of the text. However, one evident problem with
Fishelov’s approach is that it does not sufficiently reflect the possibility
that characters serve various functions simultaneously or over the course
of the narrative, as explicated in James Phelan’s division into mimetic,
thematic, and synthetic aspects and functions. The readers of comics
may need to focus on the dynamic between these functions during the
narrative progression of a story. Thereby, it must be noted that charac
ters who can be conceived as mimetically round can, at the same time,
be thematically relatively flat, and a character’s textual flatness may
function as a kind of guise for exploring the limits of the given type.
Mimetic flatness may also give way to thematic roundness, or the fore
grounding of the character’s synthetic aspect may bring the thematic
component into greater prominence. 21 The latter is especially true in
relation to the ambitious rewritings of the superhero genre in the late
1980s and the 1990s, such as Alan Moore, Dave Gibbons, and John
Higgins’s Watchmen (1987). Frank Miller’s Batman in The Dark Night
Returns (1986) maintains his superheroic label and costume (although
small changes are made to the latter), but the superhero’s age and patho
logy, as well as the dilemma of whether he could break his oath of never
killing anyone, gives the type considerable new complexity and mimetic
roundness. In The Dark Night Returns, Batman’s allegorical role as a
crusader against crime also becomes more ambiguous, as his affinity
with the villains is underscored, for instance due to his lawless methods,
rivalry with Superman, and split personality.
On the Significance of Genre and Medium

Generic expectations—understood as conditions of meaning and con
ventions of reading that relate to a specific genre—can significantly pre
determine the kind of relation that the fictional world holds with the
real world, including the characters’ affinity to or distance from peo
ple in the actual world. Such expectations may also affect the kind of
information that is given about characters. This involves, for instance,
the relevance of background knowledge, the portrayal and prominence
of the characters’ inner life and sense of development (experimental,
psychological self or signs of aging), or visual detail and verisimilitude
with regard to their appearance and environment. The impact of generic
expectations can be presumed to be strongest where the generic features
are also the strongest, that is, in genre fictions such as humour strips,
or superhero, war, romance, erotic, and horror comics that privilege
particular kinds of stories, characters, and narrative modes. By contrast,
a central expectation of contemporary first-person graphic novels, such
as Tamara Drew or The Nao of Brown, is that they portray complex
characters with considerable psychological depth.
Characters’ genre-related roles, as representatives of a social type,
moral categories, or themes for instance, or their relation to their per
ceived realism, can take precedence over real-life social knowledge,
concerning, for instance, social identity such as gender, age, and ethnic
ity, and stereotypical frameworks, as also some empirical research has
shown.22 The attention to generic expectations can also help to explain
how readers set up a certain kind of character mode and fictional world,
and channel their inferences according to what can be regarded as prob
able, possible or relevant knowledge in this world. For instance, as the
readers know that characters in superhero comics have superhuman
powers and capabilities, it may be expected of them to be able to throw
heavy trucks off the road, shape shift and shift shapes, have incredible
speed, telepathic powers, or see, smell, taste, feel, and hear more than
any normal human. The readers of this genre are also likely to be aware
of the fact that superheroes and supervillains are constantly recreated,
revised, and adapted for new audiences. Not all comics are compatible
with such expectations.
Other qualities in characterisation are medium-specific. For instance,
some of the differences between literary narrative fiction and the visual
means of characterisation in comics may become more evident if we
think of the challenges of adaptation between these two media. Why
is it that the adaptation of characters from comics into literature has
been much less common than the reverse, save alone adaptations of com
ics into film? Comics adaptations of literature have included popular
series, such Classics Illustrated or the parodies in Topolino that attempt
to encourage young readers to read the source texts. In recent decades,
numerous ambitious interpretations of literary classics have come out, in
cluding Martin Rowson’s version of Laurence Sterne’s Tristram Shandy
(2010), Eric Drooker’s adaptation of Allen Ginsberg’s Howl: A Graphic
Novel (2010), Régis Loisel’s Peter Pan (1990–2004), Posy Simmonds’
Gemma Bovery (1999), Winshluss’ Pinocchio, or Gareth Hinds’ vari
ous versions of Shakepeare’s plays, just to name a few. Dante’s Inferno
has been adapted into graphic novels such as Seymour Chwast’s Dante’s
Divine Comedy (2010) and Hunt Emerson and Kevin Jackson’s Dante’s
Inferno (2012). 23
The scarcity of literary adaptations of comics can be explained to some
extent by different kinds of audiences, or niche markets. The power of
markets is reflected, for instance, in the fact that comics are adapted
into literary fiction usually only when some series becomes extremely
popular. In continental Europe, some of the most popular comics series
for young readers, such as the French-Italian W.I.T.C.H. and the Belgian
adventure series Thorgal have been adapted into novels. 24 In the United
States, the main superheroes in the Marvel and DC Comics universe,
starting with the 1940s Superman fictions, also exist in paperback se
ries. Outside the market of adaptations, literary versions of superheroes
and manga characters are quite popular in fan fiction.
Beyond market considerations, the one-sidedness of adaptations
between literary narratives and comics may also be explained by the
possibility in comics of overcoming certain dissimilarities between the
narrative media. Comics regularly manage to tell the ‘same’ story as lit
erature or, at least, make a successful parody, but the need to give up the
characters’ recognisable visual appearance in the novelisation of comics
may seem a much bigger loss. The art of caricature, the recognisable
and memorable quality of characters in comics, based on some simple
visual traits, is difficult to transfer into literature. Protagonists in comics
tend to have physically remarkable features, such as a sketchy face with
a particular shape or expression, or head, body shape, hair, or clothing.
Thus function, for instance, Tintin’s quiff hairstyle and clothes—brown
plus fours and brown shoes, to which are later added a white shirt, a blue
sweater, and white socks—Corto Maltese’s captain’s hat and sideburns,
the superheroes’ costumes, Mickey’s ears, Donald Duck’s beak, and the
small indications of individuality that help us to distinguish between
the Smurfs.
The caricature-like and ‘cartoony’ presentation of characters allows
comics to focus on the essential in the narrative situation at hand, such
as action or the character’s mental state. Scott McCloud has called the
simplified reality of cartooning “amplification through simplification”
(McCloud 1993, 30).25 Visual simplification and exaggeration are also
means of focus. Exaggeration helps to emphasise a trait, mental state,
an idea or the nature and impact of an action, for instance. In tradi
tional Disney comics, it is usual that when characters are shocked or
surprised, their hats jump in the air; a loud noise or shout can throw
things around the room; and hits or blows to the head cause bumps
that protrude impressively. In Mickey’s Inferno, hitting and hurting are
frequently accompanied by star symbols; decisiveness and persistence by
puffs of smoke; shouting by drops of flying saliva; while fast movements
are indicated by multiplying the same figure in the panel, and so on. Big
bad Zeke Wolf also manages to run around as a mere clothed skeleton
and a head when all his flesh is blasted from his body.
The demands of graphic drawing and style, the use of caricature, and
the rich symbolic language of comics, easily compromise the sense of life
likeness in characters. The use of photographic models, or the attempt to
represent characters and reality in the photographic sense, 26 are relatively
common, but usually they occur in combination, from Hergé and Jack
Kirby, Alex Raymond and Garry Gianni, to Alison B echdel, with the
cartoony techniques of simplification and exaggeration. There are many
reasons why the use of photographic representation and photorealistic
style is thus restricted. From the cartoonists’ perspective, the systematic
use of photorealistic style can be time-consuming and expensive. At the
same time, photorealistic characterisation and reality-building has
the potential to undermine idiosyncratic cartooning that draws on the
cartoonist’s individual style of the graphic trace—a strong expectation
in the medium.
The necessity to give characters a particular visual shape clearly dis
tances comics from literary narratives. In comics, as in picture books,
external description of characters and their action can be both verbal
and visual, and can either confirm or contradict each other. Psychologi
cal description can work both through words and images. Mental states
may be suggested in many ways in pictures, as we have already seen, but
their representation may also need, as Nikolajeva and Scott argue with
regard to picture books, “the subtleties of words to capture complex
emotion and motivation” (2001, 83). At the same time, verbal exter
nal description, especially in terms of describing the character’s physical
appearance and environment, becomes superfluous. Visual description
is simply more efficient for this purpose, while speech as a means of
characterisation is verbal by definition (Nikolajeva and Scott 2001, 83).
Generally speaking, however, the visual means for depicting characters’
mental states are more versatile in comics than in picture books. This
is partly due to the possibilities of the sequential form. For example,
a sequence of panels can depict complex mental states by contrasted
perspectives, or by showing evolving emotional states in the characters’
facial expressions and body language. Moreover, action and speech, or
action and verbal description, are much more closely connected, both in
a spatial and conceptual sense, in comics than in picture books. This is
not an absolute rule, obviously, but a convention and expectation.
Conclusion
In the beginning of this chapter, I emphasised that characters do not
need to be the sole focus of narrative comics, since the story’s centre can
also be an incident, event, plot, theme, or sometimes a space or an idea,
for instance. Yet, the lack of indications of character and mental state
does point to certain narrative limitations in comics. The concept of
character is tied to a larger generic frame: the expectations and qualities
of narrative fiction. In nonfiction, and in lyrical and abstract comics, the
character category may be irrelevant, even if the representation of men
tal states or experience remains vital. In first-person narration in fiction,
the relevance of the concept may also be limited, especially if the story is
wholly focussed on the narrator-character’s personal experience.
As we have seen, comics can employ a wide scope of artistic devices
for ascribing mental states and personal traits to characters. Such de
vices include, among other things, the visual description of physical ap
pearance, facial expressions, gesture and behaviour, speech and thought
representation, the depiction of action, and the narrator’s or other char
acters’ descriptions. Slomith Rimmon-Kenan’s emphasis that any ele
ment in literary narrative fiction, in principle, may serve as an indicator
of character (2003, 36), that is, will further a sense of a character’s per
sonality and mental state, is perfectly applicable to comics storytelling,
provided that we consider the function of images and the interaction
between words and images in this process. The distinctions between a
person and a thing, personal and non-personal entities, or human and
(imagined) non-human experiences are crucial. We may also perceive the
constant violations of these distinctions by means of caricature as an
other proof of the importance of that distinction. 27 A character may lack
cognition and sentience, at least in terms of what may be expected from
humanlike characters, have a highly unnatural appearance (fantastic,
invisible, or other) or, indeed, look like a thing. The character of Valise
illustrates how the mere attribution of voice to an inanimate object,
combined with a role in action, can turn an entity into a believable and
rather complex character in comics storytelling.
The importance of external visual characterisation in this respect can
hardly be overestimated. A few lines only can create characters, the art
of caricature has the capacity to freely shape the characters’ bodies for
expressive uses, and graphic line and style offer a very efficient means
for showing signs of inner life in the character’s appearance, behaviour,
and action. Sometimes characterisation can also be based on a con
trast between visual and verbal information: a box-like robot can be
philosophical; Mickey Mouse can become a medieval poet. It is then
important to evaluate how much weight the reader gives to either visual
and verbal narration and how much our understanding of a character’s
mental state, or person-like inner life, is based on verbal information.
Although it may be much easier for literary fiction to develop the im
pression of a character’s inner life—think of Marcel Proust or Robert
Musil, for instance—comics can use most of the same verbal means in
character description, and circumvent some of the restrictions posed by
their spatial form through visual and multimodal techniques.
My choice of these examples in this chapter was deliberate in that
I wanted to give more attention to the relation between type and indi
viduality instead of focussing on graphic novels, such as Tamara Drewe
and Nao of Brown, with strongly individualised characters. This was in
order to better perceive how different strategies of caricature and typifi
cation may result in varying effects of flatness, rotundity, and complex
ity. The crucial point here is that characters in comics may suggest a
great variety of dynamic relations and combinations between type-like
and individual-like qualities.
A certain set of basic expectations about what characters are and how
they exist are similar to comics and other narrative media, such as liter
ature and film. Characters in comics can be expected to have inner and
outer qualities, and they have consistency, for instance. Yet, the outer
forms of the protagonists in the Dante adaptations—anthropomorphic
animals, a robot, and a person seemingly without an inner self—also
suggest that many aspects of human experience and embodiment may
not be that relevant here. As in much fiction, characters in comics are not
always processed as if they were real people, and the properties ascribed
to them do not need to conform to actual world regularities (Margolin
2007, 68). In addition, graphic drawing and style, the use of caricature,
and the rich symbolic language of comics set limits to lifelikeness in
characters. A wide range of options exists between roughly outlined car
icature and detailed photorealism in comics, but the general tendency is
to allow room for caricature and the subjectivity of the graphic trace.
While it is certainly possible for readers to identify with characters in
comics, or find them truly convincing as persons, especially in today’s
graphic novels, the history of the medium is marked by a fundamen
tal distance between real-life people and caricature. Our evaluation of
the reality and complexity of human-like characters in comics, however,
is also always a matter of the larger context of reception, such as ge
neric expectations, as the rich variety of today’s nonfiction comics so
well illustrates.
Notes
1 Mark Blackwell defines it-narratives as “prose fictions that take inanimate
objects or animals as their central characters, sometimes endowing them
with a subjectivity—and thus a narrative perspective—of their own, some
times making them merely the narrative axes around which other characters’
stories spin” (2012, vii). I am thankful to the members of the Comix-
Scholars discussion list, who responded to my inquiry about it-narratives in
the spring of 2015.
2 These three works are among Groensteen’s (1999, 19) examples of how to
bypass the presence of a recurrent character in comics and still tell a story
(1999, 19–20).
3 Crumb has later added other future scenarios to this work: The Fun Future,
Ecological Disaster, and The Ecotopian Solution.
4 See also comics included in the anthology Comics as Poetry. Ed. Franklin
Einspruch. New Modern Press, 2012.
5 David Herman has referred to the latter as models of self or personhood
(2013, 134, 195).
6 Genette argues that “it seems to me that by allowing the study of characteri
sation to have the privilege of shaping, and thereby governing, the analysis of
narrative discourse, we make too much of a concession to what is only one
‘effect’ among others” (1988, 136). The category of the character has, how
ever, a key role in Shlomith Rimmon-Kenan and Mieke Bal’s classical nar
ratologies, and more recently, narratological theory of character has made
important advances in Uri Margolin’s work.
7 In this regard, the historical conception of the ‘disintegration’ of the classi
cal character in the history of modern literature—a character with a proper
name, physical and moral nature, and distinct from the narrator—and that
Genette associates with Marcel Proust’s Recherche (1980, 246–247), also
requires us to hold onto this concept.
8 Phelan further connects the three dimensions with a scale of character func
tions and readerly interests in the narrative, respectively, called the mimetic,
thematic, and synthetic function or interest. See Phelan (1989, 9, 1996, 29)
and Phelan and Rabinowitz (2012, 113).
9 Henry James’s much-used dictum, from “The Art of Fiction”, about the in
terdependence between characters and action remains relevant: “What is
character but the determination of incident? What is incident but the illus
tration of character?”
10 Jimbo’s Inferno compiles seven Jimbo comic books, published by Matt
Groening in his Zongo Comics, 1995–1997, and that are reformatted to the
format of the sequel, Jimbo in Purgatory.
11 Jimbo’s adventures were first chronicled as a comic strip in the 1970s LA
punk paper Slash and later in RAW magazine.
12 Basically, any deictic linguistic marker may identify the agent of speech,
thought, or action as long as the story contains sufficient information about
the context of the depicted speech, thought, or action.
13 We must note, however, that some of these principles are also techniques of
characterisation. What Bal calls ‘accumulation’ can be perceived as an effect
of repetition, but also a means of characterisation. Rimmon-Kenan’s ‘impli
cation’, similarly, comprises various implicit means of character-building,
such as revealing personal traits through action.
14 Edward Gorey explores this freedom in his “The Inanimate Tragedy”,
which parodies the drawn inanimate figures’ supposed sentient, cognitive,
emotional, and verbal qualities, and where the characters include No.37
Penpoint, The Four-Holed Button, The Glass Marble, The Two-Holed
Button, The Half-Inch Thumbtack, and Needles and Pins.
15 In cross-media comparisons of this kind, it may also be forgotten that lit
erary fiction is not always focussed on characters as individuals or on their
psychology, but on types or behaviour.
16 Lack of change does not necessarily make characters flat, however
(Rimmon-Kenan 2003, 41; Bal 1997, 117). Much modern literary fiction
portrays, from Dickens’s minor characters such as Harold Skimpole in Bleak
House to James Joyce’s Leopold Bloom, characters who lack development,
but exhibit much depth.
17 Forster claims, for instance, that in Persuasion, Jane Austen was capable of
inflating a flat character into a round character and collapsing her back into
a flat in one sentence (1953, 73).
18 I would like to thank Mark Shackleton for this point, and also Essi Varis for
her many valuable comments on this chapter.
19 Compare with Alex Woloch’s notion of ‘distributed attention’, which refers
to the attention that a specific character has in the course of the text, and
character-space. For Woloch, character-space means “that particular and
charged encounter between an individual human personality and a deter
mined space and position within the narrative as a whole” (2003, 14).
20 Rimmon-Kenan’s ‘flat’ pole on the axis of complexity includes characters
constructed around a single trait or around one dominant trait along with
a few secondary ones, such as allegorical figures, caricatures, and types
(2003, 41).
21 As Phelan has shown in many of his studies on characterisation. See, for
instance, Phelan (2007, 222).
22 See Livingston (1998, 80–81) and Louwerse and Kuiken (2004, 171).
23 Joseph Lanzara’s Dante’s Inferno: The Graphic Novel (2012) utilises
Gustave Doré’s illustrations of the Divine Comedy to tell the story, and
Robert Sikoryak’s Inferno tells Dante’s classic in one page as a “Bazooka
Joe” in Masterpiece Comics (2009). See also Tondro (2015).
24 Hugo Pratt, the creator of Corto Maltese, has turned some of his graphic
novels into prose fiction.
25 For a more historical and technical perspective on the graphic definition of
character in modern comics, see Morgan (2009). Morgan emphasises, es
pecially, the role of reproduced hand-drawn (or written) style and the swift
creation, or speed-drawing, in portraying characters as types.
26 Detailed description does not always serve the effect of realism. The amount
of detail in the description of Hell in Martina and Bioletto’s Mickey’s Inferno
is impressive, but this is not realistic as such since no one knows what hell
looks like. Instead, detailed description emphasises the unique qualities of
this world, while it also refers to the visual imagination of Gustave Doré,
and his version of that world (1861).
27 David Herman argues that the distinction between a person and a thing, or
between personal and non-personal entities, is anchored in humans’ embod
ied experience, but is also shaped by circulating cultural models of what a
person is, and of how persons relate to the world at large (2013, 193–194).
The division is, thus, a persisting one.
Part IV
Speech and Thought

in Narrative Comics
8 Presenting Minds in Comics
Narrative theory has recently shown increasing interest in the representa

tion of consciousness in narratives. Some approaches, especially those that
redefine narratives in cognitive terms, have emphasised the centrality of
the mediating mind for all narrative understanding. For instance, Monika
Fludernik demonstrates that narrativity, a set of properties that character
ise narrative, fundamentally involves the presentation of a consciousness,
reflecting on and reacting to events, and evaluating them. In developing
her influential concept of narrative experientiality, Fludernik suggests that
any extended narrative relies on the experiential portrayal of sequences of
events and human consciousness (1996, 28–30). In the multimodal envi
ronment of comics, Alan Palmer’s idea of mental functioning as action
in narrative fiction (2004) provides a compelling approach by developing
Fludernik’s concept further. Palmer problematises what he calls the “ver
bal norm” in the predominant speech-category approach to narratologi
cal analysis of characters’ thought processes. By the verbal norm, Palmer
specifically means the preoccupation with the highly verbalised flow of
self-conscious inner speech and thought (2004, 14, 63–67), and free indi
rect perception where the narrator uses a character’s consciousness as the
perceptual angle on narration (2004, 48). Palmer contends that the verbal
bias in narratological research has favoured modern literary fiction that
employs techniques of inner speech and thought, instead of more indirect
means of thought presentation, and has contributed to a limited notion of
thought and mind in narrative analysis.
The analysis of mind presentation in comics storytelling might sug
gest ways to loosen the grip of the ‘verbal norm’ in narratology, while
also helping us to evaluate what might work across narrative media in
the speech-category approach. What makes comics, and their medium-
specific constraints and preferences, especially interesting in this respect
is that the medium stimulates the viewer’s engagement with the minds of
characters by recourse to a wide range of verbal modes of narration in a
dynamic relation with images that show minds in action. Many aspects
of this multimodal interaction between words and images are similar to
the way in which film narratives function, for instance in the relatively
limited use of narrators in both media. Various cognitive approaches in
204 Speech and Thought in Narrative Comics
film studies have also examined the audience’s engagement with char
acters whose minds and emotions are predominantly displayed through
images, with little or no direct access to the characters’ verbal thoughts.1
Yet comics storytelling is also fundamentally different from film narration
in this respect.2 In regard to speech-act categories, further challenges to
narratological analysis of mind presentation in comics are posed by the
spatial demands of the medium, which specifically limits the use of certain
categories of speech and thought that are common in literary fiction. For
instance, extended employment of inner speech, direct thought presenta
tion, and thought report is relatively rare in comics. In this chapter, I shall
focus on three issues pertaining to the presentation of minds in comics and
graphic narratives: the mimetic aspect of the image, the problem of free
indirect discourse (FID), and the interaction between visual focalisation
and verbal narration in first-person narration. My theoretical discussion
will be illustrated by some representative extracts from comics, moving
from third-person to first-person narration, both in fiction and nonfiction.
It needs to be emphasised, however, that what I refer to as first- or third-
person narration in comics is not exactly the same as in literary fiction.
The use of such grammatical persons to describe particular modes of tell
ing is compromised by the need to base them principally on the mode of
verbal telling, i.e. the continuing presence of a narrative voice, who can
be identified as a character or a narratorial voice. Visual narration can
always take considerable distance from the character-narrator’s voice and
realise a third-person framework on the visual level. At the same time,
and as was illustrated in the previous chapters of this book, comics have
various visual means at their disposal for subjectifying the image, regard
less of the identity of the narrative voices (first- or third-person narration,
or no narrator), which can also affect our understanding of the degree
of subjectivity in the narrative voice. Equally, the linguistic categories of
direct, indirect, and free indirect discourse, which describe the grammati
cal construction of speech and thought representation, need to be rede
fined appropriately in the multimodal environment of comics.
The Mimetic Image and the Redundant Word

Among the most obvious constraints of verbal mind construction in
comics is contextual character portrayal: the rendering of a character’s
thoughts and sensations with the help of the physical context. The rea
sons for this are two-fold. The narrator’s verbal report of a character’s
thought, feeling, and perception may be redundant with regard to the
visual content and information in the images. On the other hand, re
porting syntax can create an arbitrary shift in the narrative from images
to words, at the risk of turning the images into illustrations of the text.
These restraints are especially pertinent in the case of third-person nar
rative comics, but they also affect first-person varieties.
Presenting Minds in Comics 205
In classical French-language third-person perspective comics and
graphic novels, from Rodolphe Töpffer to Hergé, the presentation of
a character’s thought as mental action is realised predominantly in im
ages, distributed through what is seen on the page, often from an im
personal viewpoint. In Töpffer’s Monsieur Vieux Bois (“The Story of
Mr Wooden Head”, 1827), the narrator’s reporting voice is quite domi
nant and often mentions things that can be seen in the images. Yet even
here, it is the visual image that directs the reader’s attention to the social
situation, justifies action, and embeds a character in a physical setting.
In two panels from the story (Figure 8.1), we first see the main character,
Mr Wooden Head, asking for his loved one’s hand in marriage at her
parents’ home, and then we see him in his own house, jumping for joy as
his proposal has been accepted. All of his furniture seems to have been
overturned. The juxtaposition of these two images, if we look only at
the visual information given in the two panels, suggests an intense posi
tive experience. The physical shape of the body, body language, facial
expression, and scenery are shown rather than described. In contrast,
what the images cannot fully reveal on their own is the precise meaning
of the visit to his loved one’s house, the difference of location (it is possi
ble to imagine that he is jumping in the same room as in the first image),
the precise meaning of his emotions while he is jumping (joy), and the
duration of this jumping (3 hours).
In contrast, many features of time condensation or expansion that
Dorrit Cohn associates specifically with reported speech in modern
literary fiction—or what she calls ‘psycho-narration’, “the narrator’s
discourse about a character’s discourse” (1978, 14)—can be shown
in comics storytelling directly through images, without resorting to
words. The different functions of summary, its iterative, durative, and
Figure 8.1 Rodolphe Töpffer. Monsieur Vieux Bois (1827).

mutative rhythms or patterns of recurrence, persistence, and gradual
change (Cohn 1978, 35), can be established through purely visual cues
or through the narrative breakdown between the panel images. For in
stance, in an open-frame panel from Hugo Pratt’s Corto Maltese. La
Ballade de la mer salée (The Ballad of the Salty Sea) 1967–69, we see
the repetition of the same information in a chain of changing interlocu
tors. The verbal content of the information—that a white young woman
has been seen with a Maori man on the island—need only be related
once. Similarly, the sense of the expansion of the moment, often realised
in reported thought in modernist literature, can be achieved by purely
visual means, for instance by enlarging the size of the picture panels or
by eliminating the panel frame. Twentieth-century comics invented such
visual conventions and devices to portray repetition, temporal duration,
and spatial change in order to avoid the use of a sustained narrative
voice. The expressive function of the panel frame’s shape, size, and fo
calisation, pointing to changes in narrative space and time, as well as
in the character’s state of mind, was already part of Töpffer’s narrative
inventiveness. Changes in the panel size and focalisation between the
images also accentuate changes in the narrative focus: the move from
the polite encounter between suitor and parents in the living room to the
expression of intensive private joy in one’s own room. The second image
brings the man closer to the reader’s view, showing him from the front
so that we can focus on the expression on his face.
In written narratives, the indirect discourse of reported thought and
speech are often indistinguishable from scenic description. For exam
ple, when a character sees or hears something, the description of that
sensory experience, introduced by verbs of perception, links his or her
psyche with the scene. 3 In comics, verbal scenic description is usually
redundant, as images show scenes all at once, creating the effect of
panoramic views, or presenting a character gazing at something and
reacting to it. Verbal “telescoping” can of course serve an interpretive
purpose, as a pointer to the image and its details: in Töpffer’s stories,
the narrator’s comment actually accompanies each image. An extended
use of verbal report of a character’s emotions and thoughts, however,
can slow down the visual narrative flow considerably. The presentation
of abstract thoughts and prolonged soliloquies poses special challenges
to comics in this light. Reported thought is generally restricted to lo
calized use, as in a scene from the epic western Marshall Blueberry
(Le Spectre aux balles d’or/The Ghost with the Golden Bullets) 1972,
where the narrative box summarizes the character’s diffuse feelings
(Figure 8.2). The narrator explains that the dry basin just discovered
by the character Prosit Luckner is already strangely familiar to him.
The sensation of familiarity, the narrator explains, contributes to a
growing sense of fear and the emergence of memories from the shady
corners of his mind.
Figure 8.2 Charlier & Giraud, Blueberry: Le Spectre aux balles d’or (1972)
© Bluberry – tome 12, Le spectre aux Balles d’or DARGAUD by
Giraud & C
harlier. All rights reserved www. dargaud.com
Third-person narration that focusses on dialogue and action easily

allows occasional lapses into a character’s monologue, direct thoughts
in thought balloons, or passages of reported thought that can have the
narrative function of informing the reader about certain facts in the
story (Saraceni 2003, 66–67). For instance, in Jason Lutes’s Berlin
(2000–2008), the shifts in visual perspective are further complicated by
the way that verbal narration, in a number of localised manifestations,
shifts into the direct discourse of character-narrators. Two characters,
Kurt Severing and Marthe Müller, take turns as narrators, speaking
as if directly through quotations from their diaries; in such passages,
thought report is also used. Indirect discourse of other characters’
speech and thought may also be the principal means of storytelling, for
instance when this is characteristic of the personality and activity of
a character-narrator, or a first-person narrative voice. In the first vol
ume of Joann Sfar’s five-volume The Rabbi’s Cat (Le Chat du rabbin. 1.
La Bar-Mistva, 2002), the character-narrator is a cat, whose narrating,
describing, and reporting voice in the captions frames the whole story.
His discourse is mostly in present tense, thus reporting what his master
the rabbi, the rabbi’s daughter, the rabbi’s rabbi, or other people say,
think, and do in the given scenes, as well as explaining his own current
emotions, thoughts, and plans. The cat’s initial inability to speak justi
fies, in a sense, the narrative situation, where indirect discourse clearly
outweighs direct speech. At first, the cat can supposedly only understand
what the humans say but cannot speak back—thus, it is logical for him
to have a kind of ongoing conversation in his head—but after having
eaten the rabbi’s parrot, he also begins to speak. While dialogue is given
more room in the subsequent albums, the cat’s thought and speech re
port, nevertheless, remains prevalent in the story.4
In classic twentieth-century third-person comic albums, narrative
comments are much more severely limited. The most common means for
verbally presenting minds is dialogue, in interaction with visual infor
mation. The ironic treatment of dialogue in Tintin au Tibet (1960), for
instance, where Sherpa Tharkey calls for the yeti in the mountains of
Tibet and gets no answer, and Captain Haddock shouts to his own echo,
presents the characters’ minds in action. The scene is yet another in
stance of dialogue, and of action through reaction, that reveals some
thing of the puzzling workings of the captain’s mind. After Tintin has
explained that what they have heard is only the echo of Haddock’s voice,
the captain still insists that the echo is a person with whom one could
have a conversation (“The echo! No one asked him anything, the one up
there!”). The interaction highlights the captain’s inclination to respond
to any verbal stimulus, to take every word literally, and to assume that
there must always be a speaker behind every word.
The localised use of the direct discourse of inner or silent speech is
relatively common in classic French-Belgian third-person examples. Un
like reported thought, which clearly belongs to the margins of the im
age, or to the narrative box detached from the image, inner or silent
thoughts can be placed within the picture panel with visual markers of
its own, mainly by the thought balloon. Most typically, in the Western
tradition of comics, the thought balloon is connected to the character
by a line of smaller bubbles, instead of the arrow-like tail common in
speech balloons. The thought balloon forges a double link between the
text and the image in that it simultaneously constitutes a message while
its presence, position, and visual shape also convey information relating
to the character’s state of mind. The example above from Blueberry in
volves an instance of inner speech, indicated by the thought balloon. In
the balloon, we hear Prosit’s direct reaction to what he sees. The bold
letters of the sentence “au pied de ce monolithe, que j’ai…” (“at the foot
of this monolith that I…”) indicate the most important and emotionally
charged content of these thoughts, the recollection of what the charac
ter has done at the foot of the monolith. The possibility of integrating
inner speech and thought within the graphic image, while indicating the
emphasis or other qualities of the thoughts, makes this speech category
convenient for the medium. Any extended use of this mode, however,
naturally alters the balance between the visual and the verbal compo
nents in favour of the latter.
Thus, third-person behaviourist comics like Tintin and Corto Maltese,
a genre still predominant in the medium, very rarely explore characters’
psyche through indirect verbal discourse like reported thought.5 Instead,
these narratives resort to direct verbal discourse through dialogue and in
ner speech, as well as multimodal and visual narrative techniques, which
we have discussed earlier in this book (graphic showing, focalisation).
Free Indirect Discourse in Comics

To examine some of the medium-specific features and limitations of in
direct speech and thought in comics, I would like to turn to two vastly
different adaptations of Gustave Flaubert’s novel, Madame Bovary.
Posy Simmonds’ graphic novel Gemma Bovery (1999) is a self-reflective
rewriting of the novel set in our contemporary world, and it follows
Flaubert’s story only partially. Bardet and Janvier’s adaptation in the
album format Madame Bovary (2008), in contrast, faithfully quotes
Flaubert’s language and closely follows the order of the events in the
novel.6 My focus in this comparison will be on the literary style of free
indirect discourse (FID), and the accompanying technique of shifting
viewpoints (or focalisation) between internal and external positions,
for which Flaubert’s novel is so well known. FID entails, without being
clearly internal or external, both a reference to the perceiving subject
(the character) and the narrating instance (the narrator).
Before moving onto a more detailed textual analysis of the examples,
however, we must examine the difficulty of transferring FID to visual
storytelling and multimodal texts, where the scope of the verbal element
is limited and combined with other kinds of signs. Analysing the major
film versions of Madame Bovary, Mary Donaldson-Evans underscores
the difficulty in transferring FID to the visual medium:
Whereas the shifting narrative viewpoint created by Flaubert’s use

of this narrative style can be easily transferred to the screen, the
camera naturally embracing the perspective of different characters,
the simultaneous representation of two viewpoints and the ambigu
ity and irony that often result are difficult to convey in film, where
differing perspectives must be represented sequentially.
(2009, 31–32)
The only exception that Donaldson-Evans can think to this rule of

non-simultaneity of different perspectives in films is “the use of a voice
over narration that mocks the picture on the screen” (2009, 32, n3).
While the ambiguity that can result from the use of FID in literature
is hard to convey in film, the claim about perspective-taking in cinema
seems to rest on questionable premises.
First, it is not clear how literal simultaneity in viewpoints could
be achieved in the temporally organised order of literary narratives.
Changes in the windows of focalisation in Flaubert’s Madame Bovary
are presented sequentially, as the narrative moves from Charles to Emma
and back (or between other characters and the extradiegetic narrator).
FID can maintain, by linguistic means, the third-person reference while
reproducing the character’s own language. This relation of voices may
remain ambiguous as to the source of the words or, equally, constitute a
clear hierarchy where the narrator cites some character’s words without
quotation marks. Still, to equate the relationship between the narratorial
context and a character’s voice with ‘perspective’ or ‘simultaneity’ in a
literal sense is quite another matter. The various theories of ‘embedded
focalisation’ or ‘dual focalisation’ that have been proposed in literary
narratology usually try to come to terms with the temporal ambiguity of
this concept.7 Donaldson-Evans seems to uncritically collapse voice and
focalisation, perhaps misled by the fact that in literature, our interpre
tation of these elements is based on the same linguistic features in the
text. The separation between these two structures of narration remains
important in novels where the perspective, which is usually closely tied
to the characters, keeps shifting and where the narrator has varying de
grees of presence.
Second, there are other pertinent disparities between media. In visual
storytelling, and in the spatial arrangement of comics in particular, sev
eral focalisers can appear simultaneously. For instance, fields of vision in
a panel can include a sense of a subjective vantage point without being
limited to it, thus violating clear boundaries between external and in
ternal perspectives. In Bardet and Janvier’s version of Madame Bovary,
the close-up of Charles’s eyes, as he waits for a sign as to whether Emma
wants to marry him, brings us close to Charles’s vision (Figure 8.3). The
subsequently represented field of vision then embraces the character’s
viewpoint. The illusion of subjectivity in this sequence is based both on
the juxtaposition of the perspectives in the first image—the focus on the
man looking and his gaze, which is directed at something that we do not
see—and on the shift in perspective between the panels.
In fact, many common techniques of embedded point of view in films,
including ones that are also commonly used in comics, have been likened
to FID. For instance, Charles Forceville has suggested that studying cine
matic techniques like character-bound camera movement in terms of FID,
specifically when such techniques create ambiguity between the external
“narrator” and the character, could contribute to a transmedial narra
tology (2002, 133). Another use of embedded viewpoints is the over-the-
shoulder shot, in which the external focaliser in a narrative watches “along
with a person”, but without leaving focalisation entirely to this character
(Bal 1997, 159). The technique, as Mieke Bal has suggested, is compara
ble to FID, in which the narrative discourse approximates a character’s
voice without letting him or her speak directly (1997, 159). I believe these
to be interesting suggestions for the study of comics and visual narratives
in general, as long as we are ready to modify our understanding and
Figure 8.3 M adame Bovary par Daniel Bardet et Michel Janvier © Éditions
Adonis, 2008.
use of speech-act categories in the multimodal and mixed-media environ

ments. However, these cases chiefly represent forms of visual ambiguity
even if they may filter a sense of subjective perception and cognition, and
not ambiguity in the sense of speech and thought.
All these (and many other techniques) can accommodate the effects
of FID on the visual plane as they combine a character’s viewpoint with
some more objective perspective or manifest one character’s thoughts
inside another’s. Visual narratives have, in fact, so many means at
their disposal for creating embedded perspectives that they are vital,
in adaptation, to compensate for the limitations the narratives face
in the representation of thoughts by visual means. To better see where
the medium-specific constraints lie in this respect, it might help us
to distinguish between free indirect speech or thought and free indi
rect perception. In literature, free indirect perception, sometimes also
called represented perception, means the presentation of a character’s
non-verbalised perceptions as they occur in his or her verbalised con
sciousness (Prince 2003, 35). Visual narratives, by contrast, are charac
terised by their capacity to show fields of vision as if they were someone’s
non-verbalised perceptions. However, at the same time, it is challeng
ing to describe the subjective meanings of these perceptions by visual
means alone. Thus, the difficulty in transferring novelistic techniques of
thought representation to comics does not lie in the simultaneity of the
viewpoints—which may, in fact, be more viable in the spatial arrange
ments of visual narratives than in literary discourse—but the way that
certain techniques of indirect speech and thought, such as FID in liter
ature, can give the reader access to some subjective psychic reality and
mediate that access through narratorial sympathy or irony. In F laubert’s
Madame Bovary, FID typically reflects a subjective mind onto the ob
jective structure of the language of narration, emphasising expressions
of emotion or the irony in the mind’s illusion.
Take the passages towards the end of the first part of Flaubert’s novel,
when Emma is growing more and more dissatisfied with her country doc
tor husband and their life in the commune of Tostes. Here, her thoughts
and emotions are given in alternating stretches of reported thought and
FID, while the perspective is mediated through internal focalisation:
Emma le regardait en haussant les épaules. Que n’avait-elle, au moins,

pour mari un de ces hommes d’ardeurs taciturnes qui travaillent la
nuit dans les livres, et portent enfin, à soixante ans, quand vient l’âge
des rhumatismes, une brochette de croix, sur leur habit noir, mal
fait. Elle aurait voulu que ce nom de Bovary, qui était le sien, fût il
lustre, le voir étalé chez les libraires, répété dans les journaux, connu
par toute la France. Mais Charles n’avait point d’ambition!
(Flaubert 1993, 76)
Emma looked at him and shrugged her shoulders. Why, at least, was
not her husband one of those men of taciturn passions who work at
their books all night, and at last, when about sixty, the age of rheu
matism sets in, wear a string of orders on their ill-fitting black coat?
She could have wished this name of Bovary, which was hers, had
been illustrious, to see it displayed at the booksellers’, repeated in
the newspapers, known to all France. But Charles had no ambition.
(Trans. Eleanor Marx-Aveling, Flaubert 1946, 69–70)
The viewpoint is clearly Emma’s. At the end of the passage, further,

the exclamation point functions as an emotion marker, suggesting a
shift from reported thought to free indirect thought. Some telling trans
mutations to the narrative mode occur in Bardet and Janvier’s adapta
tion of this scene. The line “Mais Charles n’avait point d’ambition…”
(“But Charles had no ambition…”), now accompanied with an ellipsis,
becomes part of the narrator’s thought report. A few panels later, an
other passage in FID, this time in Charles’s perspective, is shortened and
transmuted into a brief dialogue between Emma and Charles:
Il en coûtait à Charles d’abandonner Tostes après quatre ans de sé

jour et au moment où il commençait à s’y poser. S’il le fallait, cepen
dant! Il la conduisit à Rouen voir son ancien maître. C’était une
maladie nerveuse: on devait la changer d’air.
(Flaubert 1993, 78)
It cost Charles much to give up Tostes after living there four years
and “when he was beginning to get on there.” Yet if it must be! He
took her to Rouen to see his old master. It was a nervous complaint:
change of air was needed.
(Trans. Eleanor Marx-Aveling, Flaubert 1946, 69–70)
The predominant narrative mode in Flaubert’s novel is what F.K. S tanzel

has called a ‘figural narrative situation’, 8 centred around the main
reflector-characters, Emma and Charles. This entails that the readers
are allowed to access their inner world, but this access is mediated by
the narrator, as indicated by the fact that the characters’ reflections are
largely cast in language that they would not use. There are only brief
instances of direct thought where the main characters articulate their
thoughts.
As the transmutations above suggest, it is difficult to find equivalents
in comics storytelling for narration that severely restricts the use of di
rect discourse, such as dialogue or inner speech and thought, and pre
fers more indirect forms of speech and thought representation. Bardet
and Janvier’s version of Madame Bovary simply dispenses with the free
indirect passages even if, in other respects, their version strives to be as
loyal to Flaubert’s language as possible. The outcome is a neat distinc
tion between the direct discourse of the speech and thought balloons
on the one hand and the indirect narratorial discourse of the narrative
boxes on the other hand. There are only a few brief exceptions to this
arrangement.9
Bardet and Janvier’s adaptation radically reduces the variety of tech
niques of representation of thought and speech in Flaubert’s novel. This
does not concern only FID, but all more complex cases of direct dis
course or shifts between different modes for speech and thought repre
sentation as well. To take another example, we may think of the scene
in which Rodolphe writes a letter to Emma to dismiss her. In the novel,
the scene is narrated through a complex alternation between indirect
thought report, FID of the writer’s reflective consciousness, and direct
discourse, including Rodolphe’s speech to himself and thoughts or quo
tations from the letter that he is writing. In the comic book version of
the same passage, there are quotations from the letter in the panels and
some brief thoughts given in direct discourse and thought report. The
narration thus becomes much more straightforward, as if underlining
the importance of the story (order of the events) over the discourse (order
of presentation) in adaptation. Certain visual elements in this scene,
nevertheless, manipulate the distinction between internal and external
perspectives in a way that may remind us of the effects of FID. In parti
cular, the incorporation of Rodolphe’s handwriting and the showing of
his hands in the foreground subjectify the perspective, although they do
not render it fully subjective.
In contrast, Posy Simmonds’ adaptation of Madame Bovary as
Gemma Bovery takes advantage of a great variety of types in thought
and speech representation. A crucial structural element that makes this
variety possible is the employment of a consistent first-person narra
tor, who is a witness to the main events of the story, while his personal
perceptions are at the core of the narration. With regard to the tech
nique of FID, however, both of these versions avoid it. In Simmonds’
Gemma Bovery, the main verbal mode in thought representation is the
narratorial report of Gemma’s diary, which is interrupted by frequent
direct quotations from her writing or the narrator’s reflections and ob
servations. What is quite different from Bardet and Janvier’s Bovary
is the use of a homodiegetic witness-narrator, accompanied by the
‘objective’ focalisation in the images, which allows smooth perspec
tival changes, and changing degrees of irony between the narrator,
called Raymond Joubert, and the other characters. Despite the focus
on Gemma’s adventures—her marriage, exile, and adultery—Joubert’s
personal experiences, perceptions, and evaluations of Gemma’s be
haviour remain at the core of the story, creating a continuous “voice”
that frames the story and the quotations. What adds to the irony is
that the narrator is obsessed with what he sees as coincidences in his
British neighbour Gemma Bovery’s life and that of her near name
sake, Emma Bovary. Thus, irony, unlike in Flaubert, extends to the
narrator himself, particularly with regard to amorous Joubert’s spying
and infrequent obtrusiveness, such as when he sends, out of jealousy
or due to his delusions of coincidence between Emma’s and Gemma’s
life, anonymous letters to Gemma and her friends in order to influence
their behaviour.
The narrator’s comments, interpretations, and summaries in direct
discourse, concerning the events or Gemma’s diary, make an excep
tion to the general divide between direct and indirect discourses on
the verbal plane. These commentaries are marked off from the main
body of the narration by decreased font size. Other devices for marking
more direct speech or thought within indirect discourse is the inclu
sion of Gemma’s writing in the images, or the use of French in quoted
dialogue in the images, with the accompanying English translation.
The question of interlinguistic transaction becomes also occasionally
evident in Joubert’s difficulty with colloquial English expressions that
he cannot find in his dictionary. An even more important means of
manipulating the relation between direct and indirect discourses in
this narrative, however, is the visual perspective and visual showing
that give the reader information beyond what could be included in any
written diary, both in terms of visual details and people’s thoughts,
speech, and dreams. The images may, for instance, cast the narrator in
an ironic light.
Interaction of Verbal and Visual Narration
in First-Person Comics
With regard to first-person narration in literary fiction, the roles of the
narrator and the character may diverge and become more independent
from each other, for instance when the character-narrator tells about
things that he has not witnessed or seen. Gérard Genette has dubbed this
technique, where the narrator gives more information than is authorised
by the overarching form, paralepsis (1980, 195). Similarly, in third-person
figural narration, as happens with the narrator’s rhetorical questions in
Flaubert’s Madame Bovary, or in some descriptive passages in the novel,
the narrative may deviate from the main characters’ represented percep
tions to a clearly external and nearly omniscient perspective.
Comics storytelling may take advantage of a similar alteration in nar
rative mode, while it can rely on this disparity in a much more system
atic way: the voice is subjective, but the perspective in the images shifts
between various more or less objective and subjective positions. The di
vergence between the verbal narration of Joubert’s consciousness and the
perspectives in the images suggests varying degrees of distance between
telling and showing, the narrator’s voice, and his embodied existence in
the fictional world. The complex narrative situation in Simmonds’ graphic
novel, and one that is somewhat unusual for the medium, achieves similar
effects to that of FID in Flaubert’s Madame Bovary, making the role and
significance of the narrators’ involvement and moral stance central to the
story. Other examples of the difficulty in pinpointing the exact degree of
subjectivity in the images in Simmonds’ Gemma Bovery are the documents
that are included in the book, such as menus, restaurant reviews, letters,
notes, advertisements, magazine articles, or objects mentioned in Gemma’s
diary. To whom can we attribute these images? Are they part of the narra
tor’s direct discourse or paratexts that are neither direct nor indirect? Are
they shown as they were seen by the characters? Or does it really matter?
Alan Palmer (2004, 15) defines a ‘continuing-consciousness frame’ as
the reader’s creation of a sense of consciousness out of the isolated pas
sages of the text that relate to a particular character, and ‘thought-action
continuum’ as the idea that action and consciousness descriptions are
often inseparable in fictional narratives. These are useful tools for ana
lysing homodiegetic graphic narratives where the narrator is a character
in the story and where the narrator’s voice is sustained throughout the
story. For instance, in Guy Delisle’s travel memoir from North Korea,
Pyongyang: A Journey in North Korea, the narrator’s words provoke a
sense of a continuous ‘voice-over’ in the reader’s mind. Tensions between
verbal narration, scenic showing, and visual focalisation are crucial to
understanding the story. The narrative employs various means of vi
sual focalisation from the subjective perspective to the objective camera
eye. The story begins with the narrating ‘I’’s arrival at a North Korean
airport where no electric light can be used. The scene is first shown from
an impersonal, exterior perspective. It soon becomes evident, however,
that the viewpoint actually belongs to the passenger going through cus
toms, the narrator himself. The subjective viewpoint is finally evoked in
the panel where the eyeline is matched with the character, who is looking
at the shadowy man who greets him. The subjective perspective is then
affirmed gradually through several changes in the windows of focali
sation, from the external showing of the scene to the main character’s
angle of vision and finally to his verbal self-communion.
Another aspect that is significant in this passage, and typical of many
first-person comics today, is the presence of two first persons: the narrat
ing voice and the experiencing self who is seen. The voice and the view
of the narrating ‘I’ can coalesce in one panel, but they can then again be
severed: the narrator re-emerges in the following panels as a figure seen
from the outside. The transition is gradual so that we are first shown the
narrator’s shadow emerging from the side of the panel’s frame, as if in
between the interior and the exterior of the panel, whereas the next panel
shows him from an external viewpoint. It is important to note that nar
ration in Pyongyang builds on transitions of this sort between the narra
tive of the mind and the showing of the body, and the act of perception,
which is itself split between the perspectives of the experiencing and the
narrating self. A whole range of effects is produced by these transitions.
By contrast, in Paul Hornschemeier’s autobiographical fiction Mother,
Come Home (2003), there is an unusually sustained use of reported
thought, related to two specific features of the story: it helps to create the
temporal perspective of self-narration—the narrator’s investigation of his
childhood experience—and it contributes to the narrator’s concern about
the state of another person’s mind. Self-narration here manipulates not
only the distance between the narrating, verbal ‘I’ and the experiencing
self of the visual figure, but also the disparity between the narrator’s pres
ent self and his childhood self. The narrative presents the narrator’s child
hood self as if to assess it. The story of the death of the narrator’s mother
and his father’s subsequent deep depression, eventually leading to suicide,
is told through a perspective many years in the future. Reported thought
is employed frequently when the narrator tries to understand the diffuse
thoughts and feelings that he had at the age of seven, and the decisions
he made at that time. For instance, the narrator explains that one long
walk to the graveyard, which we see in the images, marked an important
turn in the boy’s understanding and experience. In his summary of these
memories, the narrator also elaborates on the cognitive and linguistic dis
parity between the narrating adult and the child. On several occasions he
points to the fact that as a boy he did not understand what was happening
around him, especially in the mind of his father. The child’s alienation,
and his cognitive and temporal distance from the narrating ‘I’, is further
visually dramatised by the lion mask that is often drawn on his face.
Conclusion
What does this survey of the strategies for presenting speech and thought
in comics reveal about the specific constraints and options for presenting
minds in this medium?
It shows that graphic narration subjects to doubt certain theoretical
presuppositions prevalent in much narrative theory that is based on lit
erary narrative fiction, and requires us to critically examine and redefine
various key distinctions, such as those between first- and third-person
narration, between direct, indirect, and free indirect discourses, and
between telling and showing. In a multimodal narrative comic, vast ar
eas of mind presentation are not suitable for analysis using the concepts
of narrator and voice or the speech-category approach. The question
of narratorial authority, enunciation, and control—Who tells the tale?
Who authorizes narrative meaning? Who controls it?—must also be
posed differently when speech categories interact with graphic images.
In reading and viewing visual narratives, we often see the mind in
action from a focalised perspective or through a figure in action. How
ever, it is often doubtful whether third-person comics involve a sustained
continuing-consciousness frame in the same sense as literary heterodiegetic
narration pertaining, for instance, to varieties of “omniscient narrator”.
The framing and the focalisation of the panel image typically bear traces
of personal or impersonal intelligence, but the attribution of these traces
to a continuum of a single individual consciousness is often impossible.
Understanding the workings of the characters’ and the narrator’s
minds is important in most graphic novels and nonfiction. But typically,
comics homogenise their visual compositional techniques and sensibil
ities to create a sense of a coherent narrative whole or to foreground
elements of a persistent graphic style. With so-called behaviourist nar
ratives, the reader can also always construct a sense of consciousness
from dialogue or mere descriptions of behaviour when there are no in
side views. What helps in this construction is that impersonal points of
view, though revealing things that the character could not see, are often
related to, or encompass, or are synchronised with, the character’s field
of vision. They show scenes as if from behind the character’s back or by
his or her side, present him or her looking at something, or include his or
her point of view within a broader field of vision in other ways.
The examples that I have discussed above all portray characters in ac
tion within social environments. The showing of the characters and their
actions from panel to panel creates a sense of a continuing tale, and estab
lishes what Palmer calls a characters’ ‘frame’.10 Keeping in mind the major
role of the authorial focalisation of the impersonal viewpoint (‘camera eye’)
in all comics storytelling, we may conclude that the sense of a continuing
subjective perspective is not a necessary semiotic channel for mediating
information in such narratives. As the ongoing debate over the notion of
a cinematic narrator has proved, it is not as simple to argue about visual
narratives—where the presence of the narrative agent is only implicit—as
it may be about heterodiegetic literary narratives, that the viewers neces
sarily attribute stylistic features to a hypothetical narrator persona.11 Be
haviourist and first-person comics, likewise, constantly exploit the various
possibilities of external, objective viewpoints. While developing what she
calls “visual narratology”, Mieke Bal has admitted that even if the distinc
tion between external and internal focalisers may hold in visual arts, this
distinction is not always easy to point out (1997, 163). In her comments
on the American artist Ken Aptekar’s paintings, Bal refers to a similar
and potentially ambiguous relation that I have discussed above, between
the “external” positioning of the figures who are shown in the image and
the direction of their gaze, hands, or other gestures that may call our at
tention to specific aspects within the space of the image. Other types of
challenging contrasts and fusions of internal and external perspectives in
comics storytelling include the function of the panel frame as a potentially
identifiable frame of vision. Within a sequence of images, the panel frame
can easily alternate between indicating either external or internal vision,
or narratorial or figural (character-bound) focalisation. Furthermore, the
images and words in a panel can point to events that take place outside the
frame and are, therefore, only indirectly linked with what is seen.
Study of the cognitive construction of consciousness, especially with
regard to behaviourist fictional narrative and contextual character port
rayal, can be helpful in understanding the presentation of minds in
comics. Recent narratological work on fictional minds is important in
pointing out the often unjustified privileging of ‘intrinsic’ character port
rayal in classical narratological approaches, based on modernist figural
literary narratives of inner speech, at the expense of reported speech and
thought or the showing of a mind in action. Nevertheless, third-person
graphic novels differ from third-person realist and modernist novels,
which are Palmer’s main focus in Fictional Minds and Social Minds
in the Novel (2010), in that both inner speech and reported thought
usually serve only limited, local purposes. At the same time, the mean
ing of a ‘behaviourist’, ‘neutral’, or ‘objective’ narrative may have to be
rethought in this medium-specific context where the reader’s cognitive
ability to read minds is tested not only in words and images but also in
placing the two components in interaction. Transitions between internal
and external visual focalisation are often essential in the understanding
of the story, while shifts between direct and indirect verbal modes of
narration are not as smooth as they may be in literary narratives. In
ner or silent speech can be visually marked, integrated within the space
of the image in thought balloons. Indirect speech, in contrast, such as
the reported thought of a third-person consciousness, remains doubly
indirect, difficult to place within the image and always in danger of re
peating the information given in the image. In the multimodal context
of comics storytelling, the directness of speech or thought can only be
adequately evaluated against the visual information of the image, just
as the personality or impersonality of the viewpoint in the image must
be assessed against the verbal content. I hope that it has become evi
dent that the study of mind construction in comics needs to go beyond
the speech-category approach, and beyond the categories of narrative
situation and focalisation that are based on literary examples, to con
sider speech acts in interaction with visual information and narration by
showing. Moreover, the study of mind construction in comics storytell
ing can benefit from examining the act of showing as a form of thinking,
and caricature (of face, gesture, pose) as a form of perception.
Notes
1 See, for instance, M. Smith (1995), G. Smith (2003), Bordwell (1985), and
Branigan (1992).
2 On the formal relationship between comics and film, see Boillat (2010),
Lefèvre (2011), and Kolp (1992).
3 See Cohn (1978, 49).
4 Saraceni, while helpfully discussing the presentation of speech and thought
in comics through linguistic categories, mistakenly claims that speech and
thought can only be reported directly in comics (2003, 62, 69).
5 Gerald Prince defines a behaviourist narrative as a “narrative characterized
by external focalisation and thus limited to the conveyance of the characters’
behavior (words and actions but not thought or feelings), their appearance,
and the setting against which they come to the fore” (2003, 10).
6 Bardet and Janvier’s version belongs to the Romans de Toujours series,
which is marketed with the intention of increasing interest in “treasures of
world literature” among young readers. To further this purpose, Bardet and
Janvier’s Madame Bovary also includes a CD version of the novel and a dos
sier that gives relevant information on the writer and the social, economical,
and political situations in France 1815–1848.
7 See, for instance, Phelan (2005, 118–119).
8 In this type of narrative mediacy, a reflecting character “thinks, feels and
perceives, but does not speak to the reader like a narrator. The reader
looks at the other characters of the narrative through the eyes of this
reflector-character. Since nobody ‘narrates’ in this case, the presentation
seems to be direct” (Stanzel 1984, 5).
9 These include, for instance, the moment when Emma considers leaping out
of the attic window after she has received Rodolphe’s letter of dismissal.
Here, the first two sentences, “Pourquoi n’en pas finir? Qui la retenait
donc?” preserve the sense of free indirect style, placing Emma’s thoughts
within the narrator’s discourse.
10 The reader’s processing of fictional minds based on bidirectional informa
tion flows, such as hearing about the characters for the first time or setting
up initial hypotheses about them (top-down information), and being fed new
information about the character in the text and thus modifying earlier infor
mation about them (bottom-up information) (2004, 176).
11 Fludernik argues that in terms of reader response to individual literary texts,
“the tendency to attribute stylistic features to a hypothetical narrator per
sona and/or a character is a simple fact” (2001, 622).
9 Dialogue in Comics
Medium-Specific Features
and Basic Narrative Functions
Conversation is a basic element in the medium of comics, where much of

the narrative appeal is derived from the interplay between dialogue and
action. The speech balloon, a favoured visual symbol for voice and utter
ance in the medium since the mid-twentieth century, has become a sym
bol for comics. In Italian, famously, the word fumetto—the word for a
speech or thought balloon—also refers to the art form itself, whether in
the form of a comic strip or a comic book. In fact, dialogue is such a cen
tral feature in the medium that it may sometimes be difficult to think of
it as a distinct element. A character who speaks his thoughts aloud when
apparently nobody is listening is a much-used convention, and many
comics, for instance, ‘talking heads’ or humoristic comic strips that de
liver a verbal gag, focus on speaking. Perhaps paradoxically, dialogue
scenes may be more distinguishable when their use is more restricted,
for instance, in comics when action is predominant and only occasion
ally interrupted by a scene of talk or when first-person verbal narration
is predominant, as in autobiographical comics that occasionally lapse
into dialogue.
The reason for the popularity of the dialogue form in comics is at
least partly related to medium-specific constraints and affordances that
encourage its use and, concomitantly, restrict the employment of more
indirect forms of speech and thought representation. In contrast with
dialogue, forms of indirect discourse, such as free indirect discourse or
the narratorial reporting of a character’s speech, tend to demand more
space for words. Conventional strategies for distinguishing between
these modes of verbal narration have included their visual form and
placement in relation to the images. The dichotomy between narrato
rial voice in caption boxes and dialogue or other forms of direct speech
in text balloons is not always clear-cut, let alone all-inclusive. Speech
in comics can also occur in captions, verbal narration can take place in
text balloons, the narrator’s and the character’s voices may intermingle,1
and neither verbal narration nor direct speech or thought must be placed
in boxes or balloons. Moreover, text in comics can occur outside these
two categories in the image background or as part of the image. How
ever, the continued assertion of the difference between direct speech and
Dialogue in Comics 221
other modes of verbal narration in comics also needs to be taken into
consideration as an important convention in the medium.
This chapter focusses on the dialogue form as a key narrative device
and technique, and it examines the main compositional principles and
narrative functions that characterize conversational scenes in comics.
The starting point in this investigation is the multimodal character of
speech and conversational exchange in comics. This requires us to focus
on the interaction between the utterance and the elements of the image.
Thus, on the one hand, I will discuss the ways in which dialogue, in the
form of written speech, interacts with what is shown in the image, such as
the interlocutors’ facial expressions, gestures, body language, and other
visual cues of mental states and participant involvement. Furthermore,
this necessitates an investigation of the visual possibilities and expres
sive functions of typography, the graphic style of writing, onomatopoeia
or imitatives, 2 visual symbols, and standalone non-letter marks in the
written rendering of conversation. On the other hand, I will discuss the
function of speech balloons as metaphors for an utterance—‘utterance’
meaning here a specific piece of dialogue—voice, and turn-taking, and
their narrative role in organising the time of the speech event and the
order of its reading. Utterances in comics are characterised by their dual
role as both instances of imagined speech in the world of the story and
written language to be read. As to their latter function, it must be taken
into account that readers of comics need to process the relations between
the various utterances both in a single panel, when it includes several
utterances, and between the panels in order to create a sense of a con
tinuous conversation. Finally, I will briefly discuss some strategic uses
of contrast and emphasis between visual and verbal narration in speech
representation in comics.
The ultimate goal of this chapter is to develop a medium-specific un
derstanding of the dialogue form in comics and outline the basic narra
tive functions of scenes of talk in comics. In this investigation, different
examples will be drawn from innovative uses of dialogue in this me
dium. The subject is admittedly very broad. Within the bounds of this
chapter, I can merely hope to highlight the main features of interest in
this crucial and often central form in comics.
The Embodied Speech Situation in Comics

Given the multimodal nature of the medium and the importance of visual
showing in comics, the question of dialogue in comics requires us to think
of the areas of interaction between the image content, such as the por
trayal of the participants in the conversational scene, the utterance that is
placed in the image, and the main formal aspects of the composition, such
as panel relations and the page layout. First, let us consider the ways in
which the participants in such scenes are visually shown to be engaged in
the speech situation by means of non-verbal communication. Such means
include, especially, facial expression, posture, hand gesture, eye contact
(or gaze), and the expressive distortion of the interlocutors’ bodies.
Existing research on the gesture-utterance connection in comics sug
gests that the use of gestures as signs of emotion largely follows real-life
models in everyday speech situations (Fein and Kasher 1996; Forceville
2005). Both in everyday real-life conversation and comics, body lan
guage and posture are elemental communicative resources. At the same
time, however, research has also suggested that in comics, since they
commonly simplify and exaggerate bodily forms through caricature, the
speaker’s and the recipient’s gestures often have a more prominent role
than in real life (see Forceville 2005, 85; Fein and Kasher 1996, 795).
In particular, facial expressions that are based on elemental features,
such as eyebrows, eyes, gaze, mouth, furrows, and wrinkles, or the head
position, are conventionally exploited as signs of emotion, thought, at
titude, and stance. Similarly, speech in comics, while it may seek to be
verisimilar and can provide the linguist with useful examples of spoken
language, can take on wilfully distorted forms, such as simplification or
exaggeration, that are different from uses of spoken language in real-life
speech situations.3 As dialogue in comics also necessarily has a written
form and often an ostentatiously graphic and handwritten quality, the
study of speech in comics needs to be sensitive to graphic features and
the visual effects of written language.
Rodolphe Töpffer, who many see as the inventor of modern comics,
claimed in his essay “Essai de physiognomonie” (1845) that a graphic
trace has unique expressive potential, especially in relation to the draw
ing of a human face. For Töpffer, all faces in drawings, however, naively
or poorly completed, even in the form of simple scribbling, possess a
fixed expression. He further surmised that the viewer can recognize such
expressions without education, knowledge of art, or any experience in
drawing a face.4 Similarly, one of the basic tenets in today’s psychologi
cal research in face recognition is that people identify faces from very
little information. In such identification, as in recognizing an emotional
expression, the eyes and eyebrows are among the most salient regions to
pay attention to, followed by the mouth and the nose (Sadrô et al. 2003;
Sinha et al. 2005). Töpffer saw, similarly, that in order for the drawing
of a face to be effective, one needs to focus only on a limited number
of key aspects, such as the eyes, eyebrows, nose, nostril(s), chin, fore
head, wrinkles or folds of skin, and the shape of the head.5 Töpffer also
thought that the relation between these facial features and the person’s
posture—the form of his or her upper body, gestures, and attitudes—
mattered, even though he did not see them as important as the internal
features of a face (the eyes, nose, and mouth).
As recent psychological and sociological conversation analysis has
shown, facial expressions can enhance or disambiguate the speaker’s
and the recipient’s stances towards what is being said in real-life speech
situations (Ruusuvuori and Peräkylä 2009). In comics, likewise, a basic
element of speech representation is the relation between verbal utter
ances, facial expressions, and other features of body language such as
eye contact, typically accompanied by the sense of perspective and field
of vision that are inscribed in the image. For instance, a way of speaking
and listening can be revealed by an exchange of looks in subsequent
gaze images, images portraying someone looking at something or some
one, or reaction images, that is, images showing someone’s reaction to
something that is said. A recipient’s look can, for instance, indicate pen
siveness, concentration, or confusion, affiliation with the topic or the
speaker, the sharing of an understanding, or the rejection of an idea, or
it can reveal what is important and salient in the conversation situation
as a whole.6
Notice, for instance, the significance of facial expression, gaze, body lan
guage, and hand gestures in this scene from Abel Lanzac and Christophe
Blain’s Quai D’Orsay. Chroniques diplomatiques II (Weapons of Mass
Diplomacy 2012), which depicts a meeting between the French Minister
of Foreign Affairs, Alexandre Taillard de Vorms (inspired by Dominique
de Villepin), his speech writer Arthur Vlaminck, and a representative of
the logistics department, Gilles Mande (Figure 9.1). In this scene, the
furious minister protests to Mande about not being able to have a bigger
aeroplane (Airbus) for himself and his advisors on a diplomatic visit to
Russia. The intensity of the minister’s gaze and his facial expression, em
phasised in the close-up image framed to show only his piercing eyes and
part of his gigantic nose, convey the persistence of his stance, as well as
his manipulative attitude towards the others. The minister pours forth
a tirade of complaints, evidently fuelled by a sense of self-importance,
about the tightness of space in the smaller Falcon aircraft that has been
offered to him and his staff. All this is accompanied by expressive and
manipulative hand gestures.
Besides facial expressions and gaze, hands, hand gestures, and arm
positions can also have a significant function in speech situations in
comics, communicating meaning themselves or specifying the words’
meaning. Two likely reasons for the significance of hands in comics are
that we can gesture meaningfully and simulate shapes and things much
more accurately with our hands than with other body parts, and that
they can relatively easily be drawn to demonstrate this.7 Hand gestures
may be used as forms of illustration, specifying a type of action, a spa
tial relation, or a physical shape of something, or as a form of empha
sis, while a hand can also point to an object, place, or the interlocutor.
Waving, pointing, and beckoning can have a conversational function,
for instance, as an expression of the participant’s emotion, attitude, and
personality, and also as a conversational signal. The salience of hand
gestures in the image, or facial expressions, for that matter, can be
Figure 9.1 Abel Lanzac and Christophe Blain Weapons of Mass Diplomacy
(2012/2014). Trans. Edward Gauvin © 2014 SelfMadeHero.
further emphasised by means of layout, perspective, foregrounding, or

visual means of emphasis.
The meaning of the participants’ positions in an interaction, and what
psychology calls (inter)personal space behaviour or proxemics—how
people use the personal space around them as they interact with others—
can be effectively portrayed in comics by showing how participants in a
scene of talk take their space or relate to each other and the surrounding
environment. Focus on a particular person in a close-up image or fram
ing the image close to a participant or his or her field of vision may also
suggest a (narratorial) sense of proximity to that participant.8 This is also
common in film narratives. A more medium-specific aspect of significant
body language in comics is the non-realistic manipulation and distortion
of body shapes through caricature, that is, the relative malleability of
the drawn body. We can observe this, for instance, in the above example
from Weapons of Mass Diplomacy where Alexandre Taillard de Vorms’s
shoulders and nose change their proportional size from panel to panel.
The speaker’s body is thus modified to reflect his speech, attitude, and
personality; the body has an expressive function in itself.
Conversational scenes in comics, as in film narration, have an advan
tage over dialogue scenes in literature in that they may show various
non-verbal communication cues, which co-occur with verbal commu
nication and can combine the effect of such cues. All visually observ
able aspects of non-verbal communication that may be integrated in a
face-to-face dialogue in real life can also be portrayed in comics: facial
expression, posture, gesture, eye contact, touch, adornment, physiologi
cal responses, position and spatial relations, personal space, locomotion,
and setting.9 While comics, at least in the traditional forms of printed
strips or books, cannot usually represent sounds, they have developed
various ways of suggesting auditory signals and vocal behaviour, such as
onomatopoeia, sound effects, and symbols. All these cues are potentially
relevant in conversational scenes in comics, where they co-occur with
the verbal utterance. As they interact with each other and the utterance,
these devices help the reader to create a sense of a continuing speech
event, or what is meant by what is said; better perceive the participant’s
mental state, attitude, and intention; and grasp the nature of the rela
tion between the speakers. Yet, the ways in which cartoonists may take
advantage of the rich possibilities of non-verbal communication in the
medium vary greatly. For instance, while facial expressions are generally
important, from children’s comic strips to adult-oriented graphic novels,
or from superhero comics to nonfiction reportage, some cartoonists also
simplify facial expression cues or minimise their use.10 Thus, the vary
ing aspects of non-verbal communication, and in some cases even facial
expressions, can be conceived of as optional tools of visual showing and
narration in conversational scenes.
Symbols of the Speaker’s Mental State and Engagement

In much comics storytelling, the use of visual symbols and verbal-visual
signs that emanate from the characters may also contribute significantly
to speech representation and dialogue scenes. In the passage of Weapons
of Mass Diplomacy above, Mande’s heavy sweating, shown with drops
of sweat, his changing facial skin colour, and later also his gradually
shrinking head and body clearly point out his submission to the minis
ter’s authority. The visual symbols around his head, which the cartoon
ist Mort Walker has called ‘emanata’ and John M. Kennedy identified as
‘pictorial runes’ (1982, 600),11 portray emotions (agony), mental states,
and an internal condition (submission). These and similar graphic de
vices, such as drops of sweat or more symbolic signs such as wiggly lines,
starbursts, circles, halos, and clouds, often have little or no relation to
the outer signs of emotion and attitude in real-life speech situations. As
conventions that are used in modern narrative drawings from cartoons
to comics, emanata are metonymically motivated signs that result from
a character’s emotion and thought or some immediate sensory stimuli
and effect. Typically, they specify the force of the speech act, a speaker’s
enthusiasm or uncertainty, the recipient’s understanding or lack of un
derstanding of what is said, acceptance and disappointment, or, as here,
gradual submission to the speaker. Emanata and altered body shapes
can also portray types of perception and reactions, including the sense
of cold and warmth, smell, newness, light, and brightness or percep
tions of speed, reflection, sudden or fast movement (speed lines), the
direction of movement, or surprise and suspense. Not all comics employ
them, but when they are used, they can contribute significantly to our
understanding of the other elements in a scene of talk such as facial ex
pression, gestures, and gaze.
Beyond the emanata, or pictorial runes, conversational scenes in
comics can also comprise various other signs, including stand-alone
punctuation marks,12 pictograms,13 sound effects, imitatives, and on
omatopoeia14 that have similar or related functions. Placed in the space
in the image around the characters, or possibly continuing from panel to
panel, these signs can equally specify the characters’ emotions, thoughts,
and attitudes or a way of acting, behaving, and speaking; clarify what
is said; or express movement, sounds, and other sensory stimuli that are
relevant in the scene.
Comics imitatives, which are widely used for humorous purposes,
approximate non-linguistic sounds and action or contact between the
characters, as well as attitude, emotion, sensations, and movement by
adapting them to the phonemic system of the language. Onomatopoeia
and sound words (or descriptive sound effects), which can be regarded
as a specific case of imitatives, represent sound and voice in verbal form
and, at the same time, often aim for a visual effect, which in itself can
mime some quality of the sound or reflect its source, such as an event
causing the sound. Onomatopoeia may also indicate variation in sound
effects such as volume, pitch, timbre, and duration. Typically, onomato
poeia fit the phonology of the language in which they are used (‘boom!’,
‘wham!’, and ‘whoosh!’ in English or ‘baoum!’, ‘pff!’, and ‘vlan!’ in
French). In comics storytelling, however, it is also common that an ono
matopoeic adaptation of a sound does not necessarily have to constitute
a word or even be pronounceable. Onomatopoeic expressions in comics
are not usually reducible to the sound that they imitate—one reason
being that they are given a visual, graphic form that contributes to their
meaning and effect. The use of stand-alone descriptive words (or de
scriptive imitatives) for sensations and emotions is also common (‘snort’,
‘gasp’, ‘tickle’, ‘sigh’, etc.).
Stylistic elements of writing, such as lettering, typography, and fonts,
as well as what has been called para- or quasi-balloonic phenomena,15
can be incorporated in a dialogue scene for similar purposes. The graphic
style in which speech is written is often meaningful in such scenes in two
senses. First, the graphic style of writing can create an effect of conti
nuity between the world of the story, or the speech situation, and the
written speech. For instance, written speech can be placed and shaped in
the image field so that it reflects the visual contents of the image.16 The
graphic line that depicts the speaking figures can also give the impression
of continuity in the writing (or vice versa). Second, the style of writing
can in itself express certain aspects of the utterance, such as emphasise
the meaning of a word, a phrase, or an utterance through bold lettering,
convey humour, add a metaphorical or ironic layer through a stylistic
change, imply a way of speaking or type of voice (whispering, singing, a
broadcast voice, and so on), the intensity of speaking (by changing the
letter size, for instance), and the speaker’s attitude or emotional state.
It can also portray differences between the speakers’ register, style, or
voice. Not all comics use the rich graphic potential of writing in this re
gard, but the style of writing and the choice of typography are important
features of conversational scenes in many comics. Think, for instance,
of Walt Kelly’s Pogo, or Neil Gaiman’s The Sandman, where typograph
ical choices may reflect the characters’ personality or attitude, or René
Goscinny and Albert Uderzo’s Asterix, where changes in lettering can
indicate important vocal and linguistic differences in the characters’
speech (accent, dialect, stylistic register, language). By these means, writ
ten dialogue in comics can overcome some of the limitations that affect
the representation of spoken language in conventional literary fiction.17
All in all, the various visual and verbal-visual signs that have become
conventionalised in comics can be metaphorically motivated as indexes
of a speaker’s emotions, thoughts, attitudes, and perceptions. All these
features may also contribute to the meaning of what is said, and poten
tially influence the reader’s attribution of mental states to characters.
Frequently, such signs work together to identify the speaker’s attitude,
complementing the meanings of facial expression and body language,
and thus specify or enhance the speaker’s relation to the propositional
content of the utterance and the other participants in the scene.18
Let us take as an example the main components of a scene of talk in
Finnish cartoonist Aapo Rapi’s (auto)biographical narrative Meti (2008).
This story is based on the cartoonist’s interviews with his 80-year-old
grandmother Meeri Rapi, known as Meti, but it also has a strong auto
biographical dimension: the cartoonist pictures himself in the story,
meeting and conversing with his grandmother, taking notes during the
conversation, and relates some other events in his life at the time of the
interviews and the storytelling. The narrative perspective of Meti is of
ten ambiguous in that there are clues in the story that let the reader think
that it is told and illustrated in the way that Aapo imagines the events
have happened—Meti’s story would thus be within the frame of Aapo’s
imagination—but there are also passages in the narrative where Aapo’s
story and Meti’s memories appear to be in competition. At times, the
frame narrative and Meti’s narrative also coalesce, resulting in a kind of
intersection of stories.
Here, in this scene of five panels, the speaker’s and the recipient’s facial
expressions, posture, gaze, exchange of looks, perspective, and emanata
play a vital role together (Figure 9.2). We first see the cartoonist meet
ing with his grandmother. When Meti attempts to formally introduce
Figure 9.2 Aapo Rapi. Meti (2008) © Aapo Rapi.
herself with ‘My name is M–’, the cartoonist, visibly frustrated by this
introduction—indicated by drops of sweat springing from his face, ac
companied by a few drops of coffee spilled from his cup—interrupts her
and insists that she should speak as she ‘normally’ does, that is, not in
formal discourse. Consider also the importance of gazes and perspective
in this scene. Both speakers are present in all images, but seen from dif
ferent angles and distances. The alternating perspective of the images al
lows us to see the scene from behind both characters’ shoulders and thus
share their viewpoints to some extent. Notice also that the cartoonist’s
face is much more expressive of emotion and mental state—changing
from signs of haste and frustration to calm—than that of the stony-
faced main character. Moreover, Meti’s large non-reflective glasses are
in stark contrast with the youthful expressiveness of her face in the nar
rated memories that follow this scene.
The Bond between the Speaker and the Utterance

Speech and thought balloons were successfully incorporated into
A merican newspaper comic strips in the 1890s. In earlier European
comics and cartoons, the same device had already been widely used,
including British satirical broadsheet prints (1770–1820), but in
Töpffer’s and in many other mid-nineteenth-century European car
toonists’ works, speech was usually represented in captions that were
placed underneath the images. Only by the 1940s and the early 1950s
did the representation of speech in speech balloons become a dominant
convention in the medium in most Western countries.19 Since then,
other options for representing direct speech, such as speech quoted or
summarised in captions, have remained in relatively limited use. Many
contemporary cartoonists, however, represent utterances without re
sorting to speech balloons. For instance, in Brecht Evens’s graphic nov
els and in much of Claire Bretécher’s work, the utterances are simply
placed physically close to the speaker in the space of the image, possi
bly but not necessarily accompanied by a tail that connects the utter
ance to the vocalizing source.
Regardless of whether comics use the speech balloon format or not,
the general principle that an utterance is tied to a source that is shown in
the image or to a source that is situated close to what is shown appears to
be a default expectation in comics. The tail emanating from the balloon,
or in some cases from the text without a frame, makes this association
even more evident as it directly points to the source of the utterance.
If the speaker is not shown in the image field, the default expectation
is that the balloon and the tail indicate that someone is just outside the
visible space or is not yet or no longer in the field of vision, or that the
source of the utterance is too small or hidden to be seen (see also Force
ville et al. 2010, 69).
Thus, the speech balloon and its tail, which can take a variety of dif
ferent visual forms, express the contents of the utterance and, at the
same time, are visual symbols of a speech act. In the latter function, we
need to underscore their metaphorical function, which has something in
common with metonymy: the balloon and its tail stand for a speaking
voice (or a sound), the place, time, and duration of speaking, and the act
of speaking itself. The relationship between the balloon and the speech
act can thus be conceptualised as a structure of contiguity where, with
the written utterance representing spoken language, the visual form of
the speech balloon stands in a metaphorical relation to the source of the
voice and, possibly also, to particular aspects of that voice or sound
(intonation, for instance). In contrast, thought balloons represent the
speaker’s thoughts and inner state. The distinction between speech and
thought balloons is not always unambiguous in comics, or their differ
ence may be irrelevant—does it always matter, for instance, whether
a person speaks or thinks aloud to himself?—but in general they are
distinguished by various visual markers such as the shape of the balloon
and the tail or the background colour.
Being a visual metaphor (or metonymy) for a speech act, the balloon
and its tail also perform the function of speech tags. In fact, they can
realise the speech tag function much more efficiently and economically
than any verbs of saying that traditionally introduce an utterance in lit
erary narratives. The function of the tail, specifically, is to identify the
speaker in the image. 20 The balloon and its tail not only point out the
turn-taking, the source of the utterance, and the place of the speaker, but
often also tell us how someone is speaking—the intonation, intensity,
and volume of speech may be reflected in the shape, size, place, or colour
of the balloon and its tail—or reveal the speaker’s attitude towards what
is being said (linguistic modality). Balloon frame styles, background
colour, and tail shapes regularly depict emotional states and sensory
experiences (uncertainty, (dis)approval, ‘warm’, ‘icy’), a type of voice
(electronically relayed, distant, shrill, high, low, harsh, broken, and so
on), or volume (loud, quiet, shout, whisper). Lettering, typography, and
visual signs inside the balloon can have similar functions or can amplify
them. If in the Asterix albums typography can be a sign of a different
language and dialect; in Brecht Evens’ graphic narratives, the colour of
the text identifies the speaker (The Wrong Place, 2009; The Making Of,
2011; Panthère, 2014).
The expressive uses of the speech balloon are well known to com
ics readers and scholars, but perhaps less to academics who study the
dialogue form across media. Charles Forceville has shown how differ
ent visual variables of comics balloons—contour form, colour, fonts,
non-verbal contents, and tail use—contribute narratively salient infor
mation, for instance, with regard to the manner and topic of speaking or
the identity of the speaker (2013, 258, 268). In other words, the visual
variables of the balloon, especially in more nonstandard cases, make
salient something in what is said, how something is said, or who the
speaker is. This, again, requires that we evaluate the relation of the bal
loonic narrative information to the speaker and the speech situation as
a whole. The place of the balloon in the scene or the breakdown may
also be significant. Thierry Groensteen, who has made a theoretically
grounded description of speech balloon functions in comics, has sug
gested that the place of the balloon is always relative to three different
elements in the space of the page: the character who is speaking (the
speaker), the frame of the panel, and the neighbouring balloons (situated
in the same panel or a contiguous one) (2007, 75). Groensteen empha
sises, in particular, the interdependence between the characters and the
balloons (2007, 75, 83), claiming that their relationship is so strong that
they form a sort of functional binomial, a bipolar structure that is a
necessary organising device in comics. Moreover, Groensteen presumes
that the characters in the panels are the most salient piece of informa
tion and, subsequently, echoing Töpffer, that the character’s face and
physiognomic expression are the principal focal points of the reader’s
attention (2007, 75–76). In reading comics, then, the reader would sup
posedly first view the character’s face and expression, and then adjust
this information, reciprocally, with what is said, that is, the character’s
represented speech. 21
The claim about the bipolar structure between an utterance and an ut
terer seems highly relevant with regard to most comics. The psychologi
cal study of face recognition has also proven that the human (biological)
visual system starts with a rudimentary preference for face-like patterns,
and that our visual system has unique cognitive and neural mechanisms
for face processing (see Sinha et al. 2005). Yet, it seems worth asking
whether the functional binomial between the speaker and the utterance
is always dominant in guiding the cartoonist’s or the reader’s under
standing of conversational scenes, or the order of their reading. For one
thing, we still cannot say much that is not controversial about the read
er’s order of attention in reading comics. Do we always start reading
comics by viewing the characters’ faces?22 Comics can vary greatly with
regard to the relative amount of words they use, as well as for what
purpose they use them (what kind of information is given verbally), let
alone that the image-word ratio typically alternates within any given
story. A dialogue scene can portray the participants’ positions, gestures,
and relations in great detail, but in a ‘talking heads’ story or a verbal
gag strip, words can also be the primary focus of the reader’s attention,
whereas sometimes faces can tell next to nothing.
In addition, comics can successfully sever the relation between the
speaker, words, and space of the speech situation by various means. This
may be done, for instance, by excluding the speaker from the space of
the image or the narrative level, by multiplying the number of speakers
or utterances, and by making the connection between an utterance and
a speaker ambivalent in the space of the image. 23 The relation between
the utterance and the vocalising source may remain deliberately ambiva
lent, for instance, in panels where there is only speech and the characters
are not seen, or not clearly seen, such as in panoramic images where the
speaking figures may be shown far in the distance or are not visible at
all, or in images where the vocalising agent is visually blocked. François
Ayroles’s strip “Feinte Trinité”, which includes only speech balloons
and no figures, pertaining to a conversation between a son, a father, a
mother, and God, or the online comic strip Bande pas dessinée, chal
lenges the basic bipolar structure further by never letting us see who is
speaking.
Another challenge to the bipolar structure arises from the speaker’s
ambivalent positioning between the picture space and outside it. In some
rare cases, the speaker can also remain systematically absent from the
images. Consider, for instance, the continuous commentator track in
Altan’s Ada (1979) where a speaker, who is never seen, is emotionally
involved in the narrative as its commentator and viewer. Much more
common is that a voice may, once connected with a particular speaker,
become disconnected from that speaker on the visual level of narra
tion. This may occur, for instance, when utterances are superimposed
on what is seen in the images, thus suggesting that what is seen is the
character’s subjective vision. Towards the end of the frame narrative of
World’s End, the Chaucerian story arc in The Sandman series, the voices
of a group of characters at an inn called World’s End are superimposed
in speech balloons on a double spread with images of an enlarging win
dow pane through which they apparently look at a spectral funeral pro
cession in the sky. The reader, thus, is invited to share their field of vision
through the dialogue.
Still other challenges to the rule of the bipolar structure of speech in
comics include the multiplication of speakers for one utterance and the
use of one speaker as a representative of a group of speakers. For instance,
Martin Cendreda’s one-page story, “I want you to like me”, experi
ments with this principle by letting a conversation continue from panel
to panel while the speakers and their spaces keep changing (Chapter 3).
This creates the effect of a communal mind that apparently thinks the
same thought, and says the same thing, irrespective of the individual
sources of utterance seen in the images (speakers, billboard, dogs).
Similarly, ideas apparently voiced by one person can be attributed to a
group of people. 24 A character’s voice may also occur in many parts of
one panel. This can emphasise, for instance, the speaker’s quick move
ment, the effects of an echo, or the complexities of space, as happens in
A sterix and the Banquet when the Gaul Jellibabix, who is not seen in the
panel, says ‘Here!’ in six different corners of the maze-like alleyways of
Lugdunum (modern-day Lyon) seemingly at the same time.
All these cases experiment with the basic expectations of speech repre
sentation in comics: an utterance is visually tied to a particular speaker,
and both the utterance and the speaker belong to the space that is seen
in the panel. Yet as the exceptions above show, the bipolar structure
between the speaker and the utterance can always be modified, chal
lenged, and even discarded. The exceptions make the rule more visible,
but the flexibility of the structure also points out that, to better under
stand speech representation and dialogue in comics, it is crucial to think
beyond the speaker-utterance relation to a number of other seminal ele
ments of dialogue in the medium.
Still another important feature of conversational scenes in comics is
the interaction between the utterance, the contents of the image and
narrative captions. Narrative captions, which are typically distinguished
from speech balloons by their frames, background colour, or typography,
can also complement, evaluate, or interpret the speech acts presented in
the images. In Daniel Clowes’s first-person narrative Mister Wonder-
ful (2011), the contrasted and sometimes competing thought captions
and speech balloons of the story make clearly visible the expected inter
relations between the captions and the balloons. Here, the narrator’s
thoughts, placed in square-shaped captions with a yellow background,
are frequently superimposed on speech balloons that contain the narra
tor’s own speech or other people’s utterances, thus indicating, among
other things, the narrator’s lack of attention to what is being said. On
a few occasions, the speech balloons are also superimposed on the cap
tions, thus suggesting that what is said interrupts the flow and momen
tum of the narrator’s thoughts. Thus, also, the connection between the
speaker and the utterance in the balloon is momentarily broken.
The Temporal and Rhythmic Functions

of Speech Balloons
Having investigated some basic formal elements of speech representation
and scenes of talk in comics, we should be able to focus more specifically
on how some of these elements realise narrative functions in comics.
Character-to-character dialogue, or combined action and dialogue
scenes, are central forms of narrative organisation and development in
comics, as in literary fiction and film. 25 Dialogue scenes move the story
forward, for instance, by giving important information about the char
acters, their relationships, the milieu, and the evolving events; they can
also build suspense and reorientate the narrative. In comics, dialogue
also regularly accompanies action. In Asterix, much of the talking bet
ween Asterix and Obelix, which is a constant feature of the series, takes
place when the two characters are on the move or doing something.
Action and dialogue are constantly bound together: while moving or
acting out a scene, the characters discuss their intentions, thoughts, and
emotions or voice comments about an event or someone they have met.
What Sarah Kozloff has outlined as the main narrative functions of
dialogue in film largely apply to comics. Dialogue in films, as Kozloff
points out, can contribute to many if not all key elements of a narrative:
world construction and identification, characterisation, communication
of narrative causality (such as the relation between events or the sig
nificance of an event), enactment of a narrative event (the disclosure of
important information such as the speaker’s emotional state), adherence
to realism (plausibility), and control of the viewer’s evaluation and emo
tions (the sense of narrative rhythm, the effects of surprise and suspense)
(2000, 33–51). Inevitably, a given instance of dialogue can fulfil several
of these functions simultaneously.
What is different in comics in this respect may to some extent be
self-evident. Comics lack the sound element, the means and possibilities
of the moving image, and the actor’s work and personality is not an
issue. With regard to narrative pacing and rhythm in comics, speech
balloons play a vital role. Their arrangement in the panel, a sequence, or
on the page, modifies both the sense of the time of the narrative and the
order and time of reading. On the one hand, the utterances punctuate
the story and the dialogue scene and, thus, create a sense of the duration
of the event. Sometimes, the speech balloons can in themselves express
duration through elongated forms of tails that surpass the frame bor
ders. On the other hand, the speech balloons are part and parcel of the
spatial organisation of the comic’s page. While the speech balloon con
stitutes a space where the utterance can be read, the placement and inter
relation of the speech balloons in the space of the page also point out to
the reader an order of looking and reading, functioning as one means of
connectivity between the panels. From the reader’s perspective, thus, the
utterances in a given narrative comic mark stages in the story that need
to be attended to.26 Speech balloons placed on the picture frames, for
instance, or close to each other in neighbouring panels, can strengthen
the link between the pictures and thus affirm the order of reading.
Sometimes also, the space of the utterance can approximate the function
of a picture frame or the space between the panels. The placement of
speech balloons in a scene of talk may also emphasise, together with
other features of the scene, particular aspects of the utterance and the
speech situation.
The opening scene of Book One in Garth Ennis and Steve Dillon’s
Preacher (1996) features a conversational scene between three char
acters, Jesse Custer, Tulip O’Hare, and Cassidy, who are conversing
at a table in a diner in Texas. In this example, I would like to empha
sise the significance of three factors in the depiction of the scene: the
place of the utterance, the effect of the moving perspective, and the
means of layout. The first time we see the protagonist Reverend Jesse
Custer’s face and his clerical collar, his utterance—‘’cause lemme tell
you: it sure as hell ain’t the church’—is placed over the frame border.
Both the placement of the utterance, the particularity of which is em
phasised by the fact that speech balloons very rarely cross the panel
frames in this series, and the contrast between what Custer says and
who he is stress the importance of the utterance (Figure 9.3). Further
noteworthy elements in this panel are the angle of vision, which is
placed squarely amidst the interlocutors and very close to Cassidy’s
position in the scene, and the fact that two sides of the panel bleed
off the corner of the page. The latter feature may compel the reader
to turn the page to learn more about the contrast between the speaker
and what he has said. In the following pages that depict the conver
sation, the perspective remains close to the characters, stressing the
meaning of gazes and the exchange of looks. Moreover, and typically
of dialogue scenes in many contemporary graphic novels, the perspec
tive keeps steadily shifting around the conversing characters, moving
to one more or less subjective angle of vision in each panel. Finally,
page layout also contributes to this scene through the partial super
imposition of some of the panels, such as a close-up image of Tulip
O’Hare, on the surrounding panels, thus further aligning the inter
locutors to each other and emphasising the importance of a particular
gaze, expression, and utterance.
Concerning the sense of rhythm in such scenes, one default expec
tation is the correspondence between the utterance or an exchange of
dialogue and the speaker’s (or listener’s) posture shown in the image.
We could call this the realistic formula of time in a scene of talk. In
other words, perhaps the most basic rhythm of speech representation in
comics is one utterance per speaker, or one utterance and response per
panel. Will Eisner, for instance, has stressed the importance of preserv
ing such a bond between dialogue and action on the grounds of realism,
claiming that a protracted exchange of dialogue cannot be realistically
supported by unmoving static images. Furthermore, for Eisner, a veri
similar exchange of dialogue is one in which the utterances terminate
the endurance of the image, that is, the dialogue corresponds with the
speaker’s (or speakers’) posture in the image (1996, 60).
Figure 9.3 Garth Ennis and Steve Dillon. PREACHER. Book One (1995)
© Garth Ennis and Steve Dillon. All characters, the distinctive like
nesses thereof, and all related elements are trademarks of Garth
Ennis and Steve Dillon.
However, Eisner’s presumption, while it illustrates a basic conven

tion for representing duration in conversational scenes in much com
ics storytelling, can be contested as an all-encompassing general rule of
realistic speech. Clearly, instead of undermining the sense of veracity
in a conversational exchange, a long string or multitude of balloons in
one panel can also enhance realism in narration. On the first page of
Preacher, Cassidy’s and Tulip O’Hare’s utterances have two parts—their
difference is marked, respectively, by the conventions of one balloon
opening onto another and by a connecting tail between the balloons.
This is a common way to indicate a short pause in speech. Elsewhere,
the placement of many speech balloons in one panel can create the effect
of a speeded-up and intensified exchange of words. Strings of balloons
or a mass of balloons in one panel may, for instance, suggest the effect
of an improvised discourse, conversational intensity (as in the streets of
Lutetia in Les lauriers de César), interruption and talking over others,
the volume of speech, a cacophony of voices, and so on. Many super
imposed balloons can also indicate a disconnection between speech
and thought, as happens in Mister Wonderful, where the narrative cap
tions that are placed on the speech balloons and sometimes even on the
speakers’ faces emphasise the effect of an inner voice overriding speech.
Moreover, a protracted exchange of dialogue in one panel may suggest a
notable speeding or slowing of time in a scene of talk, instead of under
mining conversational veracity.
Furthermore, it is important to note that the relation between speech
and posture does not alone create the sense of rhythm in dialogue scenes.
The panel-to-panel transitions and other spatial relations on the page, in
cluding the sense of time in a single panel, also affect our understanding
of the time and duration of a scene of talk. In our previous example of
Meti, the sequence suggests a slowing of time during the dialogue scene:
the cartoonist’s hurry to start the interview—he is visibly out of breath
when he enters the room in the first panel—is contrasted with Meti’s re
laxed attitude. Meti’s calmness has become evident to the reader already
in the previous wordless pages of the story, which portray her leisurely
picking berries, preparing a pie, and baking it in the kitchen. The fourth
and the only wordless panel in this sequence, in which the perspective is
more distant and impersonal, powerfully suggests the passing and slow
ing of time. In these five panels, the cartoonist figure thus apparently
adjusts to Meti’s sense of time by eating lingonberry pie and drinking
coffee. Only then can the actual storytelling start.
Conversational scenes, when perceived as distinct scenes, may alter
the temporal rhythm in relation to the surrounding narrative action.
This dimension of dialogue scenes in comics corresponds with what
Kozloff refers to as the control of viewer evaluation and emotional re
sponse through dialogue. In comics, as in film, such scenes can distract,
create suspense and surprise, or control emotional response by elongat
ing a moment and stretching out a suspenseful climax or pause. The
conversation at the beginning of Preacher, which turns out to be a frame
narrative for much of the ensuing story in Book One of the series, intro
duces us to the main characters and opens up several questions about
their situation that will be dealt with in the subsequent instalments of
the story. Scenes of talk can also slow down the tempo in the narra
tive, as in the example from Meti above, to the extent that they give us
an impression of simultaneity between the time of the events and the
time of their telling and showing. In comics that include extensive dia
logue during the action, such as Asterix or other European adventure
Figure 9.4 Jérôme Mulot & Florent Ruppert. Barrel of Monkeys © 2008,
Ruppert, Mulot & L’Association, Rebus Books for the english
translation.
series, such as Spirou and Fantasio, such temporal changes may not be

apparent, however, since the action and dialogue establish such a steady
rhythm throughout the narrative.
Jérôme Mulot and Florent Ruppert’s comic books, including Safari
monseigneur (2005), Panier de singe (Barrel of Monkeys, 2006), and
Le Tricheur (2008), make visible a number of underlying principles in
speech representation in comics. For instance, they extend the tradi
tional realistic duration of speech in a panel: Ruppert and Mulot some
times place up to twenty balloons per panel for one speaker and thus
obfuscate the expectation of synchrony between the speaker’s posture
in the image and the utterance (Figure 9.4). Furthermore, their work
investigates the rules of readable information, that is, that speech bal
loons should contain informative utterances that are attributed to some
agent in the story. Generally speaking, certain constraints guarantee the
readability of speech balloons in comics. This means that one is to avoid
(a) superimposed speech balloons that block the reading of other bal
loons, unless the superimposed balloons serve a clear narrative function
such as indicating the simultaneity of many voices; (b) balloons placed in
a semantically important part of the image (such as the speaker’s face);
(c) balloons that are ‘cut’ by the image frame so that they become un
readable (this may also happen in Mister Wonderful to point out the
narrator’s lack of attention or interest); and (d) continuous nonsensical
expressions or empty balloons. However, single ‘blah-blahs’ or empty
balloons can be very revelatory of attitude or a lack of response.
Still other experiments with speech and thought balloons in Ruppert
and Mulot’s comic books involve the breaking of the flat symbolic space
of the speech and thought balloon. For instance, letters and signs regu
larly overlap the balloon contours and extend to the space of the image
in their works, thus undermining the expectation that the balloon is an
enclosed space in itself, or speech and thought balloons are treated as lit
eral containers that convey the illusion of three-dimensionality. Some of
Ruppert and Mulot’s speech and thought balloons, or their contents, can
be seen, touched, and entered, whereas others may indicate the speaker’s
movement in space as a kind of visual trace of the movement.
The Narrative Function of Visual and Verbal Contrast

in Dialogue Scenes
Still another medium-specific aspect in conversational scenes in comics
is the narrative effect (rather than function) of contrast, or narratively
motivated transition, in the balance between visual and verbal narra
tion. For instance, a scene of character-to-character dialogue in comics
can always turn into a predominantly visual narrative that fleshes out
the topic of the conversation in narrative drawings, or vice versa. This
is a typical element in Aapo Rapi’s Meti and complicates in this story
the question of the identity of the narrative agent responsible for what
is shown in the images. Lilli Carré’s The Lagoon (2008), in turn, de
picts a scene where someone is telling a tale, and the oral story is then
transformed into a visual narrative that the reader can see evolving from
panel to panel. The shift from verbal to graphic narration thus dra
matises the temporal distance between the present of the storytelling
and the past of the story events, but it also has the narrative effect of
accentuating the storyteller’s skill of inviting the listener into her world
and experiencing it from within. Such transferences between verbal and
visual narration may in some cases be compared to shifts between dif
ferent diegetic levels in a literary narrative, for instance when an inter
locutor in a conversation becomes a narrator of his or her own story.
Yet, the multimodal nature of comics allows the invention of forms of
complexity in this regard, pertaining to the relation between the time
of the events and the time of their telling, or the source and perspective
of narration, that are not available in the monomodal context of literary
narratives.
Cartoonists can set up tensions between verbal and visual narration
in conversational scenes for various other effects as well. Another device
for contrasting verbal and visual narration is to juxtapose the time and
place of an ongoing conversation and the time and place of the events
that are the topic of the conversation. For instance, at the beginning
of Jean-Claude Mézières and Pierre Christin’s Brooklyn Station Termi-
nus Cosmos (1981), where the main characters Valerian and Laureline
are engaged in a long telepathic intergalactic conversation, their dia
logue provides the story with a narrative frame. This global frame em
beds images from the speakers’ memories as short flashbacks as well
as illustrations of things and events that the speakers have heard. The
extended present moment of the dialogue thus creates a kind of intersub
jective consciousness frame that incorporates different temporalities and
changes of space, which are shown in the narrative drawings. The dia
logue may specify that the things seen in the panels have a varying rela
tion to reality—first- or second-hand information, mnemonic images, or
things seen in the speakers’ present whereabouts—or different meanings
for each speaker. The overall effect, however, is not one of simple fram
ing and embedding, but the time and space in which speakers are situ
ated occasionally also appear to coalesce with those of their stories and
memories as the speakers share the imagery through the telepathic link.
Conclusion
The ultimate goal of this chapter has been an attempt to develop a more
general understanding of the basic elements, main compositional princi
ples, and narrative functions of speech and dialogue in comics. One cru
cial area for future research that is indicated by this discussion is the way
in which the image content, especially the embodiment of the participants,
contributes to the conversational scene and the interpretative effects that
the scene generates. Typically, the images in comics show involvement in
scenes of talk through shared or contrasted perspectives, an exchange of
looks, or through gesture, posture, and other physical signs of reaction
to others. A key aspect of dialogue in comics in this respect is the depic
tion of the participants’ face and facial expressions. Visual symbols and
verbal-visual signs, such as emanata, which are added to or around the
participants’ face and head in some comics, can specify an expression,
show mental states, and emphasise a reaction to someone or something
that is said. Furthermore, comics may manipulate the characters’ body
shape and size to underline certain aspects of a speaker’s experience, atti
tude, or personality, or their reaction and engagement in the speech situa
tion. Together and in interaction with the verbal content of the dialogue,
these elements produce an integrated, but often quite complex, whole.
Finally, all compositional and spatial elements in comics can have an
expressive function that contributes to the reader’s understanding of
conversational scenes in this medium. Changing picture frames, panel
forms, panel and balloon shapes and sizes, page setup, lettering and
letter size, non-realistic backgrounds, 27 and other components of graphic
style can convey relevant information, for instance, by emphasising or
modifying the meaning of the utterances or pointing out the salient fea
tures in the situation. Moreover, the relations between the panels may
imply relevant narrative information about the scene; the gaps in what
is visually shown in the panel images need to be related to what is said
but also to the gaps in the dialogue. The precise meaning of the poten
tially meaningful formal elements in a scene of talk depends again on the
co-occurrence and combination of these elements and on their tension
and interaction with what is said and shown in the images.
Comics share various functions of narrative communication through di
alogue with other narrative media, but also employ many medium-specific
strategies that render impossible any direct comparison with dialogue
scenes in literature or film. Speech in comics is not only given in a written
form but also (usually) in a drawn form, a kind of graphic writing. In
this respect, comics vary greatly in the extent that they can maximize the
graphic and typographical effects of written speech. The speech balloons
function as a visual metaphor for a speech act, voice, and source. At the
same time, the speech balloon, the tail, and para-balloonic utterances
contribute to the organisation of the time of the narrative and the order
and time of reading. Above I have also investigated the common con
vention in comics that an utterance is physically tied to its source, the
speaker, and that this relation suggests a certain (imaginary) duration
of time. By developing Thierry Groensteen’s (1999, 2007) insights about
the elemental association between the speaker and the utterance, I have
sought to contextualise this compositional principle in relation to other
key elements of conversational scenes in comics.
Notes
1 See also Saraceni (2003, 66–67) on how this may happen in thought bal
loons and monologue.
2 Oswalt defines an ‘imitative’ as “a word based on an approximation of some
non-linguistic sound but adapted to the phonemic system of the language”
(1994, 293).
3 See also Frank Bramlett, who stresses that a linguistic investigation of lan
guage in comics needs to consider the balance of realism in the characters’
language and the amount of linguistic exaggeration and simplification that is
typical of the medium (2012, 183). See also Hatfield (2005, 34), Groensteen
(2007, 129), and Miodrag (2013, 32–36).
4 The art historian Ernst Gombrich famously named this rule Töpffer’s
law: “For any drawing of a human face, however inept, however childish,
possesses, by the very fact that it has been drawn, a character and an expres
sion” (Gombrich 1960, 339–340).
5 Bremond points out how the ‘teratological’ anatomies of certain characters
in comics allow us to pose the question of which bodily organs are abso
lutely indispensable for the realisation of gestural messages (1968, 99).
6 One type of gazing that may be equally well-portrayed in dialogue scenes
is the characters’ joint visual attention to something. For a reference in film
studies, see, for instance, Persson (2003, 68–91).
7 See Baetens (2004) on the depiction of hands in Yves Chaland’s and Jacques
Tardi’s works.
8 See, for instance, Persson on visual media and personal space (2003,
109–110).
9 Speakers in real-life speech situations can co-opt almost any physical ac
tion conversationally, that is, demonstrate by timing an action with the ver
bal communication that the non-verbal act has a communicative function
(Bavelas and Chovil 2006, 100).
10 E.S. Tan argues that some graphic novels avoid using the schema of facial
expressions altogether, “either because it is too explicit, or because the emo
tions that characters have are too complex to be ‘told’ through the face”
(2001, 45). I would argue that narration “through the face” is a matter of
stylistic choice rather than a reflection of the story’s simplicity.
11 Kennedy distinguishes actual pictorial runes that are metaphorical, such as
the state of anxiety shown by eye spirals, from graphic lines that have some
literal intent as they attempt to convey perceptual impressions, such as lines
radiating from bright light (1982, 600). Forceville has adopted Kennedy’s
term (2005, 2011). In his tongue-in-cheek lexicon, Mort Walker defines em
anata as emanating outwards “from things as well as people to show what’s
going on”, such as a character’s “internal conditions” (2000).
12 See also Dürrenmatt’s (2013, 115–127) discussion of how exclamation
points, question marks, and ellipses have become autonomous means of
description in the medium, especially for expressing characters’ emotions,
mental states, and/or silence.
13 Forceville, El Rafaie, and Meesters distinguish a pictogram from a pictorial
rune on the basis that an isolated pictogram, such as $ or ♥, has “some basic
meaning of its own when encountered outside of comics”, unlike a picto
rial rune such as motion lines, droplets, spikes, or spirals (2014, 492–493).
They admit, however, that the borderline between the two categories may be
fuzzy (2014, 494).
14 Suzanne Covey distinguishes between ‘descriptive’ sound effects, by which
she means “words, usually verbs, that don’t attempt to reproduce the sounds
they depict” and onomatopoeic words that try to approximate sounds at
least to some degree (2006).
15 Forceville, Veale, and Feyaerts include in para- and quasi-balloonic phenom
ena the various non-bordered zones of the picture that display onomato
poeia and sound effects (2010, 65). On onomatopoeia in French-language
comics, see Fresnault-Deruelle (1977, 185–199).
16 Some examples are discussed, for instance, in Dürrenmatt (2013, 165–167).
17 Compare with Chapman (1984, 18–24) on the difficulties of reproducing
speech in written dialogue.
18 Forceville emphasises, importantly, the combined effect of non-verbal signs
in comics in the representation of emotions such as anger (2005, 84–85).
19 See Smolderen (2002, 2009, 119–127) on why the speech balloon was rarely
utilised as a citation of a character’s speech before Richard F. Outcault’s
“The Yellow Kid”. There are important exceptions, however (see the last
chapter of this book). Lefèvre discusses the gradual spread of the balloon
device in European comics since its final breakthrough in the 1930s (2006).
20 Saraceni argues succinctly that the “function of the tail is equivalent to that
of clauses like ‘he said’ or ‘Ann thought’ in reported speech or thought”
(2003, 9).
21 Lawrence Abbott’s educated guess about eye movements and the order of
reading comics is similar to Groensteen’s suggestions, but Abbott puts the
main stress on words and verbal narration (1986, 159–162).
22 Will Eisner’s caution in this matter seems justified, even if eye-tracking re
search has made important advances recently: “In comics, no one really
knows for certain whether the words are read before or after viewing the
picture. We have no real evidence that they are read simultaneously. There
is a different cognitive process between reading words and pictures. But in
any event, the image and the dialogue give meaning to each other—a vital
element in graphic storytelling” (1996, 59).
23 See also Forceville (2013), who discusses some effects of tailless balloons
and tails that do not point toward an identified or identifiable speaker.
24 Carrier (2000, 42–43) associates this effect with a page from Joe Sacco’s
Palestine, but does not explicate how the effect is created. See also Force
ville (2013, 265–266) on a panel in Régis Franc’s Nouvelles Histoires: Un
dimanche d’été, where a substantial number of tails do not point toward any
identifiable speaker, thus creating the effect of a palaver where “it does not
matter very much who is saying what”.
25 See also Phelan and Rabinowitz (2012, 37–38).
26 In Groensteen’s formulation, the positioning of the balloons in the space of
the page creates a rhythm in reading as “each text fragment retains some
moment of our attention, introducing a brief pause in the movement that
sweeps across the page” (2007, 83).
27 On how pictorial metaphors in the image background may express a person’s
emotional state in manga, see Shinohara and Matsunaka (2009, 283–290).
Part V
Narrative Form and

Publication Format
10 Picture Story and Narrative
Organisation in Early
Nineteenth-Century British
Caricature and Comic Strips
Just as it is the task of the literary historian to evaluate in what way

and how far the term “novel,” in the nineteenth-century sense, can be
applied to an example from the sixteenth or seventeenth century, so it
is our task to determine how far the term “narrative,” as exemplified in
the nineteenth- and twentieth-century strip, really fits a given sequence
from an earlier period.
—David Kunzle (1973, 5)
Comics are a continuing saga, and there lies the rub: there is no point
in their history where we can pick up a particular paper and proclaim
it Comic Number One. This makes comics intriguing for the historian,
infuriating to the collector.
—Denis Gifford (1976, 6)
Voici mon credo, résumé par une simple proposition: pour l’historien du
9e art, la démarche correcte n’est plus d’expliquer l’histoire de la bande
dessinée à partir de la forme que nous connaissons, mais d’expliquer la
forme que nous connaissons à partir de son histoire.
—Thierry Smolderen (2012b)
To conceive comics as narratives, and therefore as something that is of

interest to narratology, is not only a matter of identifying the narrative
forms of the medium as we know them today. Various historically spe-
cific circumstances of artistic creation, publishing, and reception also
condition the conception of comics as narratives. Such factors include
the genre, the publication format, the significance of paratexts,1 the
evolving intermedial relations, as well as what the cartoonists and the
readers expect of stories in this medium in the first place. While these is-
sues may be said to exit narratology proper, that is, the description of the
means and techniques of narration, they become highly relevant when-
ever narratology moves beyond that descriptive task, and engages with
the history of the medium and the interpretation of individual works.
The given publication format of comics, 2 from newspaper strips to
trade paperbacks and graphic novels, or from yonkoma (the Japanese
246 Narrative Form and Publication Format
four-panel comic strip) and manga magazines to tankōbon books, is
associated with specific audiences and their expectations about what
kinds of stories can be told in that format. The publication format is
also tied to expectations involving the time of reading, the kind of read-
ing experience (such as emotional response, suspense, and the like),
and genre-related expectations concerning plot, characters and milieu.
Furthermore, the format affects the cartoonist’s choices on various lev-
els. Such choices may concern, for instance, the amount of complexity in
characterisation and plot, the extent of visual detail, the layout style, the
relationship between the images and the spatial form of the composition,
and the ratio between text and image.
Despite the emphasis on the means and techniques of narration, the
narratological enterprise can also be grounded in the historical develop-
ment of the given genre, publication format, or medium. One approach
that has gained some popularity recently in this respect is a diachronic
inquiry into the history of narrative forms. This means, more precisely,
the investigation of the basic narrative techniques, the functions of these
techniques, and artistic solutions, that are available in specific forms of
production and publication, genre, or body of work at a particular time.
In this chapter, I will pursue such a diachronic investigation by examin-
ing the main options that were available for organising a picture story
in the historical context of early nineteenth-century British satirical and
narrative drawings, published in the new format of the caricature maga
zines. In this investigation, I will privilege the development of certain
formal and compositional features, in particular the conception of panel
arrangement and panel relations. Thus, the objective is to show, on the
one hand, that narratological inquiry can be significant for understand-
ing the historical relation between the publication format and the means
of storytelling and, on the other hand, to argue that the said caricature
magazines contributed in important ways to the development of narra-
tive comics as we know them today.
Conceptions of Narrative
As earlier research in the history of modern comics has shown, the art
of caricature in the satirical broadsheets published and circulated in
the Georgian era in Britain allowed rich experimentation with graphic
style, caricature, and the story form. Gradually, over the course of the
nineteenth century, coinciding with the remarkable growth of the print
industry at this time, the satirical and humorous drawings of the broad-
sheet print progressed into the newspaper cartoon and the comic strip.3
Hitherto, however, the histories of comics have said very little about
the development of the narrative form from a narratological point of
view. What makes the caricature and narrative drawings of the early
nineteenth-century British context particularly pertinent for our discussion
Picture Story and Narrative Organisation 247
is that here, before the demands of mass-produced comic strips,4 the space
of the composition and the narrative form of the picture story were con-
ceived as a dynamic field of experimentation. In part, this experimentation
was enabled by the features and allowances of the new popular printed
formats. Many basic formal features of modern comics developed in the
satirical broadsheet print, caricature magazines, and illustrated magazines
of this time, yet unhindered by standardized forms of the panel strip or ex-
pectations pertaining to what ‘comics’ are. While we may recognise many
of these works as comic strips in the contemporary sense, no ‘cartoon’ or
‘comic’ was still produced under such definitions and expectations.
The further we go back in the nineteenth century, the more contingent
and relative the notion of a ‘cartoon’, let alone the terms ‘comic strip’ and
‘comics’, become. The use of the term ‘cartoon’ in reference to humorous
and satirical drawings published in printed press can be dated back to
the weekly Punch that on 15 July 1843 featured John Leech’s caricature
“Cartoon, No. 1—Substance and Shadow” that we have discussed in
an earlier chapter. From then on, in association with the satirical topi-
cal drawings in Punch, the term transformed into a general notion for
caricature of this kind, potentially also in reference to sequential stories.
By contrast, the terms ‘comic’ or ‘comics’ derived much later and much
more gradually from the comic content of the popular newspaper strips,
the establishment of publication formats that concentrated on sequential
stories (the ‘comic strip’ and the ‘comic book’), and names of popular
titles, such as the British illustrated weekly Comic Cuts5 that affirmed
the close connection between a drawn story and comical content. The
contemporary meanings of the ‘comic strip’ and ‘comics’ were estab-
lished only after World War I.6
There is no one historical beginning for comics as a medium or a form
of art. Yet, we can locate various important, and sometimes disconti
nuous, developments of graphic expression and picture story in different
cultural contexts, artistic traditions, and publication formats that con-
tributed to its emergence. In his studies on the origins of modern comics,
Thierry Smolderen has emphasised the importance of two publication
formats that functioned as kinds of cartooning laboratories for modern
comics: the caricatural broadsheet prints in the Hogarthian lineage, and
the mid-nineteenth-century illustrated magazines in Britain and France.
The popular British mid-nineteenth-century illustrated periodicals and
weeklies, such as The Penny Magazine, Figaro in London, Punch, The
Illustrated London News, The Pictorial Times, and The Graphic, ac-
commodated what Smolderen calls ‘polygraphic experiments’, graphic
solutions in visual representation, and hybridising semiotic practices de-
rived from cartoons (in the contemporary sense of single-image draw-
ings). For Smolderen, the graphic solutions that emerged from these
periodicals included, for instance, the development of the clear line and
the modelled line, the possibility to combine various graphic styles,
the schematic depiction of an instantaneous moment, the use of speech
balloons, and the art of caricature to depict posture and physiognomic
expressions and register shock, surprise, and violent emotion (Smolderen
2012a, 2014a, 3, 2014b, 53–54). These solutions, Smolderen claims,
together with Rodolphe Töpffer’s ironic novels in print that were first
published in 1837, suggest the immediate historical formal context for
modern comics, in particular the comic strip.
Importantly, Smolderen problematises the use of an axiomatic defi-
nition of comics as a sequential art, and questions the type of comics
history that concerns itself only with formats that correspond to a con-
temporary definition of the medium (2012b, 2014b, 60). What makes
the transposition of the idea of sequential art to the nineteenth-century
context particularly problematic is that illustrators and caricaturists in
Europe at this time were engaged in hybrid semiotic practices with draw-
ing and writing that, rather than developing the sequential form, fo-
cussed on means of humor and satire (Smolderen 2012a, 2014a, 50–51).
Yet, what remains open in Smolderen’s history of comics, and perhaps
overshadowed by his emphasis on the development of the polygraphic
visual field and Töpffer’s ironic treatment of the sequential form, is the
question of the narrative form and function of the new graphic solutions.
Much more could be said about the development of the story form and
the narrative organisation of picture stories in this historical context,
both in relation to the sequential form and beyond.
The question of narrative sequentiality in early nineteenth-century
caricature and picture story is more complex than a contrast between
a Hogarthian series of images, where the linking between the images
takes considerable effort and reading/viewing is necessarily slow, and
the emerging mass-produced narrative strip design that enables a more
fluid way of reading. The opposition that Smolderen establishes along
these lines allows him to argue that the emergence of the modern comic
strip in the nineteenth-century illustrated journals, newspapers, and
comic magazines pushed to the side the Hogarthian complexity of the
picture field. Thus, also, the principles of simultaneity and sequentiality
are contrasted: the caricaturists and artists of the new illustrated maga
zines and newspapers were constantly preoccupied, Smolderen claims,
with “the contrast between simultaneous presentation (serpentine and
intricate) and sequential content (straightforward action and/or rigid
social scripts)” (2014a, 84). The emergence of the newspaper comic strip
may then be said to be the culmination of the developments that enabled
cartoonists to better link images in series through image content, and
thus create the illusion of continuous action and motion, inspired by the
new recording technologies of photography and film.
However, it needs to be stressed that in the early nineteenth-century
context the sequential forms of graphic art were not restricted to sim-
ple linear forms of reading and narrative organisation or a contrast
with simultaneous presentation. Other relevant kinds of sequentiality
include back-and-forth, up-to-bottom, and bottom-to-up sequencing,
and a variety of types of spatial contiguity and temporal relations bet
ween the panels beyond the linear order, including also forms of visual
storytelling that are engaged with the total space of the composition.
Moreover, while I agree with Smolderen that the purpose in early and
mid-nineteenth-century ‘comics’ in the illustrated magazines is fre-
quently to make comical points or offer humorous digressions by telling
the story in a certain way, instead of trying to tell a story “seamlessly”,
“effortlessly”, or “effectively” (2014a, 15, 129), this does not diminish
the interest in investigating the conventions and conceptions of the story
form at this time. On the contrary, the historical forms of organising
narrative drawings in series become particularly interesting at this time
since (1) sequentiality has many varieties in these works; (2) sequential
order may be combined with other principles of organisation such as
juxtaposition and simultaneity; and, moreover, (3) storytelling in the
strip design does not always seem so effortless as may be expected from
today’s perspective. The sequential form is not a purely axiomatic con-
cept in this context, but can be grounded historically in its multifaceted
manifestations, for instance, in relation to particular traditions of cari
cature, picture story, audiences, and the evolving publication formats.
Furthermore, it is essential to note that the sequential form in the early
nineteenth-century British context is both an aspect of layout and image
content, and that in both of these aspects, the sequential structure is
used by the cartoonists to engage the readers in certain ways. In terms of
composition, the sequential form suggests, as a type of reading protocol,
that it makes narrative sense to follow a series of images in particular
order. The strip design, however, does not have to be based on the se-
quential form, even if these two usually coincide in comics.7 There are,
for instance, a rich variety of non-narrative uses of the strip design in the
satirical broadsheet prints and the later caricature magazines: descrip-
tive image series, series of caricature portraits or stereotypes, thematic
inventories, visual alphabets and dictionaries, mock manuals, juxtaposi-
tions of people, behavior, or objects for a humorous effect or the viewer’s
critical evaluation, and so on. In some exceptional cases, the strip design
may also support a nonsequential series of images, such as when the im-
ages in a strip can be read in various orders, or need to be viewed from
many directions.8
In terms of the image content, the sequential form can be affirmed in
multiple ways, such as by the recurring character or an object, or the po-
sition, orientation, and gaze of the depicted figures. Other visual mark-
ers of sequence in the early nineteenth-century picture stories contain
formal devices and instructions for the order of reading, including num-
bered panels, the placement of the captions or the speech balloons so as
to facilitate the sequencing between the panels, and the use of frames to
suggest an order of reading and viewing. By contrast, the impression of
continuous motion or action is rare in this body of work.9
Why and how the sequential form acquires prominence in the course
of the nineteenth century, and how it does become associated with the
strip design, is a complex historical issue. To answer this question, we
need to problematise, on the one hand, their historical relationship.
On the other hand, we need to open up the question of the historically
contingent models of visual narrative in this body of work, including
especially questions of narrative organisation and form.
In his ground-breaking study The Early Comic Strip, which examines
the prehistory of the comic strip from c. 1450 until 1825, David Kunzle
made the useful distinction between a picture story and a strip design,
claiming that “all narrative strips are picture stories, although not all
picture stories are, in the first instance, narrative strips” (1973, 5). The
distinction allows that the strip design, while being “more or less narra-
tive by definition”, does not always function narratively, but may have
other functions as well. The design can take, for instance, a descriptive
function, such as the creation of a panoramic view of society or the de-
piction of types of people (1973, 5). Furthermore, the strip design is not
the only compositional form that can serve narrative communication in
the broadsheet format. For Kunzle, a ‘narrative’ in the art of broadsheet
caricature accommodates both a sequence of separate images, in parti
cular in the form of the strip design, and what he calls the “single-setting
narrative” (1973, 4). The latter applies, more precisely, to engravings in
which are depicted, within a single image, two or more episodes from
the same story.
One-panel narrative cartoons were published in different formats be-
fore the emergence of the satirical magazines of the late 1820s or the
illustrated magazines a decade later. In fact, we should specify that the
single-setting narrative, in Kunzle’s definition, has two quite distinct
forms: the proper single-image narrative, which depicts some narrative
sequence in one image, and the story told in a series of separate images
collected in a narrative album in the Hogarthian style from A Harlot’s
Progress (1731) to The Four Stages of Cruelty (1751). At the same time,
the relationship between a series of images in an album and the story
told in the strip design is close, and Kunzle points out that their dis-
tinction is to some extent only a matter of presentation: any narrative
album could be presented as a strip, and any strip can be mounted in an
album (1973, 5).
Perhaps surprisingly, however, Kunzle has relatively little interest
in the historical aspects of the narrative form and the way in which
changes in the publication format contributed to the emergence of
the early comic strip. The transition to the comic strip from the late
eighteenth century to the mid and late nineteenth century involved a
number of significant changes in the publication format, from narrative
albums and broadsheet prints to caricature magazines and illustrated
satirical magazines, then comic magazines and albums. These trans-
formations were accompanied by changes in the way in which picture
stories could be told. Narratologically relevant inquiries that could be
pursued in this context involve, for instance, the development of tech-
niques for creating narrative voice, perspective, characters, plot struc-
ture (time and causality), sense of action and movement, the setting,
and narration through showing, style, and page layout. The develop-
ment of the narrative use of the speech balloon has perhaps been given
most attention in this regard.10
The rest of this chapter will discuss the available alternatives in the
organisation of picture stories in the early nineteenth-century carica-
ture and narrative drawings with a specific emphasis on the narrative
uses of the multi-panel sequence in the satirical caricature magazines
published between 1825 and 1835. The main examples are taken from
caricature magazines published in Britain and Scotland. These in-
clude, in particular, the fortnightly Glasgow Looking Glass (1825),
today considered by some as the earliest comics magazine, and which
changed its name to Northern Looking-Glass after five issues, to run
for n ineteen instalments altogether (1825–1826); the continuation of
this publication in the monthly The Looking Glass or, Caricature
Annual (1830–1836), based in London and also known as McLean’s
Monthly Sheet of C aricatures, or The Looking Glass; and the caricatur-
ist Charles Jameson Grant’s fortnightly single-page broadsheet ‘mag-
azine’, Every Body’s Album and Caricature Magazine (1834–1835),
that ran for a total of thirty-nine issues.11 The principal illustrator of
Glasgow/Northern Looking Glass and the first issue of The Look-
ing Glass was the English artist William Heath (1794–1840). Robert
Seymour (1798–1836), who is best known for his illustrations for
Charles Dickens’ novels, continued Heath’s work from the eighth in-
stalment of The Looking Glass onwards. Unlike the weekly illus-
trated magazines, such as Figaro in London (1831–1839), for which
both William Heath (under the pseudonym “Paul Pry”) and Robert
S eymour contributed, or the more successful Punch and The Illustrated
London News that started in the early 1840s, these caricature maga-
zines mainly featured cartoons and picture stories.12 In the course of
this discussion, I will make comparisons between this body of work
and illustrative examples of the British Golden Period of caricature,
roughly extending from 1780 to 1820, in order to be able to better
contextualize the forms of narrative organisation and the graphic in-
novations in the satirical magazines. These Golden Period works will
include selected narrative drawings, published as broadsheet prints,
by James Gillray (1756–1815), Richard Newton (1777–1798), George
Woodward (1760?–1809), Charles Williams (active 1797–1830), and
George Cruikshank (1792–1878).
Principles of Narrative Organisation
The early nineteenth-century picture stories in the satirical illustrated
magazines can be conceived around three operative principles of juxta-
position, sequentiality, and simultaneity. These principles of narrative
organisation are grounded in the history of caricature and the illustrated
press, but should be seen as heuristic devices rather than as historical
categories. They are not mutually exclusive, but can be combined over
a scale of variation where most cases would have to be characterized as
intermediate. In fact, pure instances of these principles are rare.
Juxtaposition
The juxtaposition of two or several images, parts of one image, or text
and image, is widely used as a means of narrative organisation and vi-
sual communication in this body of work: juxtaposition can provoke
meanings through contrast, association, extension, similarity, differ-
ence, reversal, and antithesis, among other things. It can be used as a
technique in a single image as well, between two images, or a series of
images. The effect of juxtaposition may in some cases also extend over
several images in a long strip or between two or more strips.
Humorous contrast is perhaps the most common manifestation of this
principle in the satirical prints and caricature magazines, based both
on the visual contents of the comparison and the contrast between the
caption and the images. We can observe this in the broadsheet carica-
tures, for instance, when a delightful young woman with a slim fig-
ure, a “nobody”, is contrasted with a buxom lady, who is somebody
important (Richard Newton’s “No Body Some Body”, 1795); when a
moonstruck young man who looks at a miniature of his beloved is jux-
taposed with a corpulent old wine lover, who is angered by a new wine
tax (Richard Newton’s “Contrasted Lovers”, 1796); or when four differ-
ent nationalities (the English, the Scots, the French, and the Dutch) are
contrasted through their “civilized” and “primitive” lavatorial habits
(James Gillray: “National Conveniences”, 1796). In the first issue of The
Looking-Glass (1830), in William Heath’s “Siamese youths, our own
youths”, Siamese twins, conjoined by their bodies, are compared to “the
British youths”, Duke of Wellington and Sir Robert Peel, the long-time
close political allies. The titles of these works confirm the sense of con-
trast, or similarity with a significant difference, such as physical figure,
cultural context, social standing, between the figures thus juxtaposed.
The many narrative uses of juxtaposition in these works include,
especially, contrasted perspectival changes, speech and dialogue, or
temporal framework (“before” and “after”, “then” and “now”), and
certain devices of spatial division that call for a narrative response. For
instance, changes of perspective, such as shifts between close-up and
distance, or between different points of view on the same situation, can
suggest a magnification of the situation or an effect of accumulation.
William Heath parodies the English actor Edmund Kean (1787–1833)
in the two-part caricature entitled “Kean in America”, published in
December 1825 in Northern Looking-Glass, by juxtaposing two images
of Mr Keen performing Shakespeare for American audiences. In these
images we can read Kean citing lines from Richard III and Richard II
while he is greeted with insults from the audience, who consists mostly
of A frican-Americans. In the first image, round objects, sticks, and what
seems to be a dead cat are thrown onto him on the stage. In the second
image, we then see Kean much closer, citing Richard II in the same cos-
tume, but this time he no longer appears to be on the stage, but pleading
in a street with a copy of Thomas Moss’s “Beggar’s Petition” in front of
him. The two African-Americans who pass and speak stereotypically
broken English recognize the actor, calling him “a very bad man” and
“a naughty man”. The changing perspective and setting, as well as the
dialogue in the speech balloons, make a satirical point, but also de-
scribe a dramatic situation. Thus, this two-part image has the potential
to be conceived both as a deepening of the satirical take on the same
situation—Keen’s humiliation in front of the American audiences—by
means of juxtaposition, and as a sequence that shows how his situa-
tion is deteriorating. Our choice, whether to regard this as a narrative
sequence or not, would partly depend on how much emphasis we are
ready to put on the temporal and causal connections between the two
panels. The juxtaposition itself has weak narrativity.
Dialogue, in the literal sense of a conversational exchange, can
strengthen the sense of duration between the images and elements in
juxtaposition. The two panels in George Cruikshank and the anony-
mous artist J. Pxxxy’s broadsheet print entitled “Back & Front View of
the Ladies Fancy-Man, Paddy Carey O’Killus” (1822) feature extended
dialogue in speech balloons between the spectators gathered around
the newly erected Duke of Wellington statue in Hyde Park shown from
two juxtaposed angles. The statue, where Wellington is allegorized as
Achilles, was London’s first nude sculpture, albeit garnished with a
fig-leaf, and is shown here to be an intense object of curiosity and look-
ing. One of the spectators uses a lorgnette to better see the details, many
people point at the statue, even a telescope is directed at the monument,
and the spectators utter excited comments on both sides of the statue
(Figure 10.1). The first panel shows the statue from the back and the
second from the front; the two panels are divided by a gutter. Most spec-
tators who crowd around the monument are women, with the exception
of Wellington himself and two children. Wellington is caricatured in the
first image in uniform looking at the statue in profile and saying: “The
Honor is so great, that all I can say by the Powers, is that I’m Speech-
less”. The dominant role of women in these images is a reference to the
fact that the statue was funded by subscriptions from women in honour
of the Duke’s services for the nation.
Figure 10.1 G eorge Cruikshank and J. Pxxxy’s “Back & Front View of the
Ladies Fancy-Man, Paddy Carey O’Killus” (1822). Courtesy of
The Lewis Walpole Library, Yale University.
The spectators featured in the juxtaposed panels are not the same,
but form similar groups around the statue. This could suggest that the
perspectives represent two distinct but close moments. Some of the utter-
ances are spontaneous individual remarks, such as the comment “I see
it!!”, which seems to come from the person who is shown to use telescope
to better grasp all the details of the statue. However, most of these utter-
ances involve an exchange of words, such as the question of the fig-leaf
that extends to a longer conversation in speech balloons: “This will be a
place of great attraction in the hight of the Season/You mean the fall of
the Leaf I suppose?/I would not give a fig for it/Well, for my Part I think
it a great ugly useless thing/Pray men, have you seen the Original one—
at Rome/O’yes—the Original is much finer/I don’t think its quite the
thing” (underlining in the original for emphasis). The situation is thus
extensively satirised by means of images and dialogue alike from two
juxtaposed angles. Nevertheless, this is not a narrative in a literal sense,
for the lack of temporal development and causal connection between the
panels, but represents a situation with much narrative potential.
The juxtaposition between two or more images or elements in one im-
age can attain a stronger sense of narrativity whenever it is supplemented
by some causal connection or a sense of development in a continuing
temporal frame. This can occur, especially, when juxtaposition takes
the rudimentary narrative form of two or more polarised moments and
situations in comparison.13 For instance, William Heath’s picture story
“Politics” from Northern Looking-Glass (18 August 1825) depicts, in
two vertical strips of four panels, the daily habits of the King of France
at his castle from the morning, when the newspapers are read to him,
until the evening when he is shown to go to bed.14 There is no causal
link or sense of continuity between the depicted actions. These are just
typical events—breakfasting, going to mass, shooting sparrows in the
park, and playing with the dogs and grandchildren—during the King’s
typical day. Is this picture story, then, based on a juxtaposition of situ-
ations or a sequence of episodes? The temporal and spatial frame gives
the panels a sense of sequential order. There are, also, some thematic and
visual correspondances across the two strips, based on the structure of
juxtaposition, and that potentially emphasise the sense of repetition. For
instance, the last panels of each strip show the King lying down, “reclin-
ing” on his bed during the day and going to bed at the end of the day.
Juxtaposition thus connects the sense of temporal progress with cyclical
structure: events are in sequence, within the temporal frame of 24 hours,
but also contrasted and connected across the chronological order.
A more extensive version of a similar arrangement can be found in
Charles Williams’s broadsheet “The Two Journals” (1814) that con-
trasts, in two juxtaposed two-tier strips with eight panels, the daily life
of two royal figures: the virtuous and studious Tsar Alexander, who is
visiting England and spending a day in London in the most beneficial
way (taking a walk to Kensington, taking notes of a military drill, writ-
ing a letter to his wife, etc.), and the idle and vain Prince of Wales who
spends the day at his palace. In contrast to William Heath’s depiction
of the daily chores of the French King, we can observe here a clearly
continuing and causally motivated string of events that unfolds from one
panel to another. The sense of narrative sequence is achieved both on the
level of showing that describes connected phases of one event and, per-
haps more significantly, by means of first-person verbal narration that
provides the story with bridges between the panels. For instance, in the
first panel of the Prince of Wales’s strip journal, we see the Regent in
the bed and can read in the accompanying caption that “Boozy and
sick, with aching head,/Toss’d sleepless, on my swan down bed./Sunk
tow’rds morning in a dose,/When dreams of frightfull import rose”. The
next panel shows the Regent still in the bed, uneasily asleep, and having
a vision of a demon that his wife wards off. In the third panel, there
are three valets working on the Regent’s hair or wig. The situation has
changed, but the caption, or the citation from his journal, explains what
has happened: the Regent has finally woken up at noon and is now hav-
ing his hair done (and this will last for 2 hours). Similarly, the following
panels suggest narratively cohesive structure where both what is verbally
narrated and what is shown in the images can be followed as a sequence.
As a whole, however, the satirical point of the composition is based on
the juxtaposition between the two royal figures and their contrasting
habits and characteristics.15
Juxtaposition can also be incorporated within the genre of a ‘prog-
ress’ story in a series of images, made famous by Hogarth’s prints that
contrast various phases of a person’s life, typically reflecting the char-
acter’s career and moral development. The progress story continues to
be varied throughout the Golden Period of Caricature in the works of
James Gillray, John Nixon, Richard Newton, and George Cruikshank,
and equally, William Heath makes various versions of it in the cari-
cature magazines. In this genre, juxtaposition becomes an element of
the sequential form. For instance, a temporal and causal sequence can
be used to make a satirical point by means of contrast, or a series of
events may transform into a structure of juxtaposition between two
or more ideas, themes, or perspectives. William Heath’s picture stories
“Essay on Modern Medical Education”, from Northern Looking-Glass
(Vol. 1, No. 6), and “Life of a Soldier” (Vol. 1, Nos 10–16) are examples
of such shifting emphasis. In “Essay on Modern Medical Education”,
which is a series of ten panels satirising modern medical education,
some of the panels suggest a sense of temporal progression and nar-
rative sequence, while elsewhere the sense of sequence is weak or non-
existent. The first three images in this series do not imply a narrative
logic as a sequence, but function as distinct caricatures. They depict
a mob protesting the grave robbers who dug up bodies to sell them to
anatomists at the medical schools (“The alarm, or the Kirk Yard in
danger”), drunken behaviour at a funeral club (“The dead association”)
and a meeting to counteract bodysnatching (“Watching and warding”).
The following second instalment with three panels, however, suggests
some degree of temporal progress. All of these images follow student
life at a medical institution, from an anatomy lecture, where the stolen
body parts were dissected, to pranks in an apothecary’s shop, and the
graduation ceremony.
The numbering of the panels from one to ten, and the use of the phrase
“to be continued” at the end of the second instalment, obviously indi-
cate that the story will continue, and the mention of “continued” in the
beginning of the last instalment, further emphasise the fact. Yet, it is
important to note that the last four panels that illustrate practising doc-
tors at work are organised in two tiers of two panels based on a satirical
contrast rather than temporal development or narrative sequence. The
captions of the images present the panels in pairs of two. The practicing
of vivisection on animals (“Preparing for practice”) is contrasted with
the amputation of a man’s leg with an axe (“Actual practice”), whereas
the poor hospital conditions of home (“At home”), where skeletons
are seen to tend the sick and dying, are juxtaposed with the terrible
circumstances of a battle field, where surgeons attempt to reassemble the
mutilated soldiers’ severed legs and head (“abroad”).
However, the polarised structure of ‘before’ and ‘after’, or the con-
trasting of events within some other temporal frame such as ‘then’ and
‘now’, do not always serve a narrative function. Such temporal structures
can also make a statement, that is, have an argumentative function. For
instance, James Gillray’s broadsheet image “The Blessings of Peace, the
Curses of War” (1795), which is dedicated to the Parliament of the Great
Britain, contrasts, in two roundels or medallions, an image of peace
and war to make an anti-war statement. Here, the first image shows
happy family life in a prosperous British home with the caption “such
Britain was!”, while the the second image portrays a family outside a de-
stroyed farmhouse lamenting the death of the father (or husband), who
is stabbed to death with a bayonet somewhere in the warring Europe,
with the caption “such Flanders, Spain, Holland, now is!”. The topi-
cal contrast between “Blessings of peace & prosperity and domestick-
happiness” and “The curses of War, invasion, massacre & desolation” is
also accompanied by the inscription, placed in between the upper parts
of the roundels, “from such reverse O GRACIOUS GOD, preserve Our
Country!!!”. The inscription specifies the purpose of the comparison by
giving this contrast a logic of an argument: the possibility of a war at
home that needs to be prevented, and the threat of a French invasion
that needs to be addressed. Similarly, in George Cruikshank’s caricature
“John Bull’s Three Stages or, From Good to bad & From Bad to Worse”,
published in the political monthly The Scourge (1811–1816) on 1 March
1815, the contrasted situations in the life of John Bull and his family
make an argument against war. In this folded colour strip, the three
panels that are entitled “Before the war”, “During the war”, and “Peace
with all the world” show first how war deprives the family of the neces-
sities of life, then changes their physical appearance from rosy-cheeked,
robust, and cheery to a starved and miserable state, and finally how the
following peace only worsens their situation. John Bull and his family
eat a large meal before the war, much less during the conflict, and have
only bones on their plates when the peace arrives.
In some broadsheet caricatures as well as narrative drawings in the
satirical magazines, the structure of juxtaposition can emphasise the
spatial division of the image that reflects two neatly distinct spheres or
realms of life. On the one hand, we can observe this in images where
some aspect of the image suggest an inner division and thus potentially
undermines the distinction between a single image and two panels. For
instance, in James Gillray’s caricature entitled “BLACK-DICK turn’d
Taylor” (1788) a tailor’s shop-board suggests a division of the image field
into two relatively independent visual spaces. In the upper part we can
see Admiral Richard Howe, known to his men as “Black Dick”, sitting
on a tailor’s board and working on a naval coat. In a speech balloon,
the Admiral explains his plans for reform—new uniforms, ships, ap-
pointing young (unexperienced) officiers and admirals. In the lower
section of the image, a demon echoes the Admiral’s words and reveals
his own plan to control the Admiral’s reform (“I shall get you before
you are aware of it”). The Admiral is not aware of the demons working
underneath him, but the demons can see and hear the Admiral from
below. Similarly, in one of Charles Jameson Grant’s several cartoons in
the 15 September 1834 issue of Every Body’s Album that focus on the
theme of angling, the water line serves as a virtual frame inside the im-
age dividing it into two sections. Above the water line, a fisherman sits
waiting with his hook and line in the water, while below the water line
two fish, unseen by the fisherman, converse and mock the man and his
methods in their own virtual panel (Figure 10.2).
On the other hand, the juxtaposition of elements can contribute to
spatial division in the embedding. For instance, in Gillray’s caricature
“Hope” (1802) we perceive, in the left side of the image, the symbolic
figure of John Bull standing in the lobby outside the House of Commons,
voicing his concerns about the economical situation and the budget plans,
while through the open door on his right we see Prime Minister William
Addington making a budget speech inside the House of Commons. John
Bull’s rumination and comments (“Let me see—25 millions. How are
we ruined? Income Tax taken off! Well! Well! Well!”), concerning Add-
ington’s abolishment of the unpopular wartime income tax, function as
a means of connection between the embedding and embedded image.
The two men, furthermore, are associated and at once contrasted by
their positions and physical shape. They both stand, in profile to the
right, with one leg forward, within the same perspectival line, while the
speech balloon, superimposed on the embedded image, directs our gaze
to Addington in the House of Commons.16
In yet other kinds of broadsheet caricatures, certain visual means of
connectivity, such as speech balloons, lines of sight and perspective,
and shared background space, can create thematic, conceptual or scenic
connections between juxtaposed images without a particular narrative
function. For instance, the positioning of the speech balloons and the
shared background in James Gillray’s satirical print “Nelson’s Victory,
or, Good-News Operating Upon Loyal Feelings” (1798) forges connec-
tions between different reactions of desperation and disappointment by
the various Members of the Opposition when they have received the
news of Horatio Nelson’s victory at The Battle of the Nile. In Richard
Newton’s “Sketches in a Shaving Shop” (1794), which depicts humorous
scenes at a barber shop, some of the twelve featured barbers’ utterances
or lines of sight, which point outside the panel frames, indicate that there
is a larger social situation taking place at the barber shop beyond what
is shown in the image. In these juxtapositions of similar situations, the
juxtaposed elements may offer an incentive for further interpretation,
Figure 10.2 © The British Library Board, LOU.LON 1052. Charles Jameson
Grant. Fisherman and the fish. Every Body’s Album (September
1834).
i.e. prompt the viewers to think how the things are related. Juxtaposi-
tion, then, functions as a question, a joke, or a kind of enigma rather
than a narrative.
Sequentiality
In the historical context of the early nineteenth-century British satirical
magazines, the sequential form of narrative organisation operates on
a continuum of various kinds of temporal and causal relations. In this
regard, we can distinguish two basic forms of sequentiality: the episodic
sequential form and the narrative continuity sequence. Furthermore,
the sequential form of these picture stories can interact dynamically with
forms of organisation and layout that are based on juxtaposition and
meaningful contrast, or the sense of simultaneity between the images
that invites a more global look at the composition.
The episodic sequential organisation involves two or more images in
series where each image represents a stage or phase of a situation or
an event. The episodic picture stories in the caricature magazines and
broadsheets of 1825–1835 exhibit considerable variety in this regard,
and the balance between episodic and more continuous arrangement can
also shift in the course of the story. At one extreme, these publications
feature picture stories where an episode in a single image is more or
less self-contained and, consequently, the order of the episodes may be
rearranged without affecting the work as a whole. In these cases, the
logic of the sequence is based on a recurring character, place, or theme;
the causal connection between the visual contents, actions, or speech
situations included in the images remains weak. At the other extreme,
episodic picture stories can include, both in the images and the texts,
strong markers of causal development that necessitate that the sequence
is read in a certain order. The genre of ‘progresses’ is the most common
genre in this kind of episodic arrangement. In the ‘progress’ story, the
passing of time between one image and another can usually be observed
in the main character’s changing physique and social standing.
We can count at least three picture stories as clear cases of the progress
genre in Northern Looking Glass: “Life of a Soldier”, “Life of a Sailor”,
and “Life of an Actress”. While the principle of juxtaposition operates
in these stories as well, it is overshadowed by sequential structure. The
longest of these, “Life of a Soldier”,17 which consists of twelve numbered
cartoons (in our modern sense of the term) in several instalments, de-
scribes a soldier’s career. The contents of the images reflect the idea of
significant episodes in a developing career: the first three panels portray
the soldiers’ training and promotion, the next pair of panels show the
embarkation and the ensuing battle, and the following pair of images de-
pict a fight between a British soldier on the ground and someone else—
the opponent is dressed in what may be an Arab costume—and then
shows a wounded soldier walking on crutches. The numbering of the
panels from one to twelve, and the use of the phrases “to be continued”
and “Cont’d”, even if this is not altogether systematic, give indications of
the order of reading. The first instalment in issue 10 (14 November 1825)
has three panels, the first of these folding out from the middle of the page
to extend the width of one panel. We need to note in this respect that
panels numbered 5 and 6 (third instalment, “The Embarkation”/“The
Action”) are curiously printed in the reverse order from right to left: the
panel on the left is accompanied with the mention “to be continued”
(Figure 10.3). While this is likely to be a mistake in printing, we can also
speculate that another reason for the numbering of the panels and the
use of the phrase “to be continued” might be to guarantee that all read-
ers can understand how these narrative drawings are set in sequence.18
Yet, some features of “Life of a Soldier” also suggest that the sense of
narrative sequence is quite weak or ambivalent in this case. On the level
of the image content, the instalments show various situations in a sol-
dier’s career, from hiring a new conscript to becoming an adjutant. One
challenge in conceiving these images in terms of a continuing narrative,
at least in the first instalments, is the protagonist’s somewhat uncertain
visual identity. In fact, given that the soldiers in each panel look some-
what different, can have their backs turned towards the viewer, or wear
hats that cover parts of their face, it is possible to read “Life of a Soldier”
alternatively as a story of an individual British soldier or as a story of un-
identified conscripts who represent typical situations in the course of a
military career. That said, however, the accompanying written explana-
tions of the story in the issues of 12 and 13 of Northern Looking Glass
(12 and 26 December 1825), after the narrative had run for four panels
in two previous issues, make it evident that the story is anchored on one
particular British soldier.
Figure 10.3 William Heath. “Life of a Soldier.” Northern Looking Glass.

No. XII. 12 December 1825.
The episodic situations in this story are linked thematically within
the biographical frame of the soldier’s career. Within each instalment,
moreover, the images have more or less evident causal relations with
each other. For instance, the scene of embarking on a navy ship is fol-
lowed by a battle scene; an image showing a fallen soldier fighting on
the ground is followed by an image of a wounded soldier on crutches;
and after a British soldier is seen struggling to seize a French eagle from
a French soldier, he is, then, having captured the eagle, being presented
to the Duke of Wellington (panels 9 and 10).
In contrast to the initially loose but gradually more coherent narrative
sequence in “Life of a Soldier”, William Heath’s “Life of an A ctress”
(Northern Looking Glass, Vol. 1, No. X, 14 November 1825) is an
imaginative take on the progress genre where the eight images of the
strip create a temporally and causally tight and coherent sequence. The
strip design has also been put to an expressive use: the ascending ladder
frame of the strip, with one panel ‘placed’ on each broadening step, ac-
companied by the gradual increase in the size of the panels, provide a
clear symbolic frame for the actress’s gradual climb on the social ladder
(Figure 10.4). At the same time, the steps of the rustic ladder indicate
the order of viewing the images. The ascending viewing protocol, thus,
echoes the actress’s societal climb from the poverty of her home to the
fame of the stage, and through widowhood and wealth to a noble status.
Moreover, there is considerable narrative coherence in the image content
throughout the sequence: the heroine and the signs of her increasing
wealth, social standing, and growing corpulence are identifiable in each
panel. The contemporary readers were likely to recognize in this de-
piction the popular actress Harriot Mellon (1775–1837) who, born in
a family of strolling players in a travelling theatre company, became,
successively, the wealthy widow of a London banker and the Duchess of
St. Albans.
As “Life of an Actress” goes to show, the progress genre can be real-
ised in the strip design or, alternately, the strip design can accommodate
the progress story.19 In “Life of an Actress” we must note, moreover,
how the design of the strip supports, ironically, the idea of progress as
a form of social climbing. The order of the reading is significant, both
for the theme of the strip, but also from our historical perspective con-
cerning the development of narrative forms. “Life of an Actress”, with
panels to be read from the low right-hand corner of the strip upwards to
the left, illustrates how the idea of sequence is quite flexible with regard
to the order of reading in this context.
What makes some narrative a comic strip, in the sense that we under-
stand this form of expression today, instead of the episodic picture story?
As we have seen, the progress story and its variations are based on the-
matic, temporal, and causal relations, often in the sense of juxtaposition,
between the contents of a series of images. Narratives in the strip design
Figure 10.4 William Heath. “Life of an Actress”. Northern Looking Glass X.
14 November 1825.
in these same publications can follow a similar logic. The difference bet
ween these two, then, is the kinds of thematic, temporal, or causal links
that can be recognised between the images. The narrative use of the strip
design creates the impression of a sequenced, rather than juxtaposed,
event, experience, or situation, and of continuity rather than compari-
son. Typically, the logic of narrative sequence predominates also on the
level of the image content; it is not just implied by the juxtaposition of
phases in some temporally structured development or sense of progress.
However, in some cases, the progress story can suggest temporal and
causal relations (consequence) between images in ways that move the
genre close to a narrative comic strip. For instance, David Kunzle notes
that in John Nixon’s picture story “Progress of Passion” (1792, etched by
Isaac Cruikshank), the narration of the continuous displacement of an
emotional reaction ditinguishes this work from most previous progress
stories (Kunzle 1973, 363). Nixon’s two-tier strip print shows a sequence
of nine scenes starting, in upper left, with King George III dismissing the
Tory politician and Chancellor Edward Thurlow, followed by a chain
of events resulting from Thurlow’s anger that affects his entire family
and household, servants, the butcher, a dog, and finally a cat that kills a
mouse. Thus, we can follow a chain of causally linked events from one
panel to another. What is seen in one image explains what has been seen
before, but also points to what comes after. In each scene of this cycle
of abuse, the figure who has suffered from someone’s anger, scolding, or
physical attack, is consequently shown to perpetuate the abuse to others.
Thus, on the one hand, passion “progresses” downward in terms of the
figures’ social status while, on the other hand, the genre of the progress
story, based on comparisons between stages of the process that invite
careful study of the occurred changes, is transformed into a closely tied
chain of events.
We need to illustrate the crucial point about sequenced rather than
juxtaposed structure since the distinction between the episodic and the
narrative sequence can remain rather indeterminate. Characteristically,
in the caricature magazines of the 1820s and the 1830s, when the strip
design is used to deliver a narrative sequence, we can observe a sense of
temporal and causal continuity on the level of the image content, as well
as between image and text, and not just within the general temporal
frame of a ‘progress’ or a ‘life’. This entails, furthermore, that the tem-
poral gaps between the images are sufficiently close to maintain a sense
of continuity instead of mere juxtaposition, contrast, or comparison.
What is sufficiently close in this context is, of course, relative to the
kind of story and its representation of time; and what one reader might
perceive as narrative continuity someone else may perhaps experience as
discontinuous. Yet, some distinctions can be made.
We can observe the operation of narrative continuity sequence, for in-
stance, in the first sequential comic strip in Glasgow/Northern Looking
Glass, “The History of a Coat”, which ran from issues 4 to 6 between
23 July and 18 August 1825. The story was published in three instal-
ments, respectively, with four, three, and two pairs of panels in two
vertical columns. The first instalment does not have captions. The layout
is not without ambivalence in the sense of sequence: the order of reading
the strip design is not quite obvious from the outset due to the fact that
the two rows of panels are separated by a full line (or frame) vertically,
and a partial line horizontally. This could suggest, for instance, that the
sequence is meant to be read from top to bottom. The content of the
images, nevertheless, clearly indicates that the story of the coat evolves
horizontally. In the second and the third instalments, the order of read-
ing is further confirmed by the captions. In the images, we can follow
the gradual making of the coat, from the shearing of a sheep to the loom
and the tailor’s table, then the slow degradation of the cloth—in the true
spirit of the ‘progress’ story—from a gentleman’s coat to a scarecrow’s
wooden shoulders and, finally, a dirty cloth torn by ragged pigs in a
panel with a caption “End of the Tail”.
The strip has the basic features of a moral life story, exactly as in the
progress genre, but the difference, from the contrasted phases of a ‘prog-
ress’ story, is the greater reliance on the sense of consequence between
the panels and the need to work out their relations in the story. In other
words, the impression of temporal and causal continuity between the
panels is more crucial here than in the Hogarthian progress stories that
revolve around the principle of comparison between the phases of a life
or a career.
Another indication of how the sequential form moves closer to the
narrative use of the strip design is awareness of the gap between the
panels as a form of temporal or logical discontinuity that needs to be
bridged and explained. One illustrative case of an episodic picture story
turning more strip-like in this sense, based on quite literal awareness
of the panel relation as a meaningful gap of information, can be found
in a picture story entitled “Scenes from a Historical Drama” from The
Looking Glass (No. 36, 1 December 1832). The story describes the his-
tory of Belgium and Holland in six panels where the first three images,
referred to as the “first act”, follow an episodic logic, depicting respec-
tively the French conquests in the Netherlands in 1795, the liberation of
Holland in 1813–1814, and the enlargement of Holland at the Congress
of Vienna 1814–1815 to include what in 1830 became modern Belgium.
Importantly, there is an inscription (sideways) in the blank space between
the third and the fourth panels: “Here is supposed to occur the space of
Sixteen years from 1815 to 1831”. The space between the panels is thus
perceived as a dramatic ‘intermission’, while it also marks a shift into
a faster rhythm and more closely tied sequential structure. The gap of
16 years is followed by three panels that describe, in a faster succession of
events and with several recurring characters (Belga for Belgium, Leopold I,
and William I), the developments of the last 2 years: the B elgian revolu-
tion, the installation of Leopold I as the King of the Belgians, and King
William’s subsequent military attempt to reconquer Belgium. The theat-
rical metaphor of this historical chain of events is extended to the very
end of the sequence where we can read: “We know not the denouement
of the piece, but look for poetical justice”.
The use of reoccurring characters in continuing action or situation
distances the strip design from the episodic structure. Charles Jameson
Grant’s strip “The Adventures of the Buggings’s”, published in Every
Body’s Album (No. 14, 15 July 1834), which tells the family Buggings’
mishaps on a journey to Gravesend and back in nine panels, affirms the
sequential form at various levels simultaneously. Here we can follow the
continuing group of characters and their movement, observe the tem-
poral connections and shared framework between the panels (a day’s
journey from morning until night), and read the captions and the speech
balloons that make various relevant causal links between the panels.
In the morning, the taxman Buggings’ family of six starts off slowly
after having enjoyed twenty-four rolls and tea. Running late from their
steamer, they hurry to catch boat on a bark, then sleep over the stop in
Gravesend and end up in Margate where, just before landing, they all
fall into the sea after their son, Bob, who had been looking for eels in
the water. Finally, they return home in a Margate coach at night, all wet
from their bath and the pouring rain, only to find their house destroyed
in fire. The disaster, as it is ironically explained, thus forms “a kind of
Consumation to their day’s pleasure”. In the last panel, Mr Buggings
reassures his wife that he will make up for all the losses, in the manner
of all great men, by a few “false accounts”.
Still other means of narrative continuity involve causal connections
through juxtaposed perspectives and voices. In the five-panel narrative
entitled “Colonial Slavery” from The Looking Glass (No. 8, 1 August
1830), the narrative use of the strip design is not based on the process of
following a recurring character through the strip, but on making causal
connections between the depicted events, situations, and their speakers.
The comic involves the real story of a female slave, so-called Poor Black
Kate, who died of her owners’ brutal treatment in the Bahamas in 1826.
In the first panel, the slave owner named Mr Henry Moss of Crooked
Island is threatening to flog the slave, who lies in a plantation field, while
the woman’s companion, a male slave, explains that she’s already dead.
The second panel portrays the slave owner, standing on what seems to be
a terrace of his house, justifying his brutality. He is accompanied by his
wife, while a black body can be seen lying in the background. It is note-
worthy that in this panel, the mean-looking colonial couple has turned
towards the reader. The man, looking directly at the reader of the strip,
says: “Wot are you stareing at? Shant a man do as he likes with his own
Ax. Your Duke of N-C-L”. The reader is thus called upon to evaluate the
slaver’s actions and the justification of his claim (Figure 10.5). The three
smaller panels in the lower row of the strip portray a chain of political
reactions following the event. First, a colonial governor (Lt-General Sir
Lewis Grant) is shown to appeal to the Members of Parliament on be-
half of the slave owner, and then a Member of Parliament (Sir George
Murray, Colonial Secretary) comments on the event on behalf of the
slave owner and approvingly of the Governor’s stance. Finally, the last
panel shows two electors in conversation, reacting to the event. One
of them is crying out “MURDER!!!”, while the other wonders: “What
‘sinnifies’ It’s My Intrest”.
The verbal narration in this strip, involving both the captions and
the reported speech of the characters, is crucial in creating narrative
coherence between the five panels. The captions identify the main char-
acters and the chain of events: “THE SLAVE”, “THE PLANTER who
Murderd the Slave!!!”, “GOVENOR sorry to punish the murdrers!!!”,
“The MPs who approve of the GOVENOR”, and “the ELECTORS who
return the MP!!!”. Beyond the captions, all panels include the charac-
ters’ viewpoints, which are given either in speech balloons or in quoted
writing (the Governor is seen writing a text). These viewpoints showcase
governmental hypocrisy in the matter and provoke the reader to evalu-
ate the justice or, perhaps rather, injustice, done. Thus, the strip actively
prompts the reader to make connections between the various scenes and
Figure 10.5 Robert Seymour. “Colonial Slavery.” The Looking Glass No. 8.
1 August 1830.
situations, not just in order to undertand the chain of events, but to per-
ceive the various political stances in the case and take a stand of one’s
own as well. 20 This strip is, then, a narrative representation of a chain of
events and positions that portray how the judicial and political system
attempts to justify the system of slavery.
Simultaneity
As we have already seen, the strip design can be employed for other than
narrative purposes. This can involve, for instance, description, argument,
or the development of a theme. In strips caricaturing the same character
in different situations or sketching different types of people, the strip
design can serve the purpose of inventory and accumulation rather than
narrative sequence. The descriptive use of the strip form may also fol-
low a spatial logic. In William Heath’s full-page drawing “My House in
Town” in Northern Looking Glass (Vol. 1, No. XV, 23 January 1826),
the twenty picture panels of this work are set in eight rows, representing
an interior of a room enclosed in a town house from the coal and wine
cellars to the garret. There is no sense of narrative sequence between the
panels. Instead, most panels depict recognizable situations that may im-
ply story-like scenarios: a footman tumbling down with his tray on the
floor in the “Ladies’ Withdrawing Room”; toasting and heavy drinking
in the “Gentlemen’s Withdrawing Room”; servants playing cards in the
Laundry; a crippled man in a Study pulls the bell, while trying with his
crutch to stop a dog from biting a cat; a footman hugging a coy maidser-
vant in the Kitchen; and so on. The space of the house, thereby, allows
the humorous portrayal of a cross-section of society and many evolving
simultaneous situations in the same space.
Beyond such descriptive uses of the image sequence, humorous pic-
torial alphabets, calendars, and dictionaries, 21 or mock instructions22
can also be given in the strip design in this historical context. Thus, the
panels may be related to each other by some shared topic or conceptual
frame rather than a narrative. We must note, furthermore, that the dis-
tinction between a strip sequence and a single image is not always that
clear cut in this body of work. One case in point is the long panorama
image that has the potential to be read both as a scene where many situ
ations occur simultaneously, and as a series of situations, and thus as a
kind of strip without the panel frames. The amount of visual detail in a
typical panoramic broadsheet print, depicting for instance a procession,
a ball, a dinner party, or a pub scene, such as Gillray’s “Union Club”
(1801) or “The Grand Coronation Procession of Napoleon the 1st”
(1804), gives a scene of rich and lively simultaneity. At the same time,
certain features of the image content, such as a road or the portrayal of
movement into a particular direction, may imply that the image consists
of various distinct units of attention set in a certain order.
The sense of an image sequence can also be heightened by the material
form of the publication, such as when the work requires the reader to un-
fold or uncoil the image in order to see it in full.23 For instance, Robert
Cruikshank’s “Going to a Fight” (1819), 24 which depicts in forty-two
scenes with Londoners going to see a boxing match, and the subsequent
match, requires the viewer to uncoil eight broad picture sheets from a
spool and boxwood drum. The numbered scenes of this panorama strip,
showing various styles of clothing, journeying, and boxing enthusiasm,
and that relate some story-like situations such as pickpocketing, are
meant to be viewed from right to left. The order of viewing is further
confirmed by the direction of movement in many of the scenes. In com-
parison, the various ‘three-dimensional’ features in Glasgow Looking
Glass/Northern Looking Glass investigate the issue of perspective
rather than strengthen the sense of sequence. Perhaps the most impres-
sive of these innovations is the multi-perspective cartoon entitled “All
My-Eye”, from the cover of the fourth issue of Glasgow Looking Glass
(23 July 1825), which shows how the same eyes, forehead, and hair may
serve for two different people and facial expressions. Here, a piece of
paper, on which is depicted the features of the lower part of a face, is
superimposed on another face underneath. By flipping the paper, which
is pasted to the page by one margin, the reader can thus perceive two
faces in one image. 25
Many of the larger images in Every Body’s Album, such as “The
Century of Invention. Anno domini 2000”26 (1 February 1834) that de-
picts a scene with futuristic means of transportation, including flying
machines, moveable houses, cars, and buses, function similarly to the
richly detailed panoramic broadsheets. In addition to the global look
on a transportation utopia that this scene suggests, the many speech
balloons and dialogue situations within the image imply that the com-
position consists of various simultaneous situations and distinct units of
attention.
By contrast, in Glasgow Looking Glass/Northern Looking Glass,
panoramic scenes are relatively uncommon. However, what becomes
important in terms of simultaneity in these caricature magazines is the
question of the page as a unit of attention and reading. In some large
full-page or half-page compositional units in Northern Looking Glass
and The Looking Glass, the page layout suggests effects of s imultaneity,
multiple possible perspectives and the global look across the whole
page. In these cases, the significance of the page as a design unit in-
volves both the conception of the page as a whole and its division into
distinct cartoons, sequences, and zones of composition. One of the
most impressive compositions in this regard is “St. Michael of London”
from The Looking Glass (No. 8, 1830), which depicts Bishop Charles
Blomfield, poised on a cushion in front of a dark cloud, reminiscent of
the hovering Christ in church paintings, suspended above and in front
of the little scenes given in three panels. What is particularly remarkable
here is that the bishop’s caricature serves as a link between the surround-
ing contrasted panels and strips (Figure 10.6). The Bishop, exclaiming
“Profane Wretch!!!”, spears a leg of mutton with a lance on a poor man’s
dish in the panel below, thus shattering the dish. On the left, the Bishop’s
cloud extends to a panel that shows a group of soldiers carrying large
dishes from a kitchen, while a military band plays “Go to the Devil and
shake yourself”, and The Duke of Wellington greets them from the bal-
cony of his Apsley House, with the new statue of Achilles (as Wellington)
in the background. To the right of the bishop, in contrast, we see groups
of poor people driven by canons “back to the smoke on a Sunday”, and
two individuals who hold their ground, claiming that “Ve vont go back”.
In “St. Michael of London”, the juxtaposition of elements is more
important and suggestive as a principle of organisation than narrative
sequence. At the same time, the composition of this page experiments
Figure 10.6 Robert Seymour. “St. Michael of London.” The Looking Glass
No. 8. 1 August 1830.
with page layout in ways that are familiar in modern comics albums and
graphic novels. For instance, there are several repeated patterns and situ
ations across the page of this T-shaped composition that suggest ways
of looking beyond the strip sequence. The repeated elements include
the topoi of a military band, a scene at a street door, and The Duke of
Wellington looking at a crowd from a balcony (Apsley House and the
Buckingham Palace). Their patterns suggest multiple ways for looking at
the page beyond the single cartoon or strip and extend the influence of
the “St Michel of London” and the bishop’s figure over the surrounding
frames, dominant in the upper part of the page, to other parts of the page.
Also, the lower part of the page includes a breach with the conven-
tion of the panel frame. This occurs in a cartoon entitled “His Majisty
(sic!) has discharged the German band”, where a German band is seen
marching in a file from left to right. The new King William IV, who had
dismissed his predecessor George IV’s band, stands in the background
with his arm extended beyond the panel frame to the next image, com-
manding the musicians imperiously to “March”. The foremost marcher
to the right, approaching the frame of the panel, states that he must
leave the scene: “Oh, I must leave this festive scene”. The subsequent
cartoon, a parodical reference to Robert Southey’s new edition of his
and Samuel Taylor Coleridge’s poem “The Devil’s Walk”, is unrelated in
subject matter with the previous panel. However, the King’s arm has the
function, from the reader’s viewpoint, of a kind of viewing and reading
instruction that connects the two cartoons and gives a sense of direction
on the page, which is affirmed by the band members’ movement. The
panel frame, then, appears here as a convention that can be broken.
The crossed frame not only affirms the linear order of reading but also
ironically gestures towards the reader about the artificial nature of this
convention.
The page of “St. Michael of London” is one of the many innovations
in this body of work that suggests that the broadsheet caricature mag-
azines of 1825–1835 offered opportunities for creating new types of
relation between the units on the page and the page layout, thus de-
veloping the earlier graphic inventions in the broadsheet format. This
means not just a stronger reliance on the strip format and the image
sequence, but the treatment of the space of the page as one unit of com-
position and attention. Many front covers of Northern Looking Glass
and The Looking Glass work similarly to “St Michel of London” in
that they reflect a certain spatial dynamic and hierarchy in the global
organisation of the page. Typically, the covers feature a more centralized
image, which breaks with the grid-like layout of the unrelated cartoons.
For instance, in the eighth issue of The Looking Glass (1 August 1830),
the images of William IV and Queen Adelaide’s heads in the large cen-
tral column radiate over the frames of the surrounding cartoons. Be-
low them, in an even more prominent and central design, a terrifying
black giant in a general’s uniform intimidates a small dandy demand-
ing: “How dare you appear without a black coat?”. The caption reads:
“General Mourning!!” in reference to George IV’s death in June 1830.
The corners of the black-bordered and diamond-shaped rectangle are
superimposed on the surrounding vignettes, with no particular connec-
tion with the topic. Sometimes the cover page composition may also be
thematically unified, as is the case with the cover of the fifth issue of
Glasgow Looking Glass (6 August 1825) with the title “A State of the
Weather”. Here, all cartoons of the page portray, more or less, the theme
of torrid weather and seasonal conditions. In the central upper part of
the page, the rays of sun strike downwards, thus connecting a group of
five images and suggesting that they have a similar relation to the central
image of the flaming sun. Enclosed in this sun is a tiny figure of Apollo,
who blows the flames with bellows.
Conclusion
The caricature magazines that were published in England and Scotland
from 1825 to 1835, including especially Glasgow Looking Glass/
Northern Looking-Glass, The Looking Glass or, Caricature Annual,
and Every Body’s Album and Caricature Magazine, played a significant
historical role in the gradual development of modern comics. We can
conceive their significance as a kind of waypoint between the broadsheet
prints and the comic strip, the comic magazine, and the comic book from
several perspectives. The artists who were central creators in these pub-
lications, such as William Heath, Charles Jameson Grant, and Robert
Seymour, carried on the graphic experimentation of the Golden Period
of caricature and the broadsheet print, helping to translate the styles
and graphic innovations of the earlier masters, such as James Gillray,
into the format of the printed press and the architecture of a magazine
page. These experiments included, in particular, the sequential form, the
development of the progress story in the multipanel format of the strip
design, the narrative conception of panel relations, and the treatment of
space of the composition both as a design unit and a unit of attention.
The organisational principles of juxtaposition, sequentiality, and
simultaneity, or their combinations, play a significant role in this body of
narrative drawings. Heuristically, we can distinguish two basic forms of
the sequential form with regard to the implied temporal and causal rela-
tions between the images: the episodic sequential form and the narrative
continuity sequence, even if their distinction is not always that clear-cut.
While the sequential form is often (but not always) closely associated
with narrative function, the principle of juxtaposition can equally well
serve narrative function, besides doing something else, such as the illu-
mination of an idea, a theme, a metaphor, or contrasting viewpoints,
arguments, personalities, and behavior.
William Heath, Charles Jameson Grant, Robert Seymour, and their
colleagues explored the relationship between the strip design, the image
sequence, and the space of the composition in the architecture of the
magazine page, continuing the art of caricature of the single-plate street
broadsheet woodcuts in a new periodical format that used litographed
comic images. The new publication format offered new freedoms. Their
satirical magazines, or magazine-like broadsheet prints in the case of
Grant’s Every Body’s Album, showcase various examples of a more
global conception of the page, and the zone of composition, beyond a
single image and a strip. Thus, the page layout and the space of the com-
position become important at two levels simultaneously: as a design unit
and a unit of attention. At the same time, this exploration concerns the
relation between the strip design and the sense of narrative sequence,
and the distinction that we today may take granted between a cartoon,
that is, a single-image drawing, and a sequential narrative comic. The
idea of a panel set in relation to surrounding panels, in a sequence or
otherwise, is already a convention in this body of work and one that can
be consciously broken—even if there are also strong indications that the
sequential arrangement of narrative drawings was not that familiar with
all readers. All in all, the new effects of sequentiality in these works, of-
ten combined with the logic of juxtaposition or simultaneity, met the is-
sue of visual narrativity in a novel way, thus also implying the emergence
of new types of the reader’s ‘diegetic absorption’ through an image series.
Finally, it must be emphasised that in my choice of examples, I have fo-
cussed on picture stories and their narrative organisation. There are also
various non-narrative forms for organising series or groups of images
and using the strip design in this body of work. They comprise descrip-
tion and exposition, as in illustrations to a text (poems, stories, anec-
dotes), pictorial encyclopedia or dictionary, inventory or the cataloguing
of items, objects, and so on—based on the principles of accumulation or
aggregation, instead of juxtaposition or sequentiality—puns and jokes,
and political satire without a sense of a story. We must thus note that the
variety of forms of graphic art in this body of work go vastly beyond the
narrative sequence and the need to follow a story from one panel to an-
other. While most comics since then have been narratives, comics have
never been just narratives.
Notes
1 Such as the title page, cover artwork, copyright pages, foreword, epigraphs,
epilogue, or publishers’ blurbs.
2 Lefèvre specifies that a publication format is not only defined by its material
aspects (size, paper quality, etc.), but also by its temporal (daily, weekly,
monthly, one shot) and editorial parameters (length of an episode, regula-
tions regarding content or the public) (2013, 267), as well as by thematic
constraints and an aesthetic system (2009, 227; 2010, 88, 91).
3 See also Fox (1988, 76).
4 Including albums, such as Ally Sloper: A Moral Lesson (1873), and comic
magazines. Often referred to as the first comic magazine, it is debatable,
however, to what extent the weekly Ally Sloper’s Half-Holiday actually was
a comic magazine in its early years. The emphasis on single-image cartoons
and written articles made the weekly look much like any other satirical
periodical of this time. That said, however, we can also follow the develop-
ment of the continuing character and the sequential story in this publication.
The longest comics in the first year of its publication in 1884 include around
ten panels, while the story entitled “Toddleboy goes to see the university
crews practice” from 27 March 1886 already has twenty-one panels. On the
history of early British publication venues for comics, with a specific em-
phasis on children’s comics, see also Gifford (2004, 362–371). See Beringer
on the development of the multipanel picture story in American illustrated
magazines from the 1850s on, including Yankee Notions (1852–1875), Nick
Nax (1856–1875), and the Comic Monthly (1859–1891), and how these de-
velopments were defined “in large part by a taste for experimenting with
new and different ways of depicting narrative experience” (2015, 457).
5 Subtitled “One hundred laughs for one halfpenny”, the first issue was pub-
lished on 17 May 1890.
6 The Barnhart Dictionary of Etymology (1988) claims that the meaning of
comics as a “comic strip” is first recorded as late as 1910 in H.G. Wells’
comic novel The History of Mr. Polly where there is a reference to cheap
boys’ comics of today.
7 My distinction between the strip design and the sequential form follows
Kunzle’s corresponding definitions (quoted here) as well as Groensteen’s
distinction between a string (or a series) of panels and a sequence. For
Groensteen, a sequence is different from the mere juxtaposition of disparate
images or a series that is based on visual, iconic, or semantic correspondences
between the images, in that in a sequence the “syntagmatic linkage” in the
succession of images is “determined by a narrative project” (2014, 176).
8 As in James Gillray’s drawing “Nature display’d, shewing the effect of the
change of the seasons on the ladies garden” (1797), where the four headless
female figures need to be viewed from four different angles.
9 Compare with Beringer’s emphasis on the diversity of ways for organising
the narrative flow in comics published in the 1850s American humour maga
zines comics. In this context, Beringer argues, it was more likely that the
panel transitions were organised in ways that do not follow a tight sequence
of actions, but, for instance, through movement between various scenes, or
the change of perspective (2015, 457).
10 See, for instance, Smolderen (2014a, 137–148) and Lefèvre (2006).
11 The first issue of Glasgow Looking Glass appeared on 11 June 1825. It was
renamed Northern Looking Glass in the issue VI (18 August 1825) and
Northern Looking Glass, or Litho’s Album in the next issue (3 September
1825). In the early stages of his career, Charles Jameson Grant collaborated
with William Heath’s brother, Henry, and also produced a small number of
prints for William Heath and Robert Seymour’s publisher, Thomas McLean
(see Pound 1998).
12 Beyond the satirical and comic images, the issues of 1–4 of Glasgow Looking
Glass feature extracts from Horace Smith’s and Thomas Campbell’s poems.
The last page of the second issue of Northern Looking Glass (VII) comprises
quotes from newspapers and periodicals. This becomes a permanent feature
in the subsequent issues. The first issue of the short-lived “new series” of
Northern Looking Glass (May 1826) includes an editorial introduction and
texts relating to some of the images.
13 Kunzle refers to the antithetical method of narration in picture stories before
the modern comic strip: “To narrate is, first of all, to polarize a sequence of
events into Before and After, Then and Now, Cause and Result—and Crime
and Punishment” (1973, 3). Thus, we may add, the logic of this method is
either temporal or causal.
14 The subtitle explains that this takes place in France and, further, that “The
following is the manner in which the King of France passes his time at
St. Cloud” (Château de Saint-Cloud).
15 The contrast is further affirmed by the motto taken from Hamlet: “Look
here upon this picture and on this, The counterfeit presentment of two
brothers”.
16 “Hope” can be further contrasted with another Gillray caricature, “ Despair”,
that was published at the same time and that shows the opposition leader
Richard Bateman Robson speaking in the House of Commons and stating
that “We’re all ruinated, Sir!”.
17 The title is reminiscent of Heath’s earlier illustrated narrative poem, The
Life of a Soldier; A Narrative and Descriptive Poem (1823), but the images
in the magazine are much simpler and more caricature-like than in the en-
gravings of the illustrated book.
18 Another instance of seemingly wrong order in Northern Looking Glass
(No. X) involves a two-panel story entitled “Cambridge Sporting Intelli-
gence”. In this incident, a Dr Syntax-like character, who is carried on an-
other person’s shoulders in a piggyback race, first falls down and is then
again shown racing in the piggyback position. See also Witek, who points
out how the use of panel numbering and directional arrows in the early
history of the medium suggests how “comicness” can be conceptualised as
“a historically contingent and evolving set of reading protocols that are ap-
plied to texts” (2009, 149).
19 Another version of this are picture stories where different phases in a char-
acter’s or a group’s development are shown in a long strip-like space without
separating frames. For instance, in George Woodward and F. G. Byron’s
“The Clerical Exercise” (1791).
20 It has been claimed that the miniature version of “Harlot’s Progress” that
was published in June 1828 in the English weekly Bell’s Life in London, and
Sporting Chronicle (1822–1886) was the first newspaper strip (Kunzle 1973,
162; Gravett 1998). The claim can be contested in light of the strips featured
in Glasgow/Northern Looking-Glass in 1825–1826. That said, Bell’s Life
had an important role in the early history of comics by running a weekly fea-
ture called “The Gallery of Comicalities”, including a series of caricatures
and illustrated jokes, beginning in September 1827. These caricatures were
then collected in the broadsheet newspaper The Gallery of Comicalities, and
published every 2 years between 1832 and 1841.
21 Such as Richard Newton’s “Clerical Alphabet” (1795) or “The Pictorial Dic-
tionary” in Northern Looking Glass (No. XII).
22 Such as “Six of the most approved methods of appearing ridiculous on the
ice” or “symptoms of jolting” (in stage coaches) in George Woodward’s pa-
rodic travel book Eccentric Excursions (1796).
23 For instance, the folded political cartoons included in The Scourge and
George Cruikshank’s The Comic Almanack (1844–1853). The readers’ ac-
tivation in this way may, however, focus on perspectival effects rather than
the impression of sequentiality.
24 Subtitled “Illustrating the Sporting World in all its variety of Style and
Costume along the Road from Hyde Park Corner to Moulsey Hurst”,
accompanied by Pierce Egan’s written description, and sized 14 ft by
2.4 inches, this work is based on a true sporting event that took place in
April 1817.
25 In another issue of Northern Looking Glass (No. VII, 3 September 1825),
a cavalry officer’s hat extends beyond the panel in a separate piece of folded
paper, similarly pasted on the page.
26 Subtitled “Or the March of Aerostation, Steam, Rail Roads, Moveable
Houses, & Perpetual Motion”.
Afterword
It is in the nature of narratology to seek what is most universal,

conventional, and general about narratives, and attempt to describe and
analyse these features as effectively as possible. Yet, the relationship bet
ween what is general and what is unique, variable, or contra-standard
in storytelling has also been one of the defining tensions throughout the
history of this field. Thus, the impetus towards the generalisable has
been regularly counterbalanced by paying attention to the ways in which
individual artists and works of art explore, question, and modify com
mon practices and invent new forms of storytelling. When narratology
meets the unexpected, and the singular, the theory has to be adjusted—if
not right away, then in time. This process of adjusting, then, has become
an important outcome of the research.
In today’s narratology, the question about the universal and the unique
in narratives also extends to the relation between narratives and their
media. As narratology has reached outside the traditional object domain
of text-based literary story, it has increasingly started to pay attention
to the ways in which narrative transmission is media-related and how
the qualities of the medium affect the way in which stories can be told.
In this book, I have attempted to take the medium-specific argument as
far as possible, but always keeping in mind the shared qualities of nar
ratives across many different systems of communication and expression,
especially with literature and cinema. The principle that I have followed
is that for narratology to have general relevance in the field of Comics
Studies, it needs to adopt a self-critical and flexible attitude towards
theoretical propositions in the light of actual poetics and artistic choices
in this form of art. And, this is how I see the relevance of the examples
that I have discussed in this study: they illustrate the main narrative con
ventions in the modern Western comics, reflecting diversity in historical
context, markets, language, genre, and publication format, while many
of them also illustrate how individual works can successfully modify the
existing narrative traditions and forms to establish their rightful place in
the canons of literary expression and visual art.
I hope to have demonstrated that I do not conceive narratology as
a solution to all interpretive challenges that may be faced in studying
278 Afterword
comics. Narratological concepts and approaches are a valuable source
of insight in research, where they can stretch our awareness of the nar
rative qualities of this art and its historically contingent forms, thus
preparing the way for a more holistic interpretation. A more compre
hensive historical study of comics, for instance, must engage with many
other dimensions of comics beyond their narrative form, conventions,
and strategies, including the question of genre (generic relations, im
plicatures, and expectations), intertextual and intermedial relations,
the context of the making and reading of comics, artistic and stylistic
movements, authorial intention and readerly expectations, and relevant
aspects of the culture and materiality of comics, such as publication for
mats and distribution, fandom, criticism, and marketing. The range of
potentially relevant contexts for studying comics in the humanities and
social sciences is, in principle then, infinite.
One last matter that I wish to draw to the reader’s attention is the ap
preciation of comics as a form of reading. It has sometimes been claimed
that narratology seeks to unravel the mysteries of storytelling, i.e. ex
pose the bare essence of the narrative form, and that for this reason its
analytical procedures could result in killing the joy of reading, viewing,
or listening to narratives. Personally, I do not believe that narratology
has any such a ‘demystifying’ objective or that narratological analysis
needs to be detrimental to the pleasures of reading. On the contrary, my
experience in teaching narratology and doing research in this field, and
having discussed the point of narrative analysis with cartoonists and
students, is that narrative theory can complement personal as well as ex
pressive and artistic responses to the art of comics. Narrative theory may
even have, when used judiciously, the potential to intensify such plea
sures and responses. On a more personal note, this book has also been
inspired by my own personal experiences of reading comics, marvelling
at them, and enjoying them throughout my life. Having first learned to
read by way of comic books, I have posed questions to myself such as:
‘Why are comics so interesting and appealing as narratives?’ and ‘What
is particularly effective about telling stories in this form of art and liter
ature?’ The process of writing The Narratology of Comic Art has been
a meaningful way to explore such questions.
Bibliography
Abbott, Lawrence L. 1986. “Comic Art: Characteristics of a Narrative Medium.”

Journal of Popular Culture 19: 155–76.
Abbott, Porter H. 2005. “Closure.” In Routledge Encyclopedia of Narrative
Theory, 65–66. Eds. David Herman, Manfred Jahn and Marie-Laure Ryan.
London and New York: Routledge.
———. 2008. The Cambridge Introduction to Narrative. 2nd ed. Cambridge:
Cambridge University Press.
Altman, Rick. 2008. A Theory of Narrative. New York: Columbia University
Press.
Ault, Donald. 2004. “Preludium: Crumb, Barks, and Noomin: Re-considering
the Aesthetics of Underground Comics.” ImageTexT 29. http://www.engl…s/
v1_2/intro.shtml.
Badman, Derik A. 2010. “Talking, Thinking, and Seeing in Pictures: Narration,
Focalization, and Ocularization in Comics Narratives.” International Jour-
nal of Comic Art 12.2/3: 91–111.
Baetens, Jan. 1989. Hergé écrivain. Bruxelles: Labor.
———. 1998. Formes et politique de la bande dessinée. Leuven: Peeters Vrin.
———. 2001. “Revealing Traces: A New Theory of Graphic Enunciation.” In
The Language of Comics: Word and Image, 145–55. Eds. Robin Varnum and
Christina T. Gibbons. Jackson: University Press of Mississippi.
———. 2004. “La main parlante.” Image & Narrative V.1. http://www.imageand
narrative.be/inarchive/performance/baetens_main.htm.
———. 2008. “Graphic Novels: Literature without Text?” English Language
Notes 46.2: 77–88.
———. 2013. “Uncaging and Reframing Martin Vaugh-James’s the Cage.” In
Drawing from Life: Memory and Subjectivity in Comic Art, 67–85. Ed. Jane
Tolmie. Jackson: University Press of Mississippi.
Baetens, Jan and Hugo Frey. 2015. The Graphic Novel. An Introduction.
Cambridge and New York: Cambridge University Press.
Baetens, Jan and Pascal Lefèvre. 1993. Pour une lecture moderne de la bande
dessinée. Bruxelles: Centre belge de la bande dessinée.
Bal, Mieke. 1981. “Notes on Narrative Embedding.” Poetics Today 2.2: 41–59.
———. 1990. “The Point of Narratology.” Poetics Today 11.4: 727–53.
———. 1991. On Story-Telling. Essays in Narratology. Ed. David Jobling.
Sonoma: Polebridge Press.
———. 1997. Narratology: Introduction to the Theory of Narrative. 2nd ed.
Toronto: University of Toronto Press.
280 Bibliography
Barker, Martin. 1989. Comics: Ideology, Power, and the Critics. Manchester:
Manchester University Press.
Bateman, John A., Francisco O. D. Veloso, Janina Wildfeuer, Felix HiuLaam
Cheung and Nancy Songdan Guo. 2016. “An Open Multilevel Classification
Scheme for the Visual Layout of Comics and Graphic Novels: Motivation
and Design.” Digital Scholarship in the Humanities 1–35. doi:10.1093/llc/
fqw024.
Bavelas, Janet B. and Nicole Chovil. 2006. “Hand Gestures and Facial Displays
as Part of Language Use in Face-to-Face Dialogue.” In The Sage Handbook
of Nonverbal Communication, 97–115. Eds. Valerie Manusov and Miles
L. Patterson. Thousand Oaks, CA, London and New Delhi: Sage Publications.
Beaty, Bart. 1999. “The Search for Comics Exceptionalism.” Comics Journal
211: 67–72.
———. 2012. Comics versus Art. Toronto: Toronto University Press.
Benveniste, Émile. 1970. “L’appareil formel de l’énonciation.” Langages 5.17:
12–18.
Beringer, Alex. 2015. “Transatlantic Picture Stories: Experiments in the
A ntebellum American Comic Strip.” American Literature 87.3: 455–88.
Blackwell, Mark. 2012. “General Introduction.” In British It-Narratives,
1750–1830. Volume I (Money), vii–xxviii. Ed. Liz Bellamy. London: Pickering &
Chatto.
Boillat, Alain (ed.). 2010. Les Cases à l’écran. Bande dessinée et cinéma en dia-
logue. Genève: Éditions Georg/L’Equinoxe.
Bongco, Mila. 2000. Reading Comics. Language, Culture, and the Concept of
the Superhero in Comic Books. New York: Garland.
Booth, Wayne C. 1983. The Rhetoric of Fiction. 2nd ed. Chicago and London:
University of Chicago Press.
Bordwell, David. 1985. Narration in the Fiction Film. Madison: University of
Wisconsin Press.
———. 2008. Poetics of Cinema. New York: Routledge.
Bordwell, David and Kristin Thompson. 1986. Film Art: An Introduction. 2nd
ed. New York: Alfred A. Knopf.
———. 2008. Film Art. An Introduction. 8th ed. Boston: McGraw-Hill.
Bortolussi, Marisa and Peter Dixon. 2003. Psychonarratology: Foundations for
the Empirical Study of Literary Response. Cambridge: Cambridge University
Press.
Bramlett, Frank. 2012. “Linguistic Codes and Character Identity in Afro
S amurai.” In Linguistics and the Study of Comics, 183–209. Ed. Frank
Bramlett. London: Palgrave Macmillan.
Branigan, Edward. 1984. Point of View in the Cinema: A Theory of Narration
and Subjectivity in Classical Film. Berlin: Mouton.
———. 1992. Narrative Comprehension and Film. London: Routledge.
Bredehoft, Thomas A. 2006. “Comics Architecture, Multidimensionality, and
Time: Chris Ware’s Jimmy Corrigan: The Smartest Kid on Earth.” Modern
Fiction Studies 52.4: 869–90.
———. 2011. “Style, Voice, and Authorship in Harvey Pekar’s (Auto) (Bio)
Graphical Comics.” College Literature 38.3: 97–110.
Bremond, Claude. 1968. “Pour un gestuaire des bandes dessinées.” Langages
3.10: 94–100.
Bibliography 281
Bridgeman, Teresa. 2004. “Keeping an Eye on Things: Attention, Tracking and
Coherence-Building.” Belphégor: Littérature Populaire et Culture Média-
tique 4.1. http://etc.dal.ca/belphegor/vol4_no1/articles/04_01_Bridge_keep_
en_cont.html.
Brooke-Rose, Christine. 1981. A Rhetoric of the Unreal. Studies in the N arrative
and Structure, Especially of the Fantastic. Cambridge: Cambridge University
Press.
Bukatman, Scott. 2006. “Comics and the Critique of Chronophotography, or
‘He Never Knew When It Was Coming!’” Animation 1: 83–103.
———. 2014. “Sculpture, Stasis, the Comics, and Hellboy.” Critical Inquiry
40.3: 104–17.
Butte, George. 2008. “Suture and the Narration of Subjectivity in Film.” Poet-
ics Today 29.2: 277–308.
Carrier, David. 2000. The Aesthetics of Comics. University Park: The P
ennsylvania
State University Press.
Carroll, Noël. 2008. The Philosophy of Motion Pictures. Malden, MA and
Oxford: Blackwell Publishing.
Chapman, Raymond. 1984. The Treatment of Sounds in Language and Litera-
ture. Oxford: Blackwell.
Chatman, Seymour. 1978. Story and Discourse: Narrative Structure in Fiction
and Film. Ithaca, NY: Cornell University Press.
———. 1990. Coming to Terms: The Rhetoric of Narrative in Fiction and Film.
Ithaca, NY: Cornell University Press.
Chavanne, Renaud. 2010. Composition de la Bande dessinée. Montrouge,
France: Editions PLG/Collection « Mémoire Vive ».
Christiansen, Hans-Christian. 2000. “Comics and Film: A Narrative Per
spective.” In Comics & Culture: Analytical and Theoretical Approaches to
Comics, 107–22. Eds. Anne Magnussen and Hans-Christian Christiansen.
Copenhagen: University of Copenhagen/Museum Tusculanum Press.
Chute, Hillary. 2010. Graphic Women: Life Narrative and Contemporary
Comics. New York: Columbia University Press.
Chute, Hillary and Marianne DeKoven. 2006. “Introduction: Graphic Narra
tive.” Modern Fiction Studies 52.4: 767–82.
Cohan, Steven and Linda M. Shires. 1988. Telling Stories: A Theoretical Ana
lysis of Narrative Fiction. New York: Routledge.
Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Con-
sciousness in Fiction. Princeton, NJ: Princeton University Press.
———. 1999. The Distinction of Fiction. Baltimore, MD: The Johns Hopkins
University Press.
Cohn, Jesse. 2007. “Translator’s Comments on Benoît Peeters, ‘Four Concep
tions of the Page.’” ImageText 3.3. http://www.english.ufl.edu/imagetext/
archives/v33/cohn/.
———. 2009. “Mise-en-Page: A Vocabulary for Page Layouts.” In Teaching the
Graphic Novel, 44–57. Ed. Stephen E. Tabachnick. New York: The Modern
Language Association of America.
Cohn, Neil. 2007. “A Visual Lexicon.” The Public Journal of Semiotics 1.1:
35–56.
———. 2010. “The Limits of Time and Transitions: Challenges to Theories of
Sequential Image Comprehension.” Studies in Comics 1.1: 127–47.
282 Bibliography
———. 2013. “Navigating Comics: An Empirical and Theoretical Approach to
Strategies of Reading Comic Page Layouts.” Frontiers in Psychology 4.186:
1–15.
———. 2014a. “The Architecture of Visual Narrative Comprehension: The In
teraction of Narrative Structure and Page Layout in Understanding Comics.”
Frontiers in Psychology 5.680: 1–9.
———. 2014b. “Building a Better ‘Comic Theory’: Shortcomings of Theoretical
Research on Comics and How to Overcome Them.” Studies in Comics 5.1:
57–75.
Cortsen, Rikke P. 2012. Comics as Assemblage: How Spatio-temporality in
Comics Is Constructed. PhD dissertation. University of Copenhagen.
Covey, Suzanne. 2006. “Beyond the Balloon: Sound Effects and Background
Text in Lynn Johnston’s For Better or For Worse.” ImageTexT: Interdisciplin-
ary Comics Studies 2.2. http://www.engl…/covey/index.shtml.
Cutting, James E. 2014. “Event Segmentation and Seven Types of Narrative
Discontinuity in Popular Movies.” Acta Psychologica 149: 69–77.
Delany, Samuel R. 1999. “Politics of Paraliterary Criticism.” In Shorter Views.
Queer Thoughts & The Politics of the Paraliterary, 218–70. Ed. Samuel
R. Delany. Hanover and London: Wesleyan University Press.
Deleyto, Celestino. 1996. “Focalisation in Film Narrative.” In Narratology:
An Introduction, 217–33. Eds. Susana Onega and José Angel Garcia Landa.
London: Longman.
Donaldson-Evans, Mary. 2009. Madame Bovary at the Movies: Adaptation,
Ideology, Context. Amsterdam and New York: Éditions Rodopi.
Duncan, Randy and Matthew J. Smith. 2009. The Power of Comics. History,
Form, and Culture. New York: Bloomsbury.
Dürrenmatt, Jacques. 2013. Bande dessinée et littérature. Paris: Classiques
Garnier.
Eisner, Will. 1985. Comics & Sequential Art. Tamarac, FL: Poorhouse Press.
———. 1996. Graphic Storytelling and Visual Narrative. Tamarac, FL:
Poorhouse Press.
Feiffer, Jules. 1965. Great Comic Book Heroes. London: Dial Press.
Fein, Ofer and Asa Kasher. 1996. “How to Do Things with Words and Gestures
in Comics.” Journal of Pragmatics 26: 793–808.
Fischer, Craig and Charles Hatfield. 2011. “Teeth, Sticks, and Bricks: Calligraphy,
Graphic Focalization, and Narrative Braiding in Eddie Campbell’s Alec.”
SubStance 40.1: 70–93.
Fishelov, David. 1990. “Types of Characters, Characteristics of Types.” Style
24: 422–39.
Flaubert, Gustave. 1946. Madame Bovary. Provincial Manners. Trans. Eleanor
Marx Aveling. New York: The Modern Library.
———. 1993. Madame Bovary. Paris: Booking International.
Fludernik, Monika. 1996. Towards a ‘Natural’ Narratology. London and
New York: Routledge.
———. 2001. “New Wine in Old Bottles? Voice, Focalization, and New
Writing.” New Literary History 32: 619–38.
———. 2003. “Scene Shift, Metalepsis, and the Metaleptic Mode.” Style 37.4:
382–400.
Bibliography 283
———. 2009. An Introduction to Narratology. Trans. Patricia Häusler
Greenfield and Monika Fludernik. London: Routledge.
Forceville, Charles. 2002. “The Conspiracy in The Comfort of Strangers:
Narration in the Novel and the Film.” Language and Literature 11.2: 119–35.
———. 2005. “Visual Representations of the Idealized Cognitive Model of
A nger in the Asterix Album La Zizanie.” Journal of Pragmatics 37: 69–88.
———. 2011. “Pictorial Runes in Tintin and the Picaros.” Journal of Pragmat-
ics 43: 875–90.
———. 2013. “Creative Visual Variation in Comics Balloons.” In Creativity
and the Agile Mind: A Multi-disciplinary Study of a Multi-faceted Pheno
menon, 251–73. Eds. Tony Veale, Kurt Feyaerts and Charles Forceville.
B erlin: Mouton de Gruyter.
Forceville, Charles, Elisabeth E. Refaie and Gert Meesters. 2014. “Stylistics
and Comics.” In The Routledge Handbook of Stylistics, 485–99. Ed. Michael
Burke. London: Routledge.
Forceville, Charles, Tony Veale and Kurt Feyaerts. 2010. “Balloonics: The Visu
als of Balloons in Comics.” In The Rise and Reason of Comics and Graphic
Literature. Critical Essays on the Form, 56–73. Eds. Joyce Goggin and Dan
Hassler-Forest. Jefferson, NC and London: McFarland Press.
Forster, Edward Morgan. 1927/1953. Aspects of the Novel. London: Edward
Arnold & Co.
Fowler, Roger. 1977. Linguistics and the Novel. London: Methuen.
Fox, Celina. 1988. Graphic Journalism in England during the 1830s and 1840s.
New York and London: Garland.
Fresnault-Deruelle, Pierre, 1976. “Du linéaire au tabulaire.” Communications
24: 7–23.
———. 1977. Récits et discours par la bande. Essais sur les comics. Paris:
Hachette.
Gardner, Jared. 2011. “Storylines.” SubStance 40.1: 53–69.
———. 2012. Projections. Comics and the History of Twenty-First-Century
Storytelling. Stanford, CA: Stanford University Press.
———. 2013. “A History of the Narrative Comic Strip.” In From Comic
Strips to Graphic Novels: Contributions to the Theory and History of
Graphic Narrative, 241–53. Eds. Daniel Stein and Jan-Noël Thon. Berlin and
New York: Walter de Gruyter.
Gardner, Jared and David Herman. 2011. “Graphic Narratives and Narrative
Theory. Introduction.” Substance 40.1: 3–13.
Gaudreault, André. 1987. “Narration and Monstration in the Cinema.” Journal
of Film and Video 39.2: 29–36.
———. 1989. Du littéraire au filmique. Système du récit. Paris: Méridiens
Klincksieck.
———. 2009. From Plato to Lumière: Narration and Monstration in Literature
and Cinema. Trans. Timothy Barnard. Toronto: University of Toronto Press.
Gaudreault, André and François Jost. 1990. Le Récit Cinematographique.
Paris: Éditions Nathan, Collection Nathan-Université.
———. 1999. “Enunciation and Narration.” In A Companion to Film Theory,
45–63. Eds. Toby Miller and Robert Stam. Malden, MA: Blackwell Publish
ing Ltd.
284 Bibliography
Gaut, Berys. 2004. “The Philosophy of the Movies: Cinematic Narration.” In
The Blackwell Guide to Aesthetics, 230–53. Ed. Peter Kivy. Malden, MA:
Blackwell Publishing.
Genette, Gérard. 1980. Narrative Discourse: An Essay in Method. Trans. Jane
E. Levin. Ithaca, NY: Cornell University Press.
———. 1988. Narrative Discourse Revisited. Trans. Jane E. Lewin. Ithaca, NY:
Cornell University Press.
Gibson, James J. 1979. The Ecological Approach to Visual Perception. Boston:
Houghton Mifflin.
Gifford, Denis. 1976. Victorian Comics. London: George Allen & Unwin Ltd.
———. 2004. “Comics, Dime Novels, Pulps and Penny Dreadfuls.” In Inter
national Companion Encyclopedia of Children’s Literature, 362–84. 2nd ed.
Ed. Peter Hunt. London and New York: Routledge.
Gitelman, Lisa. 2006. Always Already New. Media, History, and the Data of
Culture. London: The MIT Press.
Gombrich, Ernst M. 1960. Art and Illusion. A Study in the Psychology of Pic-
torial Representation. London: Phaidon Press.
Gravett, Paul. 1998. “The Cartoonist’s Progress: The Inventors of Comics in
Great Britain.” In Forging a New Medium: The Comic Strip in the Nineteenth
Century, 81–103. Eds. Pascal Lefèvre and Charles Dierick. Bruxelles: VUB
University Press.
———. 2013. Comics Art. London: Tate Publishing.
Groensteen, Thierry. 1988/2014. “Narration as Supplement: An Archaeology of
the Infra-narrative Foundations of Comics.” In The French Comics Theory
Reader, 163–81. Eds. Ann Miller and Bart Beaty. Leuven: Leuven University
Press.
———. 1997. “Histoire de la bande dessinée muette.” Première partie. Neuvième
Art 2: 60–75.
———. 1998. “Histoire de la bande dessinée muette.” Deuxième partie. N euvième
Art 3: 92–105.
———. 1999. Système de la bande dessinée. Paris: PUF.
———. 2007. The System of Comics. Trans. Bart Beaty and Nick Nguyen.
Jackson: University Press of Mississippi.
———. 2011. Bande dessinée et narration (Système de la bande dessinée 2).
Paris: PUF.
———. 2013. Comics and Narration. Trans. Ann Miller. Jackson: University
Press of Mississippi.
Guaïtella, Isabelle. 2003. “Au rythme des images: multimodalité et multi
linéarité de la bande dessinée.” Semiotica 146.1/4: 519–26.
Gunning, Tom. 2009. “Preface.” In From Plato to Lumière: Narration and
Monstration in Literature and Cinema, xvii–xxv. Ed. André Gaudreault.
Toronto: University of Toronto Press.
———. 2014. “The Art of Succession: Reading, Writing, and Watching Com
ics.” Critical Inquiry 40.3: 36–51.
Hague, Ian. 2014. Comics and the Senses. New York and London: Routledge.
Harvey, Robert C. 1994. The Art of the Funnies: An Aesthetic History. Jackson:
University Press of Mississippi.
———. 1996. The Art of the Comic Book: An Aesthetic History. Jackson:
University Press of Mississippi.
Bibliography 285
———. 2005. “Describing and Discarding ‘Comics’ as an Impotent Act of
Philosophical Right.” In Comics as Philosophy, 14–26. Ed. Jeff McLaughlin.
Hatfield, Charles. 2005. Alternative Comics: An Emerging Literature. Jackson:
University of Mississippi Press.
———. 2009. “Defining Comics in the Classroom; or, the Pros and Cons
of Unfixability.” In Teaching the Graphic Novel, 19–27. Ed. Stephen
E. Tabachnick. New York: The Modern Language Association of America.
———. 2010. “Indiscipline, or, the Condition of Comics Studies.” Transatlan-
tica 1: 2–15. http://transatlantica.revues.org/4933.
Hearn, Lafcadio. 1904. Kwaidan: Stories and Studies of Strange Things.
Boston, MA: Houghton, Mifflin and Co.
Herman, David. 1999. “Introduction.” In Narratologies: New Perspectives on
Narrative Analysis, 1–30. Ed. David Herman. Columbus: Ohio State Press.
———. 2002. Story Logic: Problems and Possibilities of Narrative. Lincoln:
University of Nebraska Press.
———. 2009a. Basic Elements of Narrative. Chichester: Wiley-Blackwell and
John Wiley & Sons Ltd.
———. 2009b. “Beyond Voice and Vision: Cognitive Grammar and Focali
sation Theory.” In Point of View, Perspective, and Focalisation: Modeling
Mediation in Narrative, 119–42. Eds. Peter Hühn, Wolf Schmid and Jörg
Schönert. Berlin: Walter de Gruyter.
———. 2011. “Storyworld/Umwelt: Nonhuman Experiences in Graphic Narra
tives.” SubStance 40.1: 156–81.
———. 2013. Storytelling and the Sciences of Mind. Cambridge, MA and
London: MIT Press.
Herman, David and Jared Gardner. 2011. “Graphic Narratives and Narrative
Theory: Introduction.” Substance 40.1: 3–13.
Horrocks, Dylan. 2001. “Inventing Comics: Scott McCloud’s Definition of
Comics.” Comics Journal 234. http://www.hicksville.co.nz/Inventing%20
Comics.htm.
Horstkotte, Silke and Nancy Pedri. 2011. “Focalization in Graphic Narrative.”
Narrative 19.3: 330–57.
Jacobs, Dale. 2007. “Beyond Visual Rhetoric: Multimodal Rhetoric and News
paper Comic Strips.” International Journal of Comic Art 9.1: 502–14.
Jahn, Manfred. 1999. “More Aspects of Focalization. Refinements and Appli
cations.” GRAAT: Revue des Groupes de Recherches Anglo-Américaines de
L’Université François Rabelais de Tours 21: 85–110.
———. 2003. “A Guide to Narratological Film Analysis.” In Poems, Plays,
and Prose: A Guide to the Theory of Literary Genres. Cologne: University of
Cologne. http://www.uni-koeln.de/~ame02/pppf.htm.
Jannidis, Fotis. 2015. “Character”. In The Living Handbook of Narratology. Eds.
Peter Hühn, John Pier, Wolf Schmid and Jörg Schönert. H amburg: Hamburg
University. http://wikis.sub.uni-hamburg.de/lhn/index.php/Character.
Jenkins, Henry. 2006. Convergence Culture: Where Old and New Media
C ollide. New York: New York University Press.
Jesch, Tatjana and Malte Stein. 2009. “Perspectivization and Focalization:
Two Concepts—One Meaning? An Attempt at Conceptual Differentiation.”
In Point of View, Perspective, and Focalization. Modeling Mediation in
286 Bibliography
Narrative, 59–77. Eds. Peter Hühn, Wolf Schmid and Jörg Schönert. Berlin:
Walter de Gruyter.
Jost, François. 1983. “Narration(s): en deçà et au-delà.” Communications 38:
192–212.
———. 1987. L’Œil-Caméra. Entre film et roman. Lyon: Presses Universitaires
de Lyon.
———. 2004. “The Look: From Film to Novel: An Essay in Comparative
Narratology.” In A Companion to Literature and Film, 71–80. Eds. Robert
Stam and Alessandra Raengo. Malden, MA: Blackwell.
Kannenberg, Gene Jr. 2001. “The Comics of Chris Ware: Text, Image, and
Visual Narrative Strategies.” In The Language of Comics: Word and Image,
174–97. Eds. Robin Varnum and Christina T. Gibbons. Jackson: University
Kennedy, John M. 1982. “Metaphor in Pictures.” Perception 11: 589–605.
Kindt, Tom. 2009. “Narratological Expansionism and Its Discontents.” In Nar-
ratology in the Age of Cross-disciplinary Narrative Research, 35–47. Eds.
Sandra Heinen and Roy Sommer. Berlin and New York: Walter de Gruyter.
Kindt, Tom and Hans-Harald Müller. 2003. “Narrative Theory and/or/as The
ory of Interpretation.” In What Is Narratology? Questions and Answers Re-
garding the Status of a Theory, 205–19. Eds. Tom Kindt and Hans-Harald
Müller. Berlin and New York: Walter de Gruyter.
———. 2006. The Implied Author: Concept and Controversy. Berlin and
New York: Walter de Gruyter.
Klauk, Tobias and Köppe, Tilmann. 2015. “Telling vs. Showing.” In The Living
Handbook of Narratology. Eds. Peter Hühn, John Pier, Wolf Schmid and Jörg
Schönert. Hamburg: Hamburg University. http://www.lhn.uni-hamburg.de/
article/telling-vs-showing.
Kolp, Manuel. 1992. Le Langage cinématographique en bande dessinée.
Brussels: Éditions de l’université de Bruxelles.
Kozloff, Sarah. 2000. Overhearing Film Dialogue. Berkeley: University of
California Press.
Kress, Gunther. 2010. Multimodality: A Social Semiotic Approach to Contem-
porary Communication. London: Routledge.
Kress, Gunther and Theo van Leeuwen. 1996/2006. Reading Images: The
Grammar of Visual Design. 2nd ed. London: Routledge.
Kruger, Jan-Louis. 2012. “Making Meaning in AVT: Eye Tracking and Viewer
Construction of Narrative.” Perspectives: Studies in Translatology 20.1:
67–86.
Kukkonen, Karin. 2013a. Contemporary Comics Storytelling. Lincoln and
London: University of Nebraska Press.
———. 2013b. Comics and Graphic Novels. Malden, MA: Wiley Blackwell.
Kunzle, David. 1973. History of the Comic Strip Volume I: The Early Comic
Strip. Narrative Strips and Picture Stories in the European Broadsheet from
c.1450 to 1825. Berkeley, Los Angeles and London: University of California
Press.
———. 1990. The History of the Comic Strip Volume II: The Nineteenth Cen-
tury. Berkeley, Los Angeles and Oxford: University of California Press.
———. 2001. “The Voices of Silence: Willette, Steinlen and the Introduction
of the Silent Strip in the Chat Noir, with a German Coda.” In The Language
Bibliography 287
of Comics: Word and Image, 3–18. Eds. Robin Varnum and Christina
T. Gibbons. Jackson: University Press of Mississippi.
Labio, Catherine. 2011. “What’s in a Name? The Academic Study of Comics
and the ‘Graphic Novel.’” Cinema Journal 50.3: 123–26.
Laffay, Albert. 1964. Logique du cinema. Paris: Masson.
Larcenet, Manu. 2011. Interview by Agnès Deyzieux. Le cas des cases. http://
agnesdeyzieux-bd.blogspot.fi/2011/04/larcenet-blast-tome-2.html.
Larsson, Donald. 2000. “Every Picture Tells a Story: Agency and Narration
in Film.” Paper given at MLA Annual Conference, Washington, DC. http://
www.english2.mnsu.edu/larsson/agency1.html#text%201.
Leech, Geoffrey N. and Michael H. Short. 2007. Style in Fiction: A Linguistic
Introduction to English Fictional Prose. 2nd ed. London: Pearson/Longman.
Lefèvre, Pascal. 2005. “The Importance of Being Published.” Extended version of
original publication in Comics & Culture, 91–105. Eds. Anne M
agnussen and
Hans-Christian Christiansen. Copenhagen: Museum Tusculanum at the
University of Copenhagen, 2000. https://www.academia.edu/660771/The_
Importance_of_Being_Published_A_Comparative_ Study_of_Different_
Comics_Formats.
———. 2006. “The Battle over the Balloon. The Conflictual Institutionalization
of the Speech Balloon in Various European Cultures.” Image & Narrative VII.1.
http://www.imageandnarrative.be/inarchive/painting/pascal_levevre.htm.
———. 2007. “Incompatible Visual Ontologies? The Problematic Adapta
tion of Drawn Images.” In Film and Comic Books, 1–12. Eds. Ian Gordon,
Mark Jancovich and Matthew P. McAllister. Jackson: University Press of
Mississippi.
———. 2009. “The Conquest of Space: Evolution of Panel Arrangements and
Page Layouts in Early Comics Published in Belgium (1880–1929).” European
Comic Art 2.2: 227–52.
———. 2010. “Researching Comics on a Global Scale.” In Comics Worlds and
the World of Comics: Towards Scholarship on a Global Scale, 89–95. Ed.
Jaqueline Berndt. Kyoto: International Manga Research Center, Kyoto Seika
University
———. 2011. “Some Medium-Specific Qualities of Graphic Sequences.” Sub-
Stance 40.1: 14–33.
———. 2013. “Narration in the Flemish Dual Publication System. The Cross
over Genre ‘Humoristic Adventure.’” In From Comic Strips to Graphic
Novels: Contributions to the Theory and History of Graphic Narrative,
255–69. Eds. Daniel Stein and Jan-Noël Thon. Berlin and Boston, MA:
Walter de Gruyter.
Livingston, Sonia. 1998. Making Sense of Television. The Psychology of Audi-
ence Interpretation. 2nd ed. London and New York: Routledge.
Louwerse, Max and Don Kuiken. 2004. “The Effect of Personal Involvement in
Narrative Discourse.” Discourse Processes 38.2: 169–72.
Lyons, John. 1977. Semantics. Volume 2. Cambridge: Cambridge University Press.
Magliano, Joseph P. and Jeffrey M. Zacks. 2011. “The Impact of Continuity
Editing in Narrative Film on Event Segmentation.” Cognitive Science 35:
1489–517.
Margolin, Uri. 1995. “Characters in Literary Narrative: Representation and
Signification.” Semiotica 106: 373–92.
288 Bibliography
———. 2005. “Character.” In Routledge Encyclopedia of Narrative The-
ory, 52–57. Eds. David Herman, Manfred Jahn and Marie-Laure Ryan.
New York: Routledge.
———. 2007. “Character.” In The Cambridge Companion to Narrative, 66–79.
Ed. David Herman. Cambridge: Cambridge University Press.
Marion, Philippe. 1993. Traces en cases. Travail graphique, figuration narrative
et participation du lecteur. Essai sur la bande dessinée. Louvain-la-Neuve,
Belgium: Université Catholique de Louvain: Academia.
McCloud, Scott. 1993. Understanding Comics: The Invisible Art. Northampton,
MA: Kitchen Sink Press.
———. 2000. Reinventing Comics. New York: Perennial/HarperCollins.
McHale, Brian. 1978. “Free Indirect Discourse: A Survey of Recent Accounts.”
PTL: A Journal for Descriptive Poetics and Theory of Literature 3.2: 249–87.
———. 1993. “Whatever Happened to Descriptive Poetics?” In The Point of
Theory. Practices of Cultural Analysis, 56–65. Eds. Mieke Bal and Inge
E. Boer. Amsterdam: Amsterdam University Press.
Meesters, Gert. 2010. “Les significations du style graphique.” Textyles, revue
des Lettres belges de langue française 36–7: 519–26.
Meister, Jan C. 2015. “Narratology.” In The Living Handbook of Nar-
ratology. Eds. Peter Hühn, John Pier, Wolf Schmid and Jörg Schönert.
Hamburg: Hamburg University. http://wikis.sub.uni-hamburg.de/lhn/index.
php/Narratology.
Meskin, Aaron. 2007. “Defining Comics?” The Journal of Aesthetics and Art
Criticism 65.4: 369–79.
Miall, David S. and Don Kuiken. 1994. “Foregrounding, Defamiliarization,
and Affect Response to Literary Stories.” Poetics 22: 389–407
Mikkonen, Kai. 2008. “Presenting Minds in Graphic Narratives.” Partial
Answers 6.2: 301–21.
———. 2010. “Remediation and the Sense of Time in Graphic Narratives.” In
The Rise and the Reason of Comics and Graphic Literature. Critical Essays
on the Form, 74–86. Eds. Joyce Goggin and Dan Hassler-Forest. Jefferson,
NC: McFarland Press.
———. 2011a. “The Implicit Narrator in Comics. Transformations of Free In
direct Discourse in Two Graphic Adaptations of Madame Bovary.” Inter
national Journal of Comic Art 13.2: 473–87.
———. 2011b. “Graphic Narratives as a Challenge to Transmedial Narratology:
The Question of Focalization.” American Studies/Amerikastudien 56.4: 639–45.
———. 2012. “Focalisation in Comics. From the Specificities of the Medium to
Conceptual Reformulation.” Scandinavian Journal of Comic Art 1: 69–95.
———. 2013. “Subjectivity and Style in Graphic Narratives.” In From Comic
Strips to Graphic Novels: Contributions to the Theory and History of
Graphic Narrative, 101–26. Eds. Daniel Stein and Jan-Noël Thon. Berlin and
New York: Walter de Gruyter/Narratologia Series.
Miller, Ann. 2007. Reading Bande Dessinée: Critical Approaches to
French-Language Comic Strip. Bristol: Intellect Books.
Miodrag, Hannah. 2013. Comics and Language: Reimagining Critical Dis-
course on the Form. Jackson: University Press of Mississippi.
MODE. 2012. Glossary of Multimodal Terms. https://multimodalityglossary.
wordpress.com/.
Bibliography 289
Morgan, Harry. 2009. “Graphic Shorthand: From Caricature to Narratology in
Twentieth-Century Bande dessinée and Comics.” European Comic Art 2.1:
21–39.
Nelles, William. 1990. “Getting Focalization into Focus.” Poetics Today 11:
363–82.
Nikolajeva, Maria and Carole Scott. 2001. How Picturebooks Work. New York
and London: Garland Publishing.
Nodelman, Perry. 1991. “The Eye and the I: Identification and First-Person
Narratives in Picture Books.” Children’s Literature 19: 1–30.
O’Neill, Patrick. 1992. “Points of Origin: On Focalisation in Narrative.”
C anadian Review of Comparative Literature/Revue Canadienne de Littéra-
ture Comparée 19: 331–50.
Oswalt, Robert L. 1994. “Inanimate Imitatives in English.” In Sound Sym-
bolism, 293–306. Eds. Leanne Hinton, Johanna Nichols and John J. Ohala.
Cambridge: Cambridge University Press.
Palmer, Alan. 2004. Fictional Minds. Lincoln: University of Nebraska Press.
———. 2010. Social Minds in the Novel. Columbus: Ohio State University
Press.
Patron, Sylvie. 2009. Le narrateur. Introduction à la théorie narrative. Paris:
Armand Collin.
Peeters, Benoît. 1991. Case, planche, récit: comment lire une bande dessinée.
Tournai: Casterman.
———. 1998. Lire la bande dessinée. Paris: Flammarion.
———. 2007. “Four Conceptions of the Page.” Trans. Jesse Cohn. ImageTexT:
Interdisciplinary Comics Studies 3.3. http://www.engl…eeters/index.shtml.
Persson, Per. 2003. Understanding Cinema: A Psychological Theory of Moving
Imagery. Cambridge: Cambridge University Press.
Phelan, James. 1989. Reading People, Reading Plots. Character, Progression,
and the Interpretation of Narrative. Chicago and London: The University of
Chicago Press.
———. 1996. Narrative as Rhetoric. Technique, Audiences, Ethics, Ideology.
Columbus: Ohio State University Press.
———. 2005. Living to Tell about It. A Rhetoric and Ethics of Character Nar-
ration. Ithaca, NY and London: Cornell University Press.
———. 2007. Experiencing Fiction: Judgments, Progressions, and the Rhetori-
cal Theory of Narrative. Columbus: The Ohio State University Press.
Phelan, James and Peter J. Rabinowitz. 2012. “Character.” In Narrative Theory:
Core Concepts and Critical Debates, 111–18. Eds. David Herman, James
Phelan, Peter J. Rabinowitz and Robyn Warhol. Columbus: The Ohio State
University Press.
Postema, Barbara. 2013. Narrative Structure in Comics. Making Sense of Frag-
ments. Rochester, NY: RIT Press.
Potsch, Elisabeth and Robert F. William. 2012. “Image Schemas and Concep
tual Metaphor in Action Comics.” In Linguistics and the Study of Comics,
13–36. Ed. Frank Bramlett. London: Palgrave Macmillan.
Pound, Richard. “‘The Gallery of Comicalities’: An Introduction to the Cari
catures of C. J. Grant, (flu. C1830–1846).” http://www.gjsaville-caricatures.
co.uk/articles/GJGrant (originally published in Graham Saville’s Caricature
Catalogue 25, 1998).
290 Bibliography
Prince, Gerald. 1992. Narrative as Theme. Studies in French Fiction. Lincoln
and London: University of Nebraska Press.
———. 2003a. Dictionary of Narratology. Lincoln: University of Nebraska Press.
———. 2003b. “Surveying Narratology.” In What Is Narratology? Questions
and Answers Regarding the Status of a Theory, 1–16. Eds. Tom Kindt and
Hans-Harald Müller. Berlin: Walter de Gruyter.
Rabkin, Eric S. 2009. “Reading Time in Graphic Narrative.” In Teaching the
Graphic Novel, 36–43. Ed. Stephen E. Tabachnick. New York: The Modern
Language Association of America.
Richardson, Brian. 1997. Unlikely Stories. Causality and the Nature of Modern
Narrative. Newark: University of Delaware Press.
Rimmon-Kenan, Shlomith. 2003. Narrative Fiction. Contemporary Poetics.
2nd ed. London and New York: Routledge.
Rommens, Aarnoud. 2000. “Manga Story-Telling/Showing.” Image and N arrative
1. http://www.imageandnarrative.be/inarchive/narratology/aarnoudrommens.
htm.
Round, Julia. 2007. “Visual Perspective and Narrative Voice in Comics: Re
defining Literary Terminology.” International Journal of Comics Art 9.2:
316–29.
———. 2013. “Anglo-American Graphic Narrative.” In From Comic Strips to
Graphic Novels: Contributions to the Theory and History of Graphic Nar-
rative, 325–45. Eds. Daniel Stein and Jan-Noël Thon. Berlin and New York:
Walter de Gruyter.
Ruusuvuori, Johanna and Anssi Peräkylä. 2009. “Facial and Verbal Expressions
in Assessing Stories and Topics.” Research on Language and Social Inter
action 42.4: 377–94.
Ryan, Marie-Laure. 2004. “Introduction.” In Narrative across Media, 1–40.
Ed. Marie-Laure Ryan. Lincoln: University of Nebraska Press.
———. 2005. “On the Theoretical Foundations of Transmedial Narratology.”
In Narratologia: Narratology Beyond Literary Criticism: Mediality, Disci-
plinarity, 1–23. Berlin: Walter de Gruyter.
———. 2006. Avatars of Story. Minneapolis: University of Minnesota Press.
———. 2013. “Transmedial Storytelling and Transfictionality.” Poetics Today
34.3: 361–88.
———. 2015. “Narration in Various Media.” In The Living Handbook of
Narratology. Eds. Peter Hühn, John Pier, Wolf Schmid and Jörg Schönert.
Hamburg: Hamburg University. http://wikis.sub.uni-hamburg.de/lhn/index.
php/Narration_in_Various_Media.
Ryan, Marie-Laure and Jan-Noël Thon. 2014. “Storyworlds across Media.
Introduction.” In Storyworlds across Media: Toward a Media-Conscious
Narratology, 12–22. Eds. Marie-Laure Ryan and Jan-Noël Thon. Lincoln:
Sadrô, Javid, Izzat Jarudi and Pawan Sinha. 2003. “The Role of Eyebrows in
Face Recognition.” Perception 32.3: 285–93.
Saraceni, Mario. 2001. “Relatedness: Aspects of Textual Connectivity in
Comics.” In The Graphic Novel, 167–79. Ed. Jan Baetens. Leuven: Leuven
University Press.
———. 2003. The Language of Comics. London and New York: Routledge.
Sattler, Peter R. 2010. “Past Imperfect: ‘Building Stories’ and the Art of
Memory.” In The Comics of Chris Ware: Drawing is a Way of Thinking,
Bibliography 291
206–22. Eds. David M. Ball and Martha B. Kuhlman. Jackson: University
Schüwer, Martin. 2008. Wie Comics erzählen. Grundriss einer intermedialen
Erzähltheorie. Trier: Wissenschaftlicher Verlag Trier.
Sellors, C. Paul. 2007. “Collective Authorship in Film.” The Journal of Aesthet-
ics and Art Criticism 65.3: 263–71.
Semino, Elena. 2011. “Deixis and Fictional Minds.” Style 45.3: 418–40.
Shinohara, Kazuko and Yoshihiro Matsunaka. 2009. “Pictorial Metaphors
of Emotion in Japanese Comics.” In Multimodal Metaphor, 265–93. Eds.
E duardo Urios-Aparisi and Charles Forceville. Berlin: Walter de Gruyter.
Singer, Marc. 2012. “Time and Narrative: Unity and Discontinuity in The
Invisibles.” In Critical Approaches to Comics. Theories and Methods,
55–70. Eds. Matthew J. Smith and Randy Duncan. New York and London:
Routledge.
Sinha, Pawan, Benjamin Balas, Yuri Ostrovsky and Richard Russell. 2005.
“Face Recognition by Humans: 20 Results all Computer Vision Research
ers Should Know About.” http://web.mit.edu/bcs/sinha/papers/20Results_
2005.pdf.
Smith, Greg M. 2003. Film Structure and the Emotion System. Cambridge:
Cambridge University Press.
Smith, Murray. 1995. Engaging Characters: Fiction, Emotion, and the Cinema.
Oxford: Clarendon.
Smolderen, Thierry. 2002. “Ceci n’est pas une bulle! Structures énonciatives du
phylactère.” In Actes des Quatrièmes Rencontres Réseaux Humains/Réseaux
Technologiques, 107–39. Eds. Eric Espéret and Jean-François C erisier.
Poitiers: CNDP. http://rhrt.edel.univ-poitiers.fr/document.php?id=555.
———. 2009. Naissances de la bande dessinée. De William Hogarth à Winsor
McCay. Bologne: Les Impressions Nouvelles.
———. 2012a. “Les bandes dessinées du Graphic et de l’Illustrated London
News.” Neuvième Art 2.0. http://neuviemeart.citebd.org/spip.php?rubrique71.
———. 2012b. “Histoire de la bande dessinée: questions de méthodologie.” In
La bande dessinée: une médiaculture, 71–90. Eds. Éric Maigret and Matteo
Stefanelli. Paris: Armand Colin.
———. 2014a. Origins of Comics: From William Hogarth to Winsor McCay.
———. 2014b. “Graphic Hybridization, the Crucible of Comics.” In The
French Comics Theory Reader, 47–61. Eds. Ann Miller and Bart Beaty.
Leuven: L euven University Press/Studies in European Comics and Graphic
Novels.
Stanzel, F.K. 1984. A Theory of Narrative. Trans. Charlotte Goedsche.
Cambridge: Cambridge University Press.
Stöckl, Hartmut. 2004. “In Between Modes: Language and Image in Printed
Media.” In Perspectives on Multimodality, 9–30. Eds. Eija Ventola, Cassily
Charles and Martin Kaltenbacher. Amsterdam: John Benjamins Publishing
Company.
Surdiacourt, Steven. 2012. “Tying Ends Together: Surface and Storyworld in
Comics.” Comics Forum. http://comicsforum.org/2012/12/27/image-narrative-
8-tying-ends-together-surface-and-storyworld-in-comics-by-steven-surdiacourt/.
———. 2015. Comics and Storytelling. Towards a Mediumspecific Narrato
logy. PhD dissertation. Katholieke Universiteit Leuven Faculteit Letteren.
292 Bibliography
Swift, Jonathan. 1736/1998. “A Character, Panegyric, and Description of the
Legion Club.” In Verse in English from Eighteenth-Century Ireland, 222–30.
Ed. Andrew Carpenter. Cork: Cork University Press.
Talmy, Leonard. 2000. Toward a Cognitive Semantics. Cambridge, MA: MIT
Press.
Tan, Ed S. 2001. “The Telling Face in Comics Strip and Graphic Novel.” In The
Graphic Novel, 31–46. Ed. Jan Baetens. Leuven: Leuven University Press.
Thomson-Jones, Katherine. 2007. “The Literary Origins of the Cinematic
Narrator.” British Journal of Aesthetics 47.1: 76–94.
Todorov, Tzvetan. 1969. Grammaire du Décaméron. The Hague: Mouton.
Tomashevsky, Boris. 1965. “Thematics.” In Russian Formalist Criticism: Four
Essays, 61–95. Eds. Lee T. Lemon and Marion J. Reis. Lincoln and London:
Tondro, Jason. 2015. “‘Nightcrawlers’ Inferno and Other Hellish Tales: Com
ics Adaptations of Dante.” In Medieval Afterlives in Contemporary Culture,
291–300. Ed. Gail Ashton. London: Bloomsbury Publishing.
Töpffer, Rodolphe. 1845/1944. “Essai de physiognomonie.” In Töpffer, l’in-
vention de la bande dessinée, 185–225. Eds. Thierry Groensteen and Benoît
Peeters. Paris: Hermann.
Uidhir, Christy M. 2014. “Comics and Collective Authorship.” In The Art of
Comics. A Philosophical Approach, 47–67. Eds. Aaron Meskin and Roy
T. Cook. Chichester: Wiley Blackwell.
Van Lier, Henri. 1988. “La bande dessinée, une cosmogonie dure.” In Bande
dessinée, récit et modernité, 5–24. Ed. Thierry Groensteen. Paris: Colloque
de Cerisy.
Verstraten, Peter. 2009. Film Narratology. Trans. Stefan van der Lecq. Toronto:
University of Toronto Press.
Walker, Mort. 2000. The Lexicon of Comicana. Lincoln, NE: Authors Guild
Backinprint.com Edition.
Walton, Kendall L. 1997. “On Pictures and Photographs. Objections A nswered.”
In Film Theory and Philosophy, 60–75. Eds. Richard Allen and Murray
Smith. Oxford: Clarendon Press.
Waugh, Colton. 1947/1991. The Comics. Jackson: University Press of
Mississippi.
Widiss, Benjamin. 2013. “Comics as Non-sequential art. Chris Ware’s Joseph
Cornell.” In Drawing from Life: Memory and Subjectivity in Comic Art,
86–111. Ed. Jane Tolmie. Jackson: University Press of Mississippi.
Witek, Joseph. 2009. “The Arrow and the Grid: Creating the Comics Reader.”
In A Comics Studies Reader, 149–56. Eds. Jeet Heer and Kent Worcester.
Wolf, Werner. 2003. “Narrative and Narrativity: A Narratological Reconceptu
alisation and Its Applicability to the Visual Arts.” Word & Image: A Journal
of Verbal/Visual Enquiry 19.3: 180–97.
Wolk, Douglas. 2007. Reading Comics. How Graphic Novels Work and What
They Mean. Philadelphia, PA: Da Capo Press.
Woloch, Alex. 2003. The One vs. the Many. Minor Characters and the Space of the
Protagonist in the Novel. Princeton, NJ and Oxford: Princeton University Press.
Yannicopoulou, Angela. 2010. “Focalization in Children’s Picture Books.”
In Telling Children’s Stories: Narrative Theory and Children’s Literature,
65–85. Ed. Mike Cadden. Lincoln: University of Nebraska Press.
Index
Ada (by Altan) 231 Blake and Mortimer (by Edgar

affordance 19–20, 23, 29n27, 220 P. Jacobs) 67
alignment 102–06 Blast (by Manu Larcenet) 120–23
Ally Sloper (magazine) 274n4 Booth, Wayne C. 76
Altman, Rick 102–03, 105–06 Bordwell, David 4, 93, 110, 136,
angle 7, 38, 42, 94–6, 169, 234, 254, 148n7, 148n8
274n8; eyeline image 165–67; Branigan, Edward 4, 70n38, 167,
(page) bleed 57, 234; close up 54, 173n13, 173n17
79, 92, 101, 169, 210, 223–24, 234, Bretécher, Claire 228
252–53; over-the-shoulder image Bridgeman, Teresa 101, 107n2
94, 165–66, 168; panorama 101, broadsheet print 228, 246–47, 249,
231, 250, 268, 269; reaction image 251, 253, 268, 272–73
166, 168, 223; reverse 92, 139, 168 Brooke-Rose, Christine 53, 56
‘arthrology’ 42–3, 68n14 Brooklyn Station Terminus Cosmos
Ascension du Haut Mal (Epileptic) (by Jean-Claude Mézières and
(by David B) 141 Pierre Christin) 239
Asterios Polyp (by David Bugs Bunny (comic) 1
Mazzucchelli) 120, 123, 125n11, Building Stories (by Chris Ware)
154–55 33–35, 38, 43, 49, 50, 51, 63, 65,
Asterix (by Albert Uderzo and René 67, 67n2, 150, 176
Goscinny) 137, 227, 230, 233,
236–37; Asterix and the Banquet Cage, The (by Martin Vaughn-James)
232 174, 175–76
authorship in comics 130, 148n2; Calvin and Hobbes (by Bill Watterson)
implied author 6, 134–36 6, 23, 112, 171
autobiographical comics 10, 17, 23, caption 54, 56, 62, 92, 97, 105, 107n6,
99, 124n3, 130, 132–33, 137, 176, 220, 228, 232, 236, 249–50,
140–42, 161, 220, 227 252, 265, 266, 267; lyrical 175–76;
narrative 132, 134, 144, 232, 236
Baetens, Jan 5, 8–9, 40, 68n16, 69n23, cardinal function (in plot) 58, 59,
70n35, 70n43, 88n6, 89n14, 89n16, 70n37
108n16, 149n15, 241n7 caricature 64, 246, 247, 248, 249,
Bal, Mieke 4, 28n6, 151, 171, 173n9, 250, 251, 252, 253, 256, 257, 258,
184, 198n6, 210, 218 170, 272, 273, 275n20; and reality
Bande pas dessinée (by Bruno 164, 198; art of 195, 197, 246, 248;
Muschio) 231 as a form of perception 219; as style
Barefoot Gen (by Keiji Nakazawa) 141 114–15, 198; bodily forms 222,
Bateman, H.M. 79–81 224; characters 179, 180, 186, 188,
Beaty, Bart 13–14, 29n22, 41, 68n12 195, 197, 199n 20; Golden Period
Berlin (by Jason Lutes) 207 of 27, 251, 256, 272
294 Index
caricature magazine 27, 246, 247, 125n15; stylistic marker 42;
249, 251, 252, 256, 260, 264, 269, subjectivity 104, 109, 120, 215n15
271, 272–73 comic strips 6–7, 24, 66–67, 94, 132,
Carroll, Noël 19 220, 231, 246, 247–50, 262, 264,
cartoon 10, 12, 18, 77, 87, 88n5, 272, 274n6, 275n13
88n6, 228, 247–47, 250, 273, comics: concept of 12–15, 247, 274n6
274n4; as style 6, 101, 195 Commando (comic) 99
character: allegorical 179, 189–90, connectivity 38, 45–48, 63, 81, 91–92,
193, 199n20; and action 178–79; 95, 102, 106, 123, 233, 258
as an individual 119–20, 122, 123, constraint: artistic (formal) 20–23,
156, 158, 185, 186, 189, 190–92, 29n29, 29n30; medium-specific
195, 197, 199n15, 217, 261; 8, 156, 203, 204, 211, 217, 220,
as means of narrative continuity 237–38
90–107; basis type 179–80, 184; continuing-consciousness frame 122,
complex 178, 186–93, 199n20, 125n14, 137, 215, 217
246; concept of 177; continuing Corto Maltese (by Hugo Pratt) 58–59,
presence 90–91, 183; flat 186, 188, 195, 200n24, 208–09; La Ballade
189, 199n17; mental state 36–37, de la mer salée (The Ballad of the
50, 84, 91, 96, 120, 123, 176, Salty Sea) 206; Les Celtiques (Celtic
179–80, 183, 185, 186, 188, 190, Tales) 58
191, 195, 196–97, 225–28, 239, Cruikshank, George 251, 253, 254,
241n12; mimetic function (or 256, 257, 275n23
dimension) 107, 107n1, 107n3, Cruikshank, Robert 269
178, 179, 180, 193, 199n8; Cutting, James E. 108n12, 108n14
movement of 94, 96–100; reflecting
150, 219n8; round 186, 188, Dante (Alighieri) 180, 183, 185, 189;
199n17; shift 96, 108n12; synthetic Inferno (from The Divine Comedy)
function (or dimension) 81, 91, 106, 180, 181, 182, 189, 194, 200n23
107n3, 178, 193, 199n8; thematic Dark Night Returns, The (by Frank
function (or dimension) 107n3, Miller) 193
178, 179, 185, 187, 192, 193, deixis 120, 152, 155, 156–57, 160,
199n8; type-like 186, 188, 189, 170, 172, 172n1, 199n12
190, 192, 197 Delany, Samuel R. 10, 28n16
characterisation in comics 8, 9, 37, Delisle, Guy 58–59, 141, 215
50, 114, 120, 174–98, 199n13, Dessous Troublants (by Jeanne
233; definition of 177–78; Puchol) 176
principles of 184–85 dialogue in comics 220–40; and
Chat du rabbin, Le (The Rabbi’s Cat) action 96–97, 137, 207, 220; bond
(by Joann Sfar) 113, 207 between speaker and utterance
Chatman, Seymour 4, 6–7, 28n9, 53, 228–32; characterisation 92, 176,
134–35, 148n1 185, 208, 217; conversational
Chavanne, Renaud 68n22, 107n11 scene 53, 76, 94–95, 134, 144, 168,
Chute, Hilary 17 188, 220, 224–26, 227–28, 230,
‘closure’ 38–40, 42, 68n10 231, 232, 233–36, 238–39, 240;
Clumsy (by Jeffrey Brown) 49 duration 53, 253; embodied speech
Cohan, Steven 6 situation 221–25, 239; fundamental
Cohn, Dorrit 205–06 narrator 133–34, 144; gaze 222,
Cohn, Jesse 68n21, 69n24 223, 227–28, 234, 241n6; narrative
Cohn, Neil 68n10, 68n17, 69n27 function 232–38, 238–39;
colour: characterisation 190; non-verbal communication 222–25;
emotional state 229; means of 180° rule 107n10; repetition 62;
layout 54; perspective 150, 154, showing 76
167, 169, 230; salience 101, 230; Dickens, Charles 186, 199n16, 251
stylistic effect 70n39, 83, 120–122, diegesis and mimesis 76, 88n9
Index 295
Distant Neighborhood, A (by Jiro perceptual 158–60, 170, 171;
Taniguchi) 24, 137–48 point-of-view image 138–44, 151,
Donald Duck (comic) 1 159–60, 165–66, 168, 173n13;
Donaldson-Evans, Mary 209–10 perception image 166–67; zero-level
duration, see time in comics 139, 147, 151; see also angle
Dürrenmatt, Jacques 29n30, 89n13, following-unit, see alignment
241n12 Forceville, Charles 168, 210,
230, 241n13, 241n15, 241n18,
editing techniques 54, 81, 92–93, 242n23
107n5, 131, 167; eyeline match Forster, E. M. 186, 188, 199n17
165–67, 169, 216; match cut frame 7, 22, 38, 44–45, 65, 160–64,
(or graphic cut/match) 41, 93, 230; broken 85, 108n17, 145,
107n5, 165–67; match on action 233–34, 270–71; embedded 45–46;
92–96, 108n13 expressive function 87, 110, 113,
Eisner, Will 29n25, 141, 234–35, 117, 142, 144, 145, 166, 168, 206,
242n22 240, 262; frame of vision 218; key
emanata 109, 225–27, 239, 241n11 frame 57–60; hors-champ (off panel
Ethel & Ernest. A True Story space) 88, 89n16, 157; rhythm 54,
(Raymond Briggs) 85 69n32, salience 70n39; see also
Evens, Brecht 228, 230 multiframe
Every Body’s Album and Caricature Fred (Frédéric Othon Théodore
Magazine 251, 258, 259, 266, 269, Aristidès) 23, 47, 51, 57, 112, 119;
272, 273 Philémon 47, 51–2, 57, 112
free indirect discourse (FID) 28n14,
facial expression 47, 80, 84, 97, 167, 168, 204, 209–14, 217, 220
184, 196, 205, 241n10; conversation free indirect perception 168, 203, 211
27, 222–25, 227, 239; deep structure Fresnault-Deruelle, Pierre 5, 36, 60
of visual storytelling 22–23 Frey, Hugo 40, 69n23, 88n6, 108n16,
“Feinte Trinité” (by François Ayroles) 149n15
231 Fun Home (by Alison Bechdel) 17, 141
film narrative 54, 81, 93, 134, 154,
157, 161, 167, 203, 224 Gaudreault, André 4, 5, 81–83, 88n8,
film studies 4, 9, 11, 25, 110, 129, 88n9, 124n2, 131, 134, 148n8,
131, 136, 147, 148n3, 167, 173n17, 149n17
203–04, 241n6 Gemma Bovery (by Posy Simmonds)
Fishelov, David 188–93 194, 209, 214–15
Flaubert, Gustave 209, 212–13, 215 Genette, Gérard: characterisation
Fludernik, Monika 29n24, 152–53, 177, 198n6, 199n7; discourse-time
154, 155, 203, 219n11 36, 37, 69n31; film narratology 4;
focalisation 3, 5, 8, 28n11, 37, 49, focalisation 5, 87, 139, 150–51,
91, 103, 150–72; authorial 140, 155, 157, 160–61, 172n3, 172n7,
217; cognitive 120, 152, 157–60; 173n10, 173n18; frequency 60–61;
definition of 150–52; direct gaze global (or synchronic) look 33,
167; embedded (or simultaneous) 35, 36, 66; metalepsis 69n30,
6, 171, 173n18, 210; external 6, 138; mimesis 76, 156; narrative
151, 165, 219n5; focaliser 26, 132, mood 28n11; narratology 3, 8, 83;
139, 143, 147–48, 151–52, 154, narrator 5, 87, 149n10, 149n13;
158, 160–65, 171, 172, 173n11, paralepsis 215; rhythm 52–3, 56;
210, 218; focalised 91, 147–48, story-time 37; temporal order 37,
151–52, 155–56, 171–72, 217; 68n6, 69n29
gaze image 116, 153, 165–68, 170, genre 18, 23, 65–67, 70n44, 99, 106,
223; impersonal 165; internal 49, 109, 111, 137, 157, 167, 177,
139, 142, 149n16, 151, 165, 212; 193–96, 245–46, 278
ocularisation 8, 157–60, 173n9; Gibson, James J. 29n27
296 Index
Gillray, James 251, 252, 256, 257, Here (by Richard McGuire) 46
258, 268, 272, 274n8, 275n16 Hergé (Georges Prosper Remi) 46,
Glasgow Looking Glass (magazine) 107n8, 113, 125n8, 186, 195, 205
251, 269, 272, 274n11, 274n12 Herman, David 8, 16, 28n4, 57–58,
global look 33, 36, 40, 46, 64–66, 63, 70n42, 198n5, 200n27
181, 260, 269; see also reading historical analysis of comics 10–11,
Gorey, Edward 199n14 105–06, 245–51, 278; see also
Grant, Charles Jameson 251, 258–59, narratology, diachronic
266, 272–73, 274n11 Hogarth, William 247, 248, 250,
graphiation, see graphic style 256, 265
graphic cut, see editing techniques
graphic match, see editing techniques image: concept of 78–79
graphic narrative: definition of 17 image and word interplay 13–15,
graphic novel 8, 10, 17, 29n25, 40, 64–65, 75, 120, 147, 153–54, 156,
67, 88n6, 94, 99, 108n16, 124, 144, 184, 196–7, 204, 215–16, 219, 221,
149n15, 161, 186, 194–95, 217–18, 232, 238–39
245–46, 271 imitative 221, 226, 240n2
graphic style 9, 11, 47, 75, 86–87, implied author, see authorship in
109–24, 124–125n7, 147, 184, 217, comics
246–47; cartoony 101, 114–15, Intérieurs (by Régis Franc) 174–75
195–96; definition of 110–13; intermediality 3–4, 15, 19, 28n5
graphiation 86–87, 88n12, 88n13, “I Want You to Like Me” (Martin
88n14, 131–32; ligne Claire 113, Cendreda) 101, 231–32
115; mind style 119–24; narrative
function 11, 47, 86, 87, 111–12, Jahn, Manfred 135, 148n6, 155, 166
115–19, 119–23; photorealism 75, Jannidis, Fotis 177, 179–80
115, 148, 195–196, 198; rupture Jimbo’s Inferno (by Gary Panter)
47, 115–19, 120–23, 125n12, 147; 180–85, 189–90, 199n10,
variation 115–19; see also writing 199n11
Groensteen, Thierry 5, 42; artistic Jimmy Corrigan: The Smartest Kid on
constraints 21; bipolar structure Earth (by Chris Ware) 150
between speech balloons and Jost, François 4, 88n11, 124n2,
characters 230, 240; braiding 148n8, 157–60, 170
42–43; character 198; definition of juxtaposition, see narrative
comics 12, 29n21; frames 69n32; organisation
foyer perceptif 173; fundamental
narrator 112, 131–33, 149n11; Kindt, Tom 9–10
iconic solidarity 12, 28n12, 42, Kirby, Jack 57, 130, 195
56, 68n20; layout style 69n23; Kress, Günther 27n3, 29n27, 101, 167
multicouche 157; multiframe 43, Kukkonen, Karin 8, 107n10
45; narrative drawing 77, 88n2; Kunzle, David 108n17, 245, 250, 264,
narratology 5, 28n8; repetition 274n7, 275n13
70n41; rhythmic function of
speech balloons 242n26; salience Lagoon, The (by Lilli Carré) 238
70n39; string vs. sequence 274n7; League of Extraordinary Gentlemen,
stylistic variation 115–16; see also The (by Alan Moore and Kevin
‘arthrology’ and multiframe O’Neill) 74
Leech, Geoffrey 48–50, 119–20
Hagelberg, Matti 1, 43–44 Leech, John 77–78, 247
Harvey, Robert C. 68n11, 111 van Leeuwen, Theo 27n3, 101, 167
Hatfield, Charles 57, 70n35, 125n11 Lefèvre, Pascal 89n16, 124–125n7,
Heath, William 251, 252, 253, 255, 242n19, 273n1
256, 261, 262, 263, 268, 272, 273, Léo (Luiz Eduardo de Oliveira) 74–75
274n11, 275n17 Lichtenstein, Roy 60
Index 297
literary narrative: adaptation 196–96; Miller, Ann 8
characterisation 184, 194, 197, 218; mind style, see graphic style
dialogue 224; diegetic level 142, Miodrag, Hannah 14, 28n19, 29n20,
238; focalisation 147, 151, 154, 68n17, 111, 124n6
155–56, 159, 161, 165, 171, 172n3, mise en page, see page layout
218; narrator 82, 148n4; showing Mr. Block (by Ernest Riebe) 183
75–76; speech and thought 206, “Mr Spoonbill’s Experiences in the
209–12, 217; time 50–51, 52, 209, Art of Skating” (by John Tenniel)
238–89; utterance 229 99–100
literary value in comics 13–16, Mister Wonderful (by Daniel Clowes)
28n18 232, 236, 238
Little Nemo (by Winsor McCay) 46, Monsieur Vieux Bois (by Rodolphe
112 Töpffer) 205
“logophobia” 14, 29n20 monstration, see showing
Morgan, Harry 5, 107n2, 200n25
Madame Bovary (by Daniel Bardet Mother, Come Home (by Paul
and Michel Janvier) 209–13, 219n6 Hornschemeier) 216
Marey effect 156–57, 172n6 Motherless Oven, The (by Rob Davis)
Margolin, Uri 185, 198n6 94
Marion, Philippe 5; colour effects multiframe (multicadre) 43–45, 68n19
122; continuity 45–48; graphic multimodality 2, 9, 13, 14, 18, 19,
enunciation 82–83, 131–33; 22–23, 27n3, 29n27, 64, 91–92,
graphisme 78; “mega-narrator” 153, 154, 238–39; characterisation
112, 131–33; monstration 86, 192, 197; focalisation 160–161;
89n14, 131; narrative and painterly multimodal document 27n3, 144;
functions of a panel 60, 70n40; speech representation 203–04, 209,
stylistic rupture 119 211, 218–19, 221–22
Marshall Blueberry (Le Spectre aux Müller, Hans-Harald 9–10
balles d’or/The Ghost with the
Golden Bullets) (by Jean-Michel Nao of Brown, The (by Glyn Dillon)
Charlier and Jean Giraud) 190–91, 194, 197
206–08 narrative: behaviourist 208–09,
match cut, see editing techniques 217, 218, 219n5; causal cohesion
match on action, see editing 90–1; concept of 15–16;
techniques first-person 49, 92, 101, 124,
Maus (by Art Spiegelman) 17 130, 132, 137, 143–45, 146, 147,
McCloud, Scott 12, 29n20, 38–9, 149n13, 149n16, 161, 170, 175,
41–42, 67n1, 69n29, 195; The 179, 189, 192, 193–94, 196,
Sculptor 54, 167 204, 207, 214, 215–16, 218, 220,
medium: definition of 17–18; comics 232; it-narrative 174–76, 198n1;
as a medium 7–8, 9, 13, 15–18, 20, levels 123, 138, 145, 160;
100, 131, 159, 188, 247; see also third-person 106, 204, 205,
medium-specificity 207–09, 215, 217, 218; visual 4,
medium-specificity 3, 6–7, 8, 9, 11, 7, 36, 54, 151, 153, 157, 159, 171,
12, 18–23, 81, 86–88, 95, 106, 123, 210–11, 217–18
129–30, 148, 152, 157, 166, 171–2, narrative agency 83, 124, 129–31,
177, 194–95, 203, 211–12, 218, 135–36, 138, 139, 141–42, 143,
220–21, 224, 238, 240, 277 144, 147–48; see also narrator
metalepsis 51, 69n30, 138, 142 narrative breakdown 24, 28n15, 42,
Meti (by Aapo Rapi) 227–28, 238 48, 55, 65, 68n11, 131, 206
Mickey’s Inferno (L’Inferno di narrative organisation: juxtaposition
Topolino) (by Guido Martina and 40, 47, 205, 252–60; sequentiality
Angelo Bioletto) 180–83, 185–87, 22, 69n27, 97, 248, 249, 260–68,
189, 191–92, 195, 197, 200n26 272, 273, 275n23; simultaneity
298 Index
154–55, 248, 249, 268–73; see also page: as a unit of attention and
sequencing design 38, 42, 46, 47, 49, 51, 64,
narrative salience, see salience 65, 68–69n22, 269–72; double
narrativist bias 13–14 spread 42, 47, 49, 51, 57, 64, 65,
narrativity 2, 11, 16, 22, 27n1, 43, 66, 95–96; splash 57; see also page
45, 90–1, 106, 107n2, 178, 203, layout and global look
254–55, 273 page layout 10, 20, 22, 24, 29n30, 40,
narratological analysis 9–11, 277–78 42–43, 45, 46–48, 51, 54–56, 61,
narratology 2–11, 15, 16, 22–23, 64–66, 68n21, 68n22, 70n41, 100,
28n6, 33, 36, 75, 81, 83, 103, 131, 101, 113, 234, 249, 270–71, 273;
144, 147, 150, 153, 172, 177–78, layout styles 46–47, 54, 64, 69n23,
203, 245, 277–78; diachronic 105, 69n24, 111, 246
246; postclassical 3, 28n4, 152; Palestine (by Joe Sacco) 17
transmedial 154, 168, 210 panel: as a photograph 79, 146,
narrator in comics 5, 8–9, 16, 29n23, 161–64; ‘memorable panel’ 60;
75–76, 81–83, 112, 123–24, narrative function 40, 60, 70n40,
130–37, 148, 149n11, 150, 153, 146; painterly (or picture) function
158, 159, 161, 165, 170, 215–16, 40, 60, 146; polyptych 57; relations
217–18, 220; continuing narratorial 22, 38–43, 45–50, 54, 68n17, 252,
voice 143, 204, 206; extradiegetic 272; wordless 84–85, 117, 139,
210; fundamental 112, 131–36; 144–46, 236; see also narrative
heterodiegetic 132, 137, 144, organisation, panel transitions
148n1, 149n10, 210; homodiegetic and time in comics, duration in a
148, 149n10, 214, 215; impersonal single panel, temporality in panel
narration 135–37; intradiegetic relations
132, 148n1; mega-narrator 82, 112, panel transitions 38–43, 92, 96,
131, 135, 137; narrative voice 36, 160, 274n9; modality-to-modality
87, 109, 124, 132, 135, 137, 143, transition 42, 112; types of 41–42
147–48, 152, 154, 207, 251; paralepsis 215
self-narration 139, 143–44, 216; Peeters, Benoît 46–47, 60, 69n23
shown in the image 132–33, 170; Pekar, Harvey 124n3, 141
see also narrative: first-person perception image, see focalisation
Newton, Richard 251, 252, 256, 258, Persepolis (Marjane Satrapi) 17, 132,
275n21 137, 141
Nikolajeva, Maria 196 perspective: camera-eye 41, 93–94,
non-fiction comics 17, 23, 136, 139, 140, 142, 155, 156, 158, 167, 168,
141–42, 161, 164, 176–77, 196, 173n9, 209, 215, 217; camera
198, 217, 225 movement 54, 129, 158, 168, 169,
non-narrative comics 14, 176–77, 210; spatially determined 152,
249, 268, 273 155–57, 169, 172; see also angle
Northern Looking-Glass (magazine) and focalisation
251, 253, 255, 256, 272, 275n20 Phelan, James 107n3, 178, 193,
199n8, 199n21
ocularisation, see focalisation Philémon, see Fred
Omni-visibilis (by Matthieu photorealism, see graphic style
Bonhomme and Lewis Trondheim) pictogram 109, 226, 241n13
159–60 pictorial rune 225–26, 241n11,
180° rule 96 93, 96, 107n10 241n13
O’Neill, Patrick 6 picture books 12–13, 18, 146, 153,
onomatopoeia 100, 109–10, 149n12, 161, 165, 196
221, 225, 226, 241n15 Pinocchio (by Winshluss) 115–19, 194
OuBaPO (Ouvroir de bande dessinée Pogo (by Walt Kelly) 227
potentielle) 21–22, 29n30 point-of-view image, see focalisation
Index 299
Postema, Barbara 68n15 presentational 49–50; psychological
Prince, Gerald 3, 29n23, 219n5 48–49; see also sequential form
‘progress’ story (narrative genre) sequential art 248
250, 256, 260–65, 272, 275n20 sequential form 196, 248–50, 256,
Prosopopus (by Nicolas De Crécy) 260–68, 272, 274n7; episodic
40–41, 55 sequential form 260, 272; narrative
publication format 10–11, 18, 20, 47, continuity sequence 260, 264, 272
65–66, 83, 245–51, 273, 273n2 Seymour, Robert 251, 267, 270, 272,
Punch (periodical) 77, 79, 98, 99, 273, 274n11
247, 251 Short History of America, The
Pyongyang: A Journey in North Korea (by Robert Crumb) 174–75
(by Guy Delisle) 58–60, 215–16 showing 73–89, 131–33, 136–38, 144,
153, 156–57, 214, 215, 219, 221;
Qu…, La (by Marc-Antoine Mathieu) continuous showing of character
122 96–97, 105–06, 217–18; definition
of 74; distinction from telling
reading: conventions 18, 38, 76–78, 150–51, 217; monstration
46, 64–67, 83, 193; ‘diegetic 81–86, 89n14, 131–35, 144,
absorption’ 273; ‘tabular’ 36, 43, 149n16, 149n17; narrative function
46, 65, 66, 70n43; unit of 44, 66 74, 77, 78, 83
repetition: characterisation 184, Shires, Linda M. 6
199n13; connectivity 47, 92, 96, Short, Mick 48–50, 119–20
106; iconic 21–22; multiphase simultaneity, see narrative
image 56–57; salience 63, 70n41, organisation
101; see also connectivity and time Smith, Murray 91, 102–03, 105,
in comics, frequency 108n22, 108n23
rhythm, see time in comics Smolderen, Thierry 241n19, 245,
Rimmon-Kenan, Slomith 172n7, 184, 247–49
197, 198n6, 199n13, 199n20 sound effect, see onomatopoeia
Round, Julia 96 spatial attachment, see alignment
Rumble Strip (by Woodrow Phoenix) speech 8, 54, 55, 133–34, 177–78,
175 203–04, 214, 217–19; direct 76,
Ryan, Marie-Laure 3, 27n1, 28n5 77, 192, 207, 208, 213, 214, 217,
219n4, 220, 228; indirect 203, 206,
Sacco, Joe 17, 141, 242n24 207, 209, 212–13, 214, 217, 218,
Saga (by Vaughan, Brian K. and Fiona 220; inner 203–04, 208–09, 213,
Staples) 92–3 218; monologue 102, 207, 240n1;
salience: narrative salience 43, 58, 60, reported 205–06, 207, 218, 242n20,
91–2, 101–02, 105, 106, 107n4; 267; speech-category approach
visual salience 60, 101–02, 105, 106, 203, 204, 217, 219; speech tag 229,
107n4, 115 242n20; see also dialogue in comics
Sandman, The (by Neil Gaiman et al) and free indirect discourse
42, 104–06, 227, 231 speech balloon 1, 13, 18, 101–02,
scene 18, 21, 38, 41, 51, 52, 53–54, 129, 133, 142, 143, 185, 208, 213,
56, 57, 68n9, 93–96, 108n13, 220–21, 228–38, 241–42n19, 248,
108n14, 240; see also dialogue, 249, 251, 258; as metaphor for a
conversational scenes and time speech act 229, 240, 243n20; as
in comics, formulas of rhythmic means of connectivity 233; expressive
variation use 230; para-balloonic phenomena
Schüwer, Martin 4–5, 36–37, 67n6 226, 240, 241n15; relation to
Scott, Mary 196 characters 230–32, 234–36, 237;
sequencing 16, 48–51, 65, 77, 81, rhythmic function 232–38
125n7, 249; chronological 48; Stanzel, F.K. 213
300 Index
strip design 248–50, 262, 264, 265, a moment’ thesis 55; polychrony
266, 268, 272, 273, 274n7 63, 70n42; retrospection 37, 49;
style, see graphic style rhythm 22, 42, 45, 50–56, 57–58,
stylistics 48, 110, 115, 119 59, 63, 67, 69n32, 79–80, 232–37,
subjective point of view, 265; story time 33–37, 50, 53, 56,
see focalisation, point-of-view 60, 63, 64, 68n9, 69n31, 87; stretch
image and subjectivity 53, 54, 56, 143, 236; temporal
subjectivity 5, 109, 110, 165; author order 33–50, 63; temporality in
109; cartoonist 82–83, 86, 131, panel relations 38–45; see also
198; character 109, 146, 156; reading conventions
indicators of 176–77; narrator 83, Tintin 11, 46, 90, 113, 115, 137, 186,
136, 204; techniques of subjective 195, 208–09; L’Ile noire (The Black
focus of perception 165–69; see also Island) 156–57; The Red Sea Sharks
focalisation, point-of-view image 42; Tintin au Tibet 208
symbolia 109–110 Topolino (comic) 180, 194
synchronic reading, see global look transmedial storytelling, see
and reading intermediality and narratology,
transmedial
‘tabular’ reading, see global look and Trondheim, Lewis 29n28, 70n41, 159
reading Töpffer, Rodolphe 99, 205–06, 222,
Tamara Drewe (by Posy Simmonds) 228, 230, 240–41n4, 248; “Essai de
188, 192, 197 physiognomonie” 222
Taste of Chlorine, A (by Bastien Vivès)
38–39, 49, 61–62, 83 V for Vendetta (by Alan Moore and
Tenniel, John 97–100 David Lloyd) 73–74
“There Will Come Soft Rains” Valentina (by Guido Crépax) 36–37
(by Wally Wood) 176 Van Lier, Henri 43–44
thought balloon 54, 104, 109, 143, Verstraten, Peter 11, 89n15
149n16, 185, 208, 213, 218, 220,
228, 229, 238 Walking Dead, The (by Robert
thought representation 28n14, Kirkman, Charlie Adlard, and Cliff
177–78, 188, 204, 211–12, 213, Rathburn) 95–97, 99
220; direct 204, 207, 213; reported Walking with Samuel (by Tommi
76, 206, 207, 208–09, 212, 216, Musturi) 113–14, 122–23, 160
218, 242n20; see also free indirect Ware, Chris 23, 33, 35, 49, 51, 65, 67,
discourse 67n2, 69n25, 150, 176
time in comics 3, 5, 8–9, 20, 33–70; Watchmen (by Alan Moore, Dave
‘achronism’ 37; anachrony 37–38, Gibbons and John Higgins) 42, 193
48, 51, 52; and genre 66–67; Waugh, Colton 13, 107n1
anticipation 37, 64, 68n8, 143; Weapons of Mass Diplomacy (by Abel
chronology 16, 33–35, 36, 37, 40, Lanzac and Christophe Blain)
48, 49, 50; duration 20, 37, 40, 45, 223–25
50–58, 62, 99, 206, 226, 229, 233, Williams, Charles 251, 255
235–57, 240, 253; duration in a Wolk, Douglas 17, 70n34
single panel 55–57, 65; ellipsis 18, Woodward, George 251, 275n19,
53, 107n6; flashback 34, 37, 38, 41, 275n22
55, 68n8, 105, 125n12; formulas wordless comics 15, 38, 40–41, 61,
of rhythmic variation 53–54, 56; 75, 79–81, 83, 88n1, 113–14, 122,
frequency 37, 42, 60–63, 206; 124, 125n16, 153, 160, 161–62,
narrative (or discourse) time 3, 8, 169, 170
33–37, 48, 50, 53, 54, 55, 56, 60, writing: style of 112, 113, 133, 221,
63, 64, 68n9, 69n31, 87; ‘panel is 226–27, 240, 248

Text

Uploaded by

Copyright:

Available Formats

Text

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Text

Uploaded by

Copyright:

Available Formats

The Narratology of Comic Art

By placing comics in a lively dialogue with contemporary narrative

Kai Mikkonen is Associate Professor of Comparative Literature at the

1 Reading Art Spiegelman

2 The Modern Superhero in Film and Television

3 The Narratology of Comic Art

NEW YORK AND LONDON

List of Figures vii

Introduction: Comics, Narrative, and Medium 1

5 Narrative Agency (in Jiro Taniguchi’s A Distant

8 Presenting Minds in Comics 203

10 Picture Story and Narrative Organisation

I.1 “After that it was my turn” © Matti Hagelberg. 1

I first learned to read by reading comics with my father when I was

Figure I.1 “After that it was my turn” © Matti Hagelberg.

Later I have often wondered what it was in these children’s comics,

Narratology and the Narrative Medium

The Beginnings of Comics Narratology

The Narratology of Comics: What Should Be Included?

What Is Narratological Analysis?

Narrative and Graphic Narrative

• spatial limitations (number of pages, page size)

• Iconic (or iconographic) restriction: The exclusion of specified picto­

If we examine in a little more detail how these constraints are conceived,

Outline of This Book

The opposition between erzählte Zeit (story time) and Erzählzeit

It is a kind of commonplace in comics studies to argue that space and

Order, Duration, and Frequency

Temporality in Panel Relations

Figure 1.2 Extrait de l’ouvrage. A Taste of Chlorine (2011). Bastien Vivès

by provoking the readers to make connections, is fundamental to com­

in certain cases, it [the gutter] may be utterly unremarkable (in the

Figure 1.3 Matti Hagelberg. Kekkonen (2004) © Matti Hagelberg.

We can distinguish between several and not altogether compatible

Layout and Connectivity

• the means of composition and layout

Many formal devices of the composition can suggest a sense of conti­

Duration and Rhythm

Figure 1.4 Fred. Philémon. La mémémoire (1977) © Philémon—tome 11,

pause: NT = n, ST = 0. Thus: NT ∞ > ST

Duration in a Single Panel

The Problem of the Key Frame

Figure 1.5 Fred Philémon. La mémémoire (1977) © Philémon - tome 11,

a singulative a: a narrative tells once what happened once;

Figure 1.6 G uy Delisle. Pyongyang: A Journey in North Korea © 2006

A caption or dialogue can give the accompanying image an iterative

• The perception of meaningful relations between panels, in two di­

The techniques of sequencing and page layout precondition these conven­

Graphic Showing and Style

Narration by visually showing something that evolves from one panel to

Figure 2.1 John Leech. “Substance and Shadow” (1843).

Such a release is evident, for instance, in individual panels, sequences, or

Figure 2.2 H.M. Bateman. “The One-Note Man” (1921).

Figure 2.3 H.M. Bateman. “The One-Note Man” (1921).

The Concept of Monstration

a mode of communicating a story, which consists of showing char­

Theatrical monstration is therefore the product, at least in part, of the

The continuing presence of a character or a group of characters in a

Match on Action in Comics

• Iconic (or iconographic) restriction: The exclusion of specified picto

Figure 1.2 Extrait de l’ouvrage. A Taste of Chlorine (2011). Bastien Vivès

by provoking the readers to make connections, is fundamental to com

Many formal devices of the composition can suggest a sense of conti

Figure 1.4 Fred. Philémon. La mémémoire (1977) © Philémon—tome 11,

Figure 1.5 Fred Philémon. La mémémoire (1977) © Philémon - tome 11,

Figure 1.6 G uy Delisle. Pyongyang: A Journey in North Korea © 2006

• The perception of meaningful relations between panels, in two di

The techniques of sequencing and page layout precondition these conven

a mode of communicating a story, which consists of showing char

Figure 5.1 Graphic Novel Excerpt from PERSEPOLIS 2: THE STORY OF A

Figure 5.2 Taniguchi, Jiro. A Distant Neighborhood. Vol. 1. (2009) © by Jiro

throughout the story. In the best-known autobiographical graphic nar

Figure 5.3 Taniguchi, Jiro. A Distant Neighborhood. Vol. 1. (2009) © by Jiro

Figure 5.4 Taniguchi, Jiro. A Distant Neighborhood. Vol. 1. (2009) © by Jiro

Narrative theory has recently shown increasing interest in the representa

use of speech-act categories in the multimodal and mixed-media environ

Il en coûtait à Charles d’abandonner Tostes après quatre ans de sé

The predominant narrative mode in Flaubert’s novel is what F.K. S tanzel

However, Eisner’s presumption, while it illustrates a basic conven

Figure 10.3 William Heath. “Life of a Soldier.” Northern Looking Glass.

Abbott, Lawrence L. 1986. “Comic Art: Characteristics of a Narrative Medium.”