You are on page 1of 645

Emerging Vectors of Narratology


Contributions to Narrative Theory

Edited by
Fotis Jannidis, Matías Martínez, John Pier,
Wolf Schmid (executive editor)

Editorial Board
Catherine Emmott, Monika Fludernik, José Ángel García Landa, Inke Gunia,
Peter Hühn, Manfred Jahn, Markus Kuhn, Uri Margolin, Jan Christoph Meister,
Ansgar Nünning, Marie-Laure Ryan, Jean-Marie Schaeffer, Michael Scheffel,
Sabine Schlickers

Volume 57
Emerging Vectors
of Narratology

Edited by
Per Krogh Hansen, John Pier, Philippe Roussin
and Wolf Schmid
ISBN 978-3-11-055378-9
e-ISBN (PDF) 978-3-11-055515-8
e-ISBN (EPUB) 978-3-11-055488-5
ISSN 1612-8427

Library of Congress Cataloging-in-Publication Data

A CIP catalog record for this book has been applied for at the Library of Congress.

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data are available on the Internet at

© 2017 Walter de Gruyter GmbH, Berlin/Boston

Typesetting: fidus Publikations-Service GmbH, Nördlingen
Printing and binding: CPI books GmbH, Leck
♾ Printed on acid-free paper
Printed in Germany
Is narratology consolidating or is it diversifying?
Such was the question debated at the third conference of the European Nar-
ratology Network held in Paris on March 29 and 30, 2013: Emerging Vectors of
The essays collected in the present volume, all of them full-length research
articles developed subsequent to the lectures presented at that conference,
provide no clear-cut answer to this question or to the numerous matters sur-
rounding it. Rather, these in-depth studies are emblematic of the diverse and
sometimes complementary, sometimes conflictual, but always thought-provok-
ing ways in which the contributors frame the issues in today’s research environ-
ment, suggesting that current narratological inquiry is consolidating in some
ways while at the same time diversifying in others. All of the authors draw on
established advances in the field, and all seek to refine existing frameworks or to
venture into areas which, whether already well-known or only nascent, call for
further exploration. This, basically, is the rationale behind grouping the contri-
butions into “Contexts” and “Openings.”
As readers of the following pages will see, the “contexts” of the first section
refer partly to the various theoretical and disciplinary environments and/or nar-
rative objects focused on by the authors, and partly to context in the sense of the
cultural and historical or communicative circumstances within which narrative
discourses occur.
Dan Shen discusses a number of attempts to “contextualize” narrative
poetics, such as feminist narratology, but finds that while this approach enriches
and refines formal narrative poetics, it is not possible to “genderize” structural
distinctions, as such distinctions are by nature decontextualized. As for rhetori-
cal narratology, it stakes out a position which, although not fully appreciated for
a number of years, has considerable potential for the sociohistorical contextual-
ization of narrative theory itself, without any need to challenge the principles of
formal narrative poetics.
In the following essay, Jannike Hegdal Nilssen, in an analysis of Linn Ull-
mann’s Before Your Sleep, examines “ambiguous” discourse, a combination of
free indirect discourse with character-(in)dependent discourse which forms a
hidden layer of narrative unreliability and irony in this Bildungsroman. Whereas

1 Complete information on the conference can be found on the ENN website at: http://narratol-

DOI 10.1515/9783110555158-202
VI   Preface

unreliable narration is frequently in the first person, unreliability in this case

results from the intradiegetic-homodiegetic narrator and extradiegetic-hetero-
diegetic narrator functioning together. Refinement of another type is brought to
unreliable narration by Per Krogh Hansen. Typically associated with fictional
texts, unreliable narration undergoes a novel transformation when it is employed
in autofiction, a genre characterized by the presence of both referentiality and
fictionality and by the blurring of borders between author, narrator and charac-
ter. As attested by Hansen’s corpus, this state of affairs leads to the unexpected
conclusion that the implied author is not necessarily inseparable from unreliable
narration. David Stromberg brings yet another set of considerations to unrelia-
bility. This raises a host of questions about value judgments which, in narrative
contexts, bear on narrative faith: “a conviction that, regardless of its source in
the imagination, literary narrative retains its relevance to both the phenomenal
world and to human life.” Narrative faith often exists in a relation of tension with
narrative doubt: “hesitation between the believability of the illusion and the con-
sciousness of the story as invention.” Valery Timofeev offers a close analysis of
a page from Nabokov’s story “Ultima Thule” (in both Russian and English) that
incorporates unreliable narration into generative narratology, a model he has
developed which is predicated in part on two aspects of authorial self-conscious-
ness drawn from blending theory: self-reflexiveness and internal state awareness.
In another group of articles included in the same section, Natalya Bekhta
argues for the existence of a distinct, first-person plural narrating agent which
defines we-narrative, an emergent form of narrative situation. Nora Berning,
working within the framework of critical ethical narratology, analyzes a corpus of
autobiographical end-of-life stories, a form of factual narrative in the life-writing
mode that makes heavy use of fictional strategies. In Berning’s study, the imme-
diacy of this form of writing is made more salient by the fact that her tutor text,
Wolfgang Herrndorf’s Arbeit und Struktur, is a blog. Manja Kürschner studies how
history is fictionalized in a corpus she describes as post-constructivist metahisto-
riographic fiction—a form, she explains, that “refutes radical constructivist the-
ories that postulate an all-encompassing unreliability of historical writing and
storytelling.” The fragile dividing line between factuality and fictionality is also
taken up by Silke Lahn and Stephanie Neu, although with a transmedial focus.
On the basis of Richard Walsh’s proposal to redefine fictionality as a “rhetorical
mode” and the ensuing intensive Danish research in this field, the authors analyze
two recent and iconic examples in European television series and movies. The
Italian film Romanzo Criminale and the Danish series Klovn (subsequently a film)
mix supposedly irreconcilable elements of fictional and factual origin into the
narrative by including “forensic” elements—real events, archive footage and real
persons as characters—so as to achieve effects of fictionality. Narrative cannot be
Preface   VII

divided up only into nonfictional and fictional, though, for as Brian Richardson
observes, there is a considerable body of “unnatural” or “antimimetic” narratives
that defy verisimilitude through the use of physically impossible time schemes,
endings with multiple conclusions, etc. There is no unified narratology suited to
accounting for all narratives, Richardson contends, and for this reason he calls
for an unnatural narratology to supplement the standard mimetic models.
The next three articles examine aspects of narrative that are less bound to
context properly speaking than they are to phenomena that span narrative in a
broad sense. Stating that causation as such is not adequate for the analysis of
narrative, Göran Rossholm, after reviewing a number of causal theories of narra-
tive, presents a causal expectation model. A candidate for a narrative universal,
this model incorporates external factors (e. g., genre) as well as internal factors,
effect expectations and causal explanation, but it also allows for narrative point
and surprise, two elements essential for the narrativity of a discourse, in what-
ever medium. Wolf Schmid examines the partly related issue of eventfulness.
Events, he claims, are bound to the necessary conditions of factuality or reality
and resultativity; eventfulness, however, results from the degree of relevance,
unpredictability, persistence, irreversibility and non-iterativity through which
events are narratively portrayed. At the same time, Schmid notes that the novelty
of eventfulness is threatened by repetitiveness, thus pitting (in Šklovskij’s termi-
nology) recognition against defamiliarization. On this basis he goes on to discuss
narrative eventfulness in cultural history with reference to Lotman’s “aesthetics
of identity” as opposed to “aesthetics of opposition.” Raphaël Baroni reflects
on the virtualities of possible worlds in narrative theory. In possible worlds
semantics, he observes, opposing statements about the fabula need not always
be ordered hierarchically but can, as in Jorge Borges’s “The Garden of Forking
Paths,” accommodate multiple alternative worlds that cannot be reduced to
mutually exclusive versions. Virtualities can be expressed in the discourse (as in
Gerald Prince’s “disnarrated”), or they can be shaped by the character at the level
of the fabula or formulated by the reader. Unlike the logical structure of fabula,
the dynamics of plots open up (in David Herman’s words) “a dialectical interplay
between narrative and consciousness.”
With the following three contributions, the context is that of non-verbal nar-
ratives. In his article, Jan Alber argues that film is much better suited to depict-
ing character interiority than is commonly assumed. Through facial expressions,
the use of music and associations of characters with other entities, for example,
internal states can be conveyed in ways that are structurally similar to psychonar-
ration in prose narratives. Furthermore, dual-perspective shots, which combine
the subjective vision of a character with an ‘objective’ one, are analogous to free
indirect discourse. Finally, Alber also discusses more immediate cinematic ways
VIII   Preface

of rendering inner lives (through subtitles, captions, enacted mindscreens and

interior monologues at the auditory level) that function like stretches of direct
thought in prose texts. The question of transmedial transposition is taken up by
Małgorzata Pawłowska in her study of the extensive repertoire of musical works
(especially though not only opera) inspired by Shakespeare’s Romeo and Juliet.
The discussion of music and literature relations date back many years, of course,
but Pawłowska, referring to a growing body of research devoted to musical narra-
tology, shows how the properties of media “affordances” and the use of musical
themes (known as “topics”) enable music—particularly in the case of Romeo and
Juliet—to capture emotional content in ways that parallel the Freytag Pyramid
while at the same time following the processuality of narrative as mapped out by
Greimas in his model for the narrative scheme. The broader issue of narrativity
across media is addressed by Matthias Brütsch. In a survey of theories of narrativ-
ity, synthetically grouped into five “positions,” Brütsch analyzes the implications
of these positions for verbal, audiovisual and dramatic forms of narrative, and
he goes on to confront them with narratological models based on the categories
of story/discourse, mediacy and mimesis/diegesis. In the end, he opts for Wolf
Schmid’s “constitutional model,” built up from happenings, story, the narrative
and presentation of the narrative, as a framework for defining narrative.
Finally, with a switch to cultural contexts far removed from those of the
modern Western corpuses habitually taken into account by narratologists, the
parameters of inquiry may change significantly. Such is the case with ancient
Greek narrative which, according to Claude Calame, is less a literature per se
than it is a “poetic form of action” or an instance of “ritualized enactive speech.”
Required for such a corpus, he contends, is an “interpretive understanding of
intercultural translation demanded by cultural and social anthropology” together
with discourse analysis working along the lines of French enunciative linguistics
and pragmatics. This leads to a critique of the concept of narrativity as devel-
oped in the classical/postclassical paradigm for narratology which, among other
things, fails to account for the prefigurations, configurations and refigurations of
time as formulated by Paul Ricœur.
When it comes to literary traditions such as that of China, looking in from
the outside calls for a comparative approach. It is in this spirit that Huaiyu
Luo surveys a body of research on Chinese narrative theory, both Western and
Chinese, that got underway during the 1970s. During the 1980s and 90s, a number
of major programs for translating Western literary and narrative theory into
Chinese got underway. It was under this influence that narratological studies by
Chinese scholars has been gathering speed over the past twenty years. Involved
is not only the assimilation of Western research but also a rereading of traditional
Chinese poetics, dating back over more than two thousand years. More recently,
Preface   IX

important work has been done within a properly Chinese context, thus devel-
oping narrative theories which open up perspectives for non-Western forms of
narratological reasoning.
In a number of ways, the contributions included in the second section of this
collection—“Openings”—expand the contexts set out, sometimes implicitly, in
the first section.
This observation is confirmed by the topic of the narrative turn, a line of
inquiry that has been evolving for more than thirty years but that is now undergo-
ing important changes in response to recent developments in the research envi-
ronment. Philippe Roussin shows how the narrative turn, which ushered in an
instrumental conception of narrative and sense making in a departure from the
structuralist preoccupation with form and story structure, resulted in the reintro-
duction of narrative into the social sciences. In his commentary, where the work
of Lyotard, Bourdieu, Bruner and MacIntyre, among others, is discussed, Roussin
stresses the rise in postmodern culture of narrative as a form of knowledge that
rivals science. This tendency is felt particularly in historiography (notably in the
work of Hayden White), but also (more recently) in the rise of storytelling—“the
grand narrative of the present”—in communication, management and politics. In
another commentary on the same topic, Paul Dawson looks at the implications
of the narrative turn for interdisciplinarity. With narratology’s expansion beyond
the literary canon, on the one hand, and the growing web of the narrative turn in
the social sciences, on the other, narratology has been confronted by both centrif-
ugal and centripetal forces, leaving its boundaries ill-defined. Dawson concludes
by noting that narrative does not provide “unified knowledge about narrative as
a fundamental cognitive faculty by which humans make sense of experience,”
but rather that the different ways in which the story/discourse distinction run
through the various disciplines afford possibilities for fruitful methodological
exchanges in interdisciplinary research.
The following three chapters are devoted to some of the numerous cognitive
issues that have been studied by narrative theorists. In the first of these contribu-
tions, it is noted that the silent reading of literary fictions has long been regarded
“as an incorporeal process of mapping written words onto mental representations
and meanings.” This practice tends to make the reading of narrative a cerebral
and abstract exercise, ignoring the various ways in which embodiment affects
narrative sense making. Taking a cue from Monika Fludernik’s notion of experi-
entiality, Marco Caracciolo, Cécile Guédon, Karin Kukkonen and Sabine Müller,
working within the scope of second-generation cognitive approaches, explore the
possible avenues for an “embodied” narratology. Such a theory incorporates cog-
nitive parameters, the representational dimension and interpretive strategies for
projecting the storyworld into cultural and ideological meanings. The possibilities
X   Preface

of a neuro-narratology are taken up by Ralf Schneider. Although such a field does

not exist yet, and although it cannot be expected to explain how neural activity
might contribute to an adequate understanding of the production and reception
of narratives, neuro-narratology, nourished by neurobiological data, does have
the potential to provide “a heightened awareness of the richness and complexity
of mental processes below the level of consciousness.” Finally, Richard Walsh
questions the true scope of the cognitive sciences in narrative theory. Taking as
his example a close analysis of Alain Robbe-Grillet’s Jealousy, he argues that nar-
rative cognition, dependent on the serial logic of temporal sequence, and spatial
cognition are incommensurable and consequently that narrative cognition does
not suffice to comprehend the spatiotemporal reality we inhabit; on this basis,
Walsh calls into question any totalizing global conception of storyworlds as well
as the pertinence of immersion for the definition of fictional worlds, which he con-
ceives as mental models, not as extra-representational referents. Taken together,
these three chapters address only a small portion of cognitive narrative studies,
yet they are indicative of several important tendencies in this field of research.
Among the questions studied in the final four articles are a number of broad
conceptual frameworks within which at least some narratological reasoning cur-
rently seems to be developing. Eva Sabine Wagner scrutinizes the conditions for
narratological self-consolidation. She challenges the view according to which
narratology is built up simply by defining its object of research, arguing that
there is a feedback loop between narratology and narrativity. In view of this cir-
cular dynamics, the goal of narratological consolidation cannot be achieved by
attempting to provide narratology with a supposedly “solid” conceptual basis,
but only by exploring narratology’s “complex” processes of self-organization.
Wagner contributes to this project by analyzing, on the levels of both narrativity
and narratology, a specific dynamics of “coherence in progress” through nego-
tiations between syntagmatic and paradigmatic modes of self-organization. For
John Pier, complexity does not serve as a paradigm for narratology so much as
it offers a transdisciplinary perspective for viewing narratological principles in
terms of their possible degrees of commensurability with the loose federation of
complexity sciences. With reference to Ilya Prigogine’s nonequilibrium thermo-
dynamics, Pier shows how narrative sequentiality, whose standard formulations
reflect the laws of Newtonian classical mechanics, might profitably be viewed, in
today’s research environment, in terms of irreversibility or the “arrow of time.”
Doing so supplants equilibrium and disequilibrium as defining features of nar-
rative sequence in favor of the principles of stochastic (or random) processes,
dissipative structures and spontaneous self-organization together with the insta-
bilities engendered by nonlinearity, positive feedback and bifurcations. When
sequentiality is viewed in this way, the properties found at the micro scale of
Preface   XI

narrative are not reduplicated at the macro scale, thus severing the isomorphic
relations between parts and emergent wholes.
Another scientifically inspired approach to narrative is explored by José Ángel
García Landa. Arguing that literary and cultural phenomena are best understood
within a consilient disciplinary framework, García Landa associates this perspec-
tive with “big history,” the broad context of the evolution of societies and of life
generally. In a discussion of the philosophical works of Herbert Spencer, he finds
in nineteenth-century evolutionary theory a historiographical and narratological
perspective which is close to present-day research on natural and ecological con-
textualization of human societies. The concept of “narrative anchoring” is intro-
duced so as to provide temporal schemas with large-scale interpretive contexts
(e. g., the Christian myth from the Creation to Apocalypse) and that of “narrative
mapping” as a framework to account for the historical situatedness of narrative
Finally, Roy Sommer looks into what might be a way forward in sorting out
the multiple theories, models and paradigms that have come to dominate the
narratological landscape. Observing that much has been accomplished to estab-
lish the links between narrative and knowledge, Sommer raises the question as
to how theories of narrative, through their epistemological and methodological
stances, produce knowledge about narrative. Such an exercise in metanarra-
tology, he contends, stems from intuitive knowledge gained through aesthetic
experience and abstract knowledge. From this standpoint, he discerns three
types of compatibility in narratological theory building: “backward” compatibil-
ity (e. g., cognitive approaches integrate various principles of earlier structural
approaches); “forward” compatibility (e. g., research that opens the way to the
transgeneric and transmedial dimensions of narrative); and “sideways” com-
patibility of classical and postclassical narratology, for example, thanks either
to paradigm expansion through the revision of existing theories or to paradigm
evolution, as in the confrontation of structural and rhetorical narrative theories
or of mimetic and antimimetic theories.

Readers of these contributions who are in search of a unified narratological

theory are likely to be struck by the diverse though fruitful avenues of inquiry fol-
lowed by the authors. Yet within this diversity, and on another, more subtle level,
many points of convergence will be found thanks to the prolific dialogue taking
place in today’s research in narrative theory.

 January 2017
 Paris, Kolding, Hamburg
Table of Contents
Preface   V


Dan Shen (Beijing)

“Contextualized Poetics” and Contextualized Rhetoric: Consolidation or
Subversion?   3

Jannike Hegdal Nilssen (Oslo)

Rethinking the Unreliable Narrator: Is the Demarcation Heterodiegetic/
Homodiegetic Necessary?   25

Per Krogh Hansen (Kolding)

Autofiction and Authorial Unreliable Narration   47

David Stromberg (Jerusalem)

Beyond Unreliability: Resisting Naturalization of Normative Horizons   61

Valery Timofeev (Saint Petersburg)

Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology   77

Natalya Bekhta (Giessen)

Emerging Narrative Situations: A Definition of We-Narratives Proper   101

Nora Berning (Gießen)

Critical Ethical Narratology as an Emerging Vector of Narrative Theory and
Autobiographical End-of-Life Stories   127

Manja Kürschner (Kiel)

The Fictionalization of History in Metahistoriographic Fiction after the
Constructivist Challenge   153
XIV   Table of Contents

Silke Lahn (Hamburg) and Stephanie Neu (Mannheim)

Towards a Crossing of the Divide between Fiction and Non-Fiction in European
Television Series and Movies: The Examples of the Italian Romanzo Criminale
and the Danish Klovn   171

Brian Richardson (College Park, Maryland)

Unnatural Narrative Theory: A Paradoxical Paradigm   193

Göran Rossholm (Stockholm)

Causal Expectation   207

Wolf Schmid (Hamburg)

Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   229

Raphaël Baroni (Lausanne)

The Garden of Forking Paths: Virtualities and Challenges for Contemporary
Narratology   247

Jan Alber (Aachen)

The Representation of Character Interiority in Film: Cinematic Versions of
Psychonarration, Free Indirect Discourse and Direct Thought   265

Małgorzata Pawłowska (Cracow)

Intermedial Transposition: From Verbal Story to Music. Narrative in Musical
Works Based on Romeo and Juliet   285

Matthias Brütsch (Zurich)

How to Measure Narrativity? Notes on Some Problems with Comparing Degrees
of Narrativity Across Different Media   315

Claude Calame (Paris)

From Structural Narratology to Enunciative Pragmatics: Greek Poetic Forms
between Mythical Narrative and Ritual Act   335

Huaiyu Luo (Beijing)

Comparison of Chinese-Western Narrative Poetics: State of the Art   361
 Table of Contents   XV


Philippe Roussin (Paris)

What is Your Narrative? Lessons from the Narrative Turn   383

Paul Dawson (Sydney)

How Many ‘Turns’ Does it Take to Change a Discipline? Narratology and the
Interdisciplinary Rhetoric of the Narrative Turn   405

Marco Caracciolo (Ghent), Cécile Guédon (Cambridge, Massachusetts),

Karin Kukkonen (Oslo) and Sabine Müller (Berlin)
The Promise of an Embodied Narratology: Integrating Cognition,
Representation and Interpretation   435

Richard Walsh (York)

Beyond Fictional Worlds: Narrative and Spatial Cognition   461

Ralf Schneider (Bielefeld)

Is There a Future for Neuro-Narratology? Thoughts on the Meeting of Cognitive
Narratology and Neuroaesthetics   479

Eva Sabine Wagner (Osnabrück)

In Search of Coherence: Tacit Negotiations between the Paradigmatic and the
Syntagmatic in Narratology and Narrativity   497

John Pier (Tours and Paris)

Complexity: A Paradigm for Narrative?   533

José Ángel García Landa (Zaragoza)

The Story behind any Story: Evolution, Historicity and Narrative
Mapping   567

Roy Sommer (Wuppertal)

The Future of Narratology’s Past: A Contribution to Metanarratology   593

Notes on Contributors   609

Index   615
Dan Shen (Beijing)
“Contextualized Poetics” and
Contextualized Rhetoric:
Consolidation or Subversion?
The relation between form and history has been a hot topic for debate in the field
of narratology since the 1980s. Classical narrative poetics, because of its decon-
textualization, has been criticized by contextualist approaches. Rhetorical nar-
rative theory, although figuring as a postclassical approach since the 1990s, has
likewise been criticized for neglecting sociohistorical context. But if we examine
contextualist narratological challenges to formal narrative poetics, we may find
that the efforts to contextualize poetics actually come up with decontextualized
structural distinctions, since the investigation of generic structures requires, by
nature, leaving aside varied specific contexts.1 In the case of rhetorical narrative
theory, contextualist challenges also function to bring into play the historicizing
potential in the theory itself. That is to say, the contextualist challenges function
to consolidate rather than to subvert the fundamental principles of formal narra-
tive poetics and rhetorical narrative theory. I will start by considering the relation
between poetics and contextualization.

1 “Contextualized” Poetics2
Although the situation varies in different countries or with different perspec-
tives, many accounts of the development of narratology tell stories of evolution
either from structuralist narratology to poststructuralist narratology (Currie 1998;
Onega and García Landa 1996), from classical narratology to postclassical nar-
ratology (Herman 1999), from structuralist narratology to cultural and historical
narratology (Nünning 2000), from “a strictly formalist poetics” to a “contextu-
alist narratology” (Darby 2001, 829) or from formal investigation to pragmatic,
gender-oriented and ideological investigations that go “beyond form” (Fludernik

1 I use “generic” in the sense of pertaining to the narrative genre or to a specific genre of nar-
2 This section draws on Shen “Why Contextual and Formal Narratologies Need Each Other”
(2005). I am grateful for permission from the editors to reprint here relevant materials in that

DOI 10.1515/9783110555158-001
4   Dan Shen

2005). These stories vary, some even differ in nature, but one idea is shared in
common: the decontextualized formal investigation of generic structures has
been and should be abandoned, and narratologists should always take contexts
into account. However, if we examine respectively narratological theorizing and
narratological criticism—often occurring since the late 1980s in the same narra-
tological study—a different picture emerges. In terms of narratological criticism,
the picture is indeed one of evolution from a text-based investigation subject
to formalist limitations to a more valid and fuller investigation that takes into
account contexts and readers. In terms of narratological theorizing, however,
the picture is essentially different. Postclassical or contextual narratologies have
greatly enriched narratological theorizing in various ways (for a good survey, see
Fludernik 2005), but when the investigation is concerned with generic textual
structures and their generic functions, there is usually no room or no need for the
consideration of varied specific contexts.
In a sense, a narrative is analogous to a sentence. It is true that, in inter-
preting the meaning of sentences in specific utterances, we need to consider the
contexts in which the relevant sentences are uttered. But in classifying structural
elements such as syntactic subject, predicate and object in a grammar, we do not
need to consider the specific contexts in which the relevant language elements
are expressed. Instead, we need to lift the relevant language elements (as struc-
tural illustrations) “out of their contexts in order to distill from them the essential
structures” concerned (Warhol 1989, 4). Indeed, in classifying sentential/clausal
structures in a grammar, usually only decontextualized hypothetical language
elements are used as illustrations. This is the case even in M. A. K. Halliday’s
(2004 [1994]) functional grammar.
Because structural distinctions are necessarily decontextualized, the trend
towards contextualization over the past several decades have not affected clas-
sical structural distinctions such as that among heterodiegetic, homodiegetic
and hypodiegetic narration, or that among direct discourse, indirect discourse,
free indirect discourse and free direct discourse, or that among various modes of
focalization. Such decontextualized structural distinctions have been employed
continuously in postclassical narratologies as useful tools in contextualized nar-
ratological criticism. Moreover, formal narrative poetics (in the shape of new-
ly-established decontextualized structural concepts and models) has appeared
continuously in contextualist narratologies.
Let’s take feminist narratology for an example. Feminist narratology has
played a pioneering and a most significant role in consolidating and enrich-
ing formal narrative poetics. It came into being in North America in the 1980s
when structuralist narratology was very much excluded from the scene by the
joint forces of deconstructive and sociohistorical approaches. Under such cir-
 “Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   5

cumstances, feminist narratologists helped to save formal narrative poetics by

combining narratology with feminist criticism. While defending structuralist
narratology, feminist narratology, as a contextualist approach, is unequivocally
critical of formal narrative poetics. The criticism centers on two related issues:
one, being gender-blind, and the other, decontextualization (see, for instance,
Lanser 1981, 39).
But a close look at the theoretical distinctions feminist narratologists propose
will reveal that the investigation of generic structures, in striking contrast with
narratological criticism, defies, by nature, the consideration of specific soci-
ohistorical contexts. Let’s first consider the distinction made by Lanser between
public and private narration. In her influential essay “Toward a Feminist-Narra-
tology,” Lanser says,

By public narration I mean simply narration (implicitly or explicitly) addressed to a nar-

ratee who is external (that is, heterodiegetic) to the textual world and who can be equated
with a public readership; private narration, in contrast, is addressed to an explicitly desig-
nated narratee who exists only within the textual world. (1986, 352)

Not surprisingly, Lanser’s distinction is very much decontextualized and gen-

der-indifferent. Indeed, as far as such structural classifications themselves are
concerned, diversified specific contexts only form irrelevant digressions. As
Lanser points out, “a major benefit of narratology is that it offers a relatively
independent (pre-textual) framework for studying groups of texts” (1986, 346).
While the investigation of groups of texts or specific texts as communicative acts
needs to take account of the varied contexts, the establishment of the relatively
independent or pre-textual framework necessarily requires lifting texts (as struc-
tural illustrations) “out of their contexts in order to distill from them the essential
structures” concerned.
Interestingly, Lanser’s study of women’s texts has led her to formalize and
decontextualize “sex” as a narratological category. In “Sexing the Narrative,”
Lanser writes,

Written on the body leads me to recognize that sex is a common if not constant element of
narrative so long as we include its absence as a narratological variable. Such an inclu-
sion allows us to make some very simple formal observations about any narrative: that the
sex of its narrator is or is not marked and, if marked, is marked male or female, or shifts
between the two. […] One might well classify heterodiegetic and homodiegetic narra-
tives according to their marking or non-marking of sex and according to the ways in
which sex gets marked: overtly, through explicit designation, or covertly, through con-
ventional aspects of gender that suggest but do not prove sex. (1995, 87; original emphasis,
boldface added)
6   Dan Shen

This theoretical distinction of “sex” is as formal and decontextualized as classical

structural distinctions. In the case that the narrator’s sex is unmarked or marked
only covert, the readers’ inferring of the narrator’s sex may vary from individual
to individual or from context to context, but the theoretical distinction between
“marked” and “unmarked” or between “covertness” and “overtness” has to be
made in an abstract and decontextualized way. This case lends strong support to
the point I have been driving at, namely, that the classification of generic struc-
tures defies, by nature, contextualization, or that it requires, by nature, leaving
aside the consideration of sociohistorical contexts. In a similar fashion, we
could formalize the narrator’s race, class, religion, ethnicity, education, marital
status or sexual preference, all of which can be either “marked” or “unmarked,”
and, if marked, can be marked either “overtly” or “covertly” in the text. Once
an attempt is made to theorize those non-structural elements (sex, race, class,
religion, ethnicity, etc.) as “narratological” categories, it also becomes necessary
to lift the texts out of their contexts and to distill from them the distinguishing
properties concerned. Such non-structural elements, that is to say, cannot enter
the realm of poetics unless they are transformed into decontextualized formal dis-
tinctions. Chatman observes that contextualist narratologists “argue for the need
to inquire into the intentions, motivations, interests and social circumstances of
real authors and audiences. Failure to make this kind of inquiry, they believe,
dooms narratology to a treatment of narrative as a ‘detached and decontextu-
alized entity’” (1990, 314). But as far as the investigation of generic structures is
concerned, there is, in effect, nothing wrong with decontextualization, since this
is the only possible way to do it—even in the case of “sex.”
Significantly, feminist narratology functions to enrich narrative poetics from
a specific angle. A most valuable enrichment is found in Lanser’s discussion of
“communal voice,” which is “a category of underdeveloped possibilities that has
not even been named in contemporary narratology” (1992, 21). Lanser distin-
guishes three forms of “communal voice”: “a singular form in which one narrator
speaks for a collective, a simultaneous form in which a plural ‘we’ narrates, and
a sequential form in which individual members of a group narrate in turn” (1992,
21). The “communal voice” was previously neglected probably because, “[u]nlike
authorial and personal voices, the communal mode seems to be primarily a phe-
nomenon of marginal or suppressed communities” (1992, 19). Lanser discovered
this mode through her investigation of women’s texts. While the use of this mode
by women writers undoubtedly calls for contextualized analysis, the theoretical
classification of this mode and its different forms (singular, simultaneous and
sequential) requires lifting texts out of their various specific contexts “in order to
distill from them the essential” structural properties concerned. The gender-in-
different and decontextualized structural distinction also leaves room for the
 “Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   7

investigation of “communal voice” in men’s texts. It should be stressed that, as

in the case of “authorial” or “personal” voice (Lanser 1992, 16–21), we recognize
“communal” voice or its “singular,” “simultaneous” and “sequential” forms, not
because of the author’s gender or because of any given sociohistorical context,
but due to the “essential” structural properties involved.
In her Gendered Interventions (1989, 29), Robyn Warhol puts forward the dis-
tinction between “distancing” and “engaging” narrators. The distancing narrator
“provides so much information about the narratee” that the addresee becomes
very clearly defined just like “any character,” which “necessarily places a dis-
tance between the actual reader and the inscribed ‘you’ in the text.” In con-
trast, the engaging narrator “strives to close the gaps between the narratee, the
addressee, and the receiver.” This distinction is decontextualized and agendered
in itself, and it likewise forms a valuable contribution to formal narrative poetics.
Perhaps because of the fact that it is practically impossible to contextualize
and “genderize” structural distinctions, many recent feminist criticisms, in con-
trast to earlier attempts to forge a feminist poetics, have turned to classical narra-
tive poetics for analytical tools. In her more recent “The Look, the Body, and the
Heroine of Persuasion: A Feminist-Narratological View of Jane Austen” (1996),
Robyn Warhol makes use of the classical distinction between story and discourse
and the classical concept of focalization in her contextualized feminist criticism.
In “Gender and History in Narrative Theory,” Alison Case says,

Narratology gives us, among other things, the tools to identify and describe narrative tech-
niques more precisely, and thereby to consider their implications and significance in more
nuanced ways. In recent years, feminist critics have begun to make good use of these tools
to examine the impact of gender ideology on the form as well as the content of literary
narratives: asking, for example, how gender is encoded in the formal structures of novels,
and in the dynamic of reading those structures produce. […] But feminist narratology can
also help us to attend to gendered distinctions within texts, whether they are authored by
men or women—as, for example, by looking at the ways narrative voices are gendered in
homodiegetic narration. (2005, 312; original emphasis)

Case uses the decontextualized and non-gendered structural concepts “paradox-

ical paralipsis”3 and “ambiguous distancing”4 to investigate how narrative voices
are gendered in specific narratives in historical contexts.
Let’s shift attention to another contextualist approach, cognitive narratology.
As distinct from feminist narratology, which sets store by real authors in soci-

3 This concept has been put forward by Phelan (1996, 82–104).

4 This concept is suggested by Case herself in this essay.
8   Dan Shen

ohistorical contexts, the contextualization of cognitive narratology is oriented

toward readers who can be classified into different types (cf. Shen 2005, 155–157).
What is most relevant to the present discussion are two types of readers. One
is what we may call “generic” readers, that is, readers who are equipped with
“narrative competence” (Culler 1975, 113–130) and who share the same narrative
conventions as typically embodied by stereotypic assumptions, expectations,
frames, scripts, plans, schemata or mental models in narrative comprehension.5
The other type consists of flesh-and-blood readers with their particularity and
social identity in specific sociohistorical contexts. When cognitive narratologists
are concerned with narrative poetics or formal description in contrast with actual
reading experiences, only the “generic” type of readers come into play and varied
sociohistorical contexts are necessarily ruled out.
I’ll take David Herman’s influential book Story Logic (2002) as an example.
The book rethinks “narrative as a strategy for creating mental representations of
the world” (5) and emphasizes the necessity of contextualizing formal descrip-
tion. In chapter  9, entitled “Contextual Anchoring,” attention is directed to
“the process by which cues in narrative discourse trigger recipients to establish
a more or less direct or oblique relationship between the stories they are inter-
preting and the contexts in which they are interpreting them” (8). But the recipi-
ents Herman has in mind are generic ones, and the contexts are likewise generic
rather than sociohistorical. Herman focuses on the second-person pronoun “you”
as a special case of person deixis. While the actual occurrences of narrative you
in specific narratives are anchored in sociohistorical contexts, once an effort is
made to classify the different types of you as a theoretical framework of narrative
poetics, we have to lift the texts out of their specific contexts in order to distill
from them the generic structural properties concerned. Herman offers a compre-
hensive classification of five types of you in second-person fictions: (a) general-
ized you, (b) fictional reference, (c) fictionalized (= horizontal) address, (d) apos-
tropic (= vertical) address, (e) double deictic you (2002, 340–345). The different
types are all determined solely by “essential” structural properties. The first two
types—(a) generalized you, (b) fictional reference—are “marked by an uncoupling
of the grammatical form of you from its deictic functions” (2002, (340). Type (a)
is impersonal, generalized, “pseudo-deictic” or “non-deictic” (340). Type (b), in
contrast, refers to the narrator-protagonist by what Margolin (1984) calls “deictic
transfer.” In making the classification, the various examples Herman cites from

5 Narrative genres can be subdivided into more specific genres, such as “second-person fiction”
(see below). The same applies to literary genres, such as prose fiction (realist novels, the fantas-
tic, etc.).
 “Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   9

Edna O’Brien’s novel A Pagan Place (1984) only serve as generic structural illus-
trations, playing the same illustrative role as hypothetical examples.
In Herman’s classification, each type of you has its “generic” function which
is shared by different occurrences of the same type across specific contexts in
this genre, and which is therefore to be distinguished from contextualized signif-
icance. Precisely because Herman’s classification of the different types of “you”
is based on generic textual features without being affected by the diversity of
readers and contexts in the same genre, the classification yields valuable “new
tools for the poetics of second-person fiction” (Herman 2002, 337).
With its capacity for taking account of the interaction among textual cues,
generic conventions and generic interpretive strategies, cognitive narratology has
been widely hailed as an approach superior to the text-based formal narrative
poetics. In Psychonarratology, Marisa Bortolussi and Peter Dixon offer a compre-
hensive summary of various reception-oriented approaches, concluding that “In
all these disciplines, this emphasis on the recipient of narratives can be seen as
the result of a paradigm shift that exposed and transcended the limitations of
purely formalist models” (2003, 2). Moreover, Bortolussi and Dixon take issue
with other reader-oriented scholars for failing to carry out an empirical approach:
“In general, to expand on what might plausibly be attributed to the reader, narra-
tologists and reader-response theorists have generated a hypothetical description
of readers’ knowledge and inferences with little grounding in objective evidence.
[…] The result is a circular kind of logic: The characteristics of a text provide evi-
dence for various narrative competencies, and the existence of a particular com-
petence provides the evidence for a particular characteristic of the text” (2003,
168). What they advocate is to study “actual, real readers, and to ground one’s
analysis of the reading process in empirical evidence on how readers process
narrative forms” (168–169). Interestingly, but not surprisingly, while theoretically
excluding the text-based approach and the “generic reader”-oriented approach,6
both are adopted in Bortolussi and Dixon’s own three-step investigation: “we first
provide a framework for understanding the relevant textual features; then we
discuss some hypotheses for related reader constructions; and finally, we report
some empirical evidence that supports these hypotheses. Each of these aspects is
taken up in turn” (184–185).
While formal narrative poetics is only concerned with generic textual struc-
tures (and their generic functions) with no capacity for dealing with actual
readers’ responses in varied contexts, an empirical approach to varied actual

6 For a discussion of the distinction between “generic reader” and actual readers, see Shen
(2005, 155–157).
10   Dan Shen

readers’ responses, similarly, will be hard put to work out models of generic
textual structures. The failure to see this division of labor clearly is a fundamental
reason underlying many criticisms of formal narrative poetics. While Bortolussi
and Dixon (2003, 177–178) criticize Gérard Genette for failing to take into account
“the type of reader, the nature of the text, and pragmatics of the reading context”
in his theoretical discussion of focalization, their own theoretical discussion of
focalization, not surprisingly, is equally as reader-free and context-free. Their
“psychonarratological approach” has synthesized, “from the relevant scholar-
ship in narratology and linguistics,” three categories as the theoretical frame-
work of focalization: 1) descriptive reference frames, 2) positional constraints
and 3) perceptual attributions (186–189). The first category is further divided into
“relative reference frames” (e. g., “Some times a dog would howl in the distance,”
where we have “perceptual information relative to the location of a potential
perceiver”) and “external reference frames” (e. g., “The lamps shone from the
summits of their tall poles,” where the reference frame is “determined by axes
found in the story world, independent of any potential perceiver”). The second
category, “positional constraint,” is the textual “constraint on the location of an
agent who might have perceived the information,” while the third category, “per-
ceptual attribution,” consists of textual cues “suggesting a perceiver.” Whatever
the modification of earlier models, Bortolussi and Dixon’s classification of textual
features is every bit as decontextualized as classical narrative poetics. What is
crucial is to preserve the borderline between “the objective features of the text”
and “the potentially variable reader constructions” (Bortolussi and Dixon 2003,
198), a borderline that is kept quite clear in the authors” own three-step investi-
gation: “Having discussed some categories of textual features related to percep-
tual information, the important question that should be considered is how such
cues are processed by readers. […] We present here several ideas concerning one
aspect of the representations readers may construct in processing perceptually
salient descriptions. Subsequently, we report some evidence in support of these
ideas” (191, emphasis added).
The three-step investigation well demonstrates the co-validity of, and the
mutually-benefiting relationship among, the different kinds of inquiry: 1) the
decontextualized investigation of generic textual structures, 2) the investiga-
tion of the hypothetical generic reader’s understanding of narrative and 3) the
empirical study of actual readers’ cognitive processes. The first decontextualized
approach, which provides “a stable landing” or “a theoretical bedrock” (Phelan
and Rabinowitz 2005, 1), paves the way for the latter two. The third, that is, the
empirical, verifies or challenges the conclusions drawn by the second, concerned
with hypothetical narrative understanding. Moreover, the hypothetical and the
 “Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   11

empirical can shed light on the limitations of the first, promoting its further
By now, postclassical narratologies have enjoyed a history of two or three
decades, and narrative criticism has long been contextualized. However, attempts
to contextualize narrative poetics have invariably ended up in decontextualized
structural distinctions. But of course, we can investigate the historical or cul-
tural/national context in which a narrative technique came into being (cf. Shen
2005, 152), or trace the history of the specific uses of a narrative technique in,
say, French women’s narrative or African-American narrative. But when it comes
to a theoretical discussion of the structural features of a narrative technique, we
simply have to leave aside the specific contexts and focus on the basic structural
properties of the technique—structural properties that are shared by varied spe-
cific uses of the technique at least in a narrative genre such as second-person
fiction or in the genre of a language such as Chinese narratives (see Shen 2010).
By both using classical poetic concepts and models (which helps classical
poetics to gain current relevance) and proposing new decontextualized structural
distinctions, contextualist narratologies have made significant contributions to
formal narrative poetics. As this discussion has sought to make clear, contex-
tualist narratologists actually help to consolidate and develop formal narrative
poetics, rather than subvert it. But of course, instead of being concerned only
with universal narrative poetics, postclassical narratologists pay more attention
to the narrative structures of specific narrative genres. It could thus be said that
we have minor diversification coexisting with major consolidation.

2 Contextualized Rhetoric7
Although feminist narratology, cognitive narratology and rhetorical narratology
are all influential postclassical approaches, the rhetorical approach has been crit-
icized for failing to take historical contexts into account. This approach has been
developed mainly by the second and third generations of the Chicago School of
criticism. Both within and outside the Chicago School, there have been critical
attempts that try to rectify the decontextualization of the rhetorical approach. The
critics involved take their attempts either as rebellion against or as challenges to

7 This section draws on Shen “Implied Author, Authorial Audience, and Context: Form and His-
tory in neo-Aristotelian Rhetorical Theory” (2013, 140–158). I am grateful for permission from the
publisher to reprint here relevant materials in that paper.
12   Dan Shen

the rhetorical tradition, but I’ll argue and show that these attempts actually func-
tion to realize the contextualizing potential in rhetorical theory itself.
Let me start with the key rhetorical concept of the “implied author,” which
has been widely used and hotly debated in narratological studies. When putting
forward the concept in The Rhetoric of Fiction (1961), Wayne Booth states,

To some novelists it has seemed, indeed, that they were discovering or creating themselves
as they wrote. As Jessamyn West says, it is sometimes “only by writing the story that the nov-
elist can discover—not his story—but its writer [the novelist himself in the writing process],
the official scribe, so to speak, for that narrative.” (Miss West continues: “Writing is a way of
playing parts, of trying on masks, or assuming roles, not for fun but out of desperate need,
not for the self’s sake but for the writing’s sake.”) Whether we call this implied author an
“official scribe,” or adopt the term recently revived by Kathleen Tillotson—the author’s
“second self” (In her inaugural lecture at the University of London, published as The Tale
and the Teller. “Writing on George Eliot in 1877, Dowden said that the form that most persists
in the mind after reading her novels is not any of the characters, but ‘one who, if not the real
George Eliot, is that second self who writes her books, and lives and speaks through them.’
The ‘second self’, he goes on, is ‘more substantial than any mere human personality’ and
has ‘fewer reserves’; while ‘behind it, lurks well pleased the veritable historical self secure
from impertinent observation and criticism’” [1959, 22])—it is clear that the picture the
reader gets of this presence is one of the author’s most important effects. However imper-
sonal he may try to be his reader will inevitably construct a picture of the official scribe who
writes in this manner. […] Just as one’s personal letters imply different versions of oneself,
depending on the differing relationships with each correspondent and the purpose of each
letter, so the writer sets himself out with a different air depending on the needs of particular
works. (Booth 1961, 71; emphasis added)

If we examine Booth’s words carefully, including those between the two sets of
brackets (taken from footnotes 7 and 8), we will find that the “implied author”
is no other than the writer of the text—“the official scribe who writes in this
manner,” or “the writer [who] sets himself out with a different air,” or the “second
self who writes” the text through “playing parts,” “trying on masks, or assuming
roles,” or again “that second self who writes her books, and lives and speaks
through them.” The difference between the implied author (the second self) and
the real author (the first self) is that between the person assuming a certain air or
adopting a particular stance when writing the text and the same person in daily
life out of the writing process. If we turn our attention from the encoding process
to the decoding process, the implied author is the textual image of this writer for
the reader to infer:

Encoding Process:
the IA = the person writing in a certain manner, making all the textual choices
“Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   13

Decoding Process:
the reader infers from all the choices made by the IA (= the person writing in a certain
manner) the image of the IA (= the person who has written the text in a certain manner)

While it is biographical materials that we need to read to know the real author
(or the flesh-and-blood person),8 what we need in order to know the implied
author of a text is to read the particular text itself, for the implied author’s image
is implied by the textual choices he or she made when authoring the text.
As I analyzed in detail in my paper “What is the Implied Author?” (2011),
Booth’s frequent metaphorical expression of the real author’s “creating” the
implied author, coupled with his putting much more emphasis on the decoding
process in his earlier work, has led to the assumption that the writer of the text
is the “real author” who, when writing, literally creates the “implied author.” So
we have the division of labor that “the historical author writes […]; the implied
author means […]; the narrator speaks” (Nelles 1993, 22; Kindt and Müller 2011,
68). In this division of labor, the implied author tends to be confined within the
text, even reduced to a mere “semantic relation” (Kindt and Müller 2011, 69).
And the epithet “implied” is taken to be an antonym of “historical,” “real,” or
“actual,” thus depriving the implied author of any historicizing potential. This
fundamental decontextualization of the implied author has also played a role in
the decontextualization of its counterpart, “implied reader” or “authorial audi-
ence,” another key rhetorical concept that, in fact, also requires the considera-
tion of the relevant historical context (see below).
It is significant that, although Booth’s implied author has a textual emphasis,
it forms a key element in Booth’s revision of the text-oriented position of the first
generation of Chicago critics, as represented by R. S. Crane. While Booth aims at
distinguishing the role-playing writer of a text from the same person in daily life
and in writing other texts, Crane is concerned with the isolation of the text from
its very writer. Crane’s method is

one which depends on the analytical isolation of works of art, as finished products, from the
circumstances and processes of their origin. It is better fitted to explain those effects which

8 The term “real author” is potentially misleading, since the person outside the writing process
is referred to as “author.” In “The Resurrection of the Implied Author” (2005), Booth consistently
uses “the FBP” (the flesh-and-blood person) or the FBP plus the person’s name, instead of “the
real author.” In fact, the epithet “real” and “flesh-and-blood” could be potentially misleading
too, since the implied author is also a “real” and “flesh-and-blood” person. What Booth wanted
to distinguish is the person in the “role-playing” writing process from the same person in daily
life, out of the “role-playing” writing process (see Shen 2011, 2013a).
14   Dan Shen

would be specifically the same in any other work, of whatever date, that was constructed in
accordance with the same combination of artistic principles than those effects which must
be attributed to the fact that the work was produced by a given artist (1952, 92; emphasis

Compare the following observation by Booth (quoted above):

Just as one’s personal letters imply different versions of oneself, depending on the differing
relationships with each correspondent and the purpose of each letter, so the writer [the
implied author] sets himself out with a different air depending on the needs of particular

In contrast with Crane’s emphasis on the same effects “in any other work,”
Booth’s emphasis is on the contrast among the different textual norms created
by the different implied authors “depending on the differing relationships with
each” targeted type of reader and “the purpose of each” text. While in Crane’s
poetic theory we lose sight of the writer and only have in view a timeless and
autonomous text, in Booth’s rhetorical theory it is the communicative or rhetor-
ical purposes of the role-playing writer that form the focus of attention. Crane’s
work is very important and valuable in redressing the long-term neglect of poetic
form, but he has gone too far in decontextualization, leaving the author to work
in a realm above and beyond social and cultural history.
However, I do not mean to suggest that, when putting forward the concept of
the “implied author,” Booth had in mind both a textual emphasis and a historical
requirement. Booth followed the ahistorical position of the first generation of the
Chicago School when writing The Rhetoric of Fiction in the mid-twentieth century,
a time marked by the reign of formalism. But as discussed above, in shifting from
textual poetics to author-audience rhetoric, Booth significantly revised the rela-
tion between the author, text and reader. In contrast with the first generation of
the Chicago School’s treatment of textual effects as ‘authorless’ (“the same in
any other work”), autonomous and timeless (“of whatever date”), Booth empha-
sized the rhetorical purposes and overall textual design of the implied author of a
given text who has in mind a particular type of audience when making the textual
choices for producing specific rhetorical effects. Since the implied author writes
in a given historical period and his or her textual choices may thus be influenced
by historical factors, and since the type of audience the implied author has in
mind is one with knowledge of the relevant historical factors, Booth’s shift from
poetics to rhetoric enables the theory to take on a contextualizing potential. In
other words, rhetorical theory’s requirement for a correct understanding of the
implied author’s rhetorical purposes and textual norms is at the same time an
implicit requirement that the relevant contextual factors be taken into account.
“Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   15

In the poetic theory of the first generation of Chicago critics, as represented

by R. S. Crane, we not only lose sight of the particularity of the writer and only
see a timeless and autonomous text, but we also lose sight of the particularity
of the reader. Indeed, in Ralph Rader’s words, for Crane and Sheldon Sacks, “a
twentieth-century reader, taking Tom Jones from a drugstore rack, could find
himself in immediate contact with its moving aesthetic force, that is to say, with
the essential meaning and value of the novel” (Rader 1999, 49). The case is fun-
damentally different with the rhetorical theory of the second and third genera-
tions of the Chicago critics, as represented by Wayne Booth, James Phelan and
Peter J. Rabinowitz. As already pointed out, for Booth, the implied author of a
given text “sets himself out with a different air […] depending on” the relationship
with the particular type of reader he or she has in mind, and he designates this
particular type of reader the “implied reader.” In support of his rhetorical posi-
tion, Booth quotes Montgomery Belgion’s words: “Only when the moral beliefs of
the reader tally exactly with those on which a story is based will the reader have
the whole of the emotion which it is potentially able to produce in him” (Booth
1961, 118). In the afterword to the second edition of The Rhetoric of Fiction, Booth
(1987 [1983/61], 422–424) subscribes to Rabinowitz’s distinction between “autho-
rial audience” (the implied author’s ideal or hypothetical audience, resembling
Booth’s “implied reader”), “narrative audience” (corresponding to the narrator,
believing that the events of the story are real) and the flesh-and-blood “actual
audience” (Rabinowitz 1977, 126–128). Rabinowitz defines the “authorial audi-
ence” as unequivocally contextualized, stating:
Like a philosopher, historian, or journalist, [the author of a novel] cannot write without
making certain assumptions about his readers’ beliefs, knowledge, and familiarity with
conventions. […] Demby’s The Catacombs, for instance, takes place during the early sixties,
and the novel achieves its sense of impending doom only if the reader knows that John F.
Kennedy will be assassinated when the events of the novel reach 22 November 1963. Had
Demby assumed that his audience would be ignorant of this historical event, he would
have had to rewrite his book accordingly. Since the structure of a novel is designed for the
author’s hypothetical audience (which I call the authorial audience), we must, as we read,
come to share, in some measure, the characteristics of this audience if we are to understand
the text. (1977, 126)

In his book-length study Before Reading, Rabinowitz makes a more comprehen-

sive discussion of the diversified assumptions the implied author has in mind
when writing the text for his or her particular type of authorial audience. Accord-
ing to him, “Some assumptions are historical: Flaubert assumes considerable
knowledge of the revolution of 1848 in Sentimental Education. Some are sociolog-
ical: at least one critic has argued convincingly that The Turn of the Screw makes
16   Dan Shen

proper sense only to a reader who knows something about the conduct deemed
proper to governesses in the nineteenth century” (Rabinowitz 1987, 21).
Just as the actual writer, the role-playing implied author creates the text in
history, and his or her textual choices are often based on contextual information
accessible to readers in that particular socio-historical period. That is to say, the
authorial audience or implied reader the implied author writes to is essentially a
contextualized or historicized audience. When the implied Fielding was writing
Tom Jones in eighteenth-century England, he intended the novel for an autho-
rial audience with the knowledge of “the latitudinarians and eighteenth-century
thought” (Rader 1999, 49). According to Rader (50), they are Fielding’s “like-
minded” contemporary audience, when reading Tom Jones in the twentieth or
twenty-first century France, America or China, we need to take into account the
relevant historical information in order to enter the position of Fielding’s “like-
minded” authorial audience in that socio-cultural context. Similarly, when the
implied Edgar Allan Poe was writing “The Tell-Tale Heart” in nineteenth-century
America, he had in mind an authorial audience well informed of the insanity
debate going on in that historical context, and this historical information is indis-
pensable for entering the authorial audience in order to perceive the ironic under-
current centering on the narrator-protagonist’s self condemnation (cf. Shen 2014,
44–49). In a like manner, when Katherine Mansfield was writing “The Singing
Lesson” in early twentieth-century England, she intended the text for an autho-
rial audience who knew that Victorian England regarded a woman who could
not catch a husband as worthless, a kind of social knowledge necessary for dis-
cerning the covert textual progression as an implicit protest against phallocentric
social forces (cf. Shen 2014, 111–116). In such cases, unless we enter the position
of the implied author’s “like-minded” authorial audience in history, we cannot
gain an adequate understanding of the implied author’s textual choices and rhe-
torical purposes, and there cannot be successful communication between the
implied author and us readers. Seen in this light, the consideration of the histori-
cal context in which a text was produced is not only allowed but also required by
rhetorical narrative theory.
Curiously, this contextual requirement in rhetorical theory has been back-
grounded and very much neglected by many scholars, both outside and inside
the rhetorical camp, due in part to misunderstanding the key rhetorical concept
“implied author” and its counterpart, the “authorial audience.” In fact, Booth
himself has also played an important part in backgrounding the contextual
potential of rhetorical theory in the first edition of The Rhetoric of Fiction. Writing
at the height of formalist criticism, he “arbitrarily isolated technique from all of
the social and psychological forces that affect authors and readers” and thus
failed to consider, let alone mention, the contextual requirement that implicitly
“Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   17

exists in his implied reader. Although much more open to the historicist approach
and heartily praising Bakhtin for his ideological and historical criticism of lit-
erary works in the afterword to the second edition of his The Rhetoric of Fiction
(1987 [1983/61], 414–415), Booth still insisted on his “transhistorical” (not anti-
historical) position (413). But of course, Booth was talking about the nature of
his rhetorical project—one concerned with uses of “rhetoric in fiction” across the
history of the novel, rather than with a historicist project that will make historical
categories such as “Victorian” and “modernist” primary. When Booth introduced
Rabinowitz’s “authorial audience,” he discussed two kinds of difference between
the actual readers and the authorial audience. One is how a reader’s reading of
a text changes in the course of the reader’s development (e. g., “the changes the
years had produced in [Booth’s] reading of Anna Karenina”; 420). The other is
how different actual readers, such as male versus female readers, would come
up with divergent readings of a text. However, this concern with the “transhistor-
ical” results in backgrounding the historical dimension implicit in the “authorial

Rabinowitz’s way of talking about the authorial audience underlines a complicating fact
that my discussion does not make clear: the reader whom the implied author writes [original
emphasis] to can be found as much in the text’s silences as in its overt appeals. What the
[implied] author felt no need to mention tells us who he thinks we’ll be—or hopes we’ll be.
“Demby’s The Catacombs […] takes place during the early sixties, and the novel achieves its
sense of impending doom only if the reader knows that John F. Kennedy will be assassinated
when the events of the novel reach 22 November 1963.” Precisely because the novel remains
silent about this fact, we can infer that members of the “authorial audience” already know
it. The same thing holds for our beliefs about values: what the [implied] author feels no
need to mention, of the values the story depends on, tells us who he thinks we are before we
start to read. (Booth 1987 [1983/61], 422–423; emphasis added, except where noted)

Here we can see the essential similarity between Booth and Rabinowitz in terms
of the historical requirement: both regard the historical knowledge of John F.
Kennedy’s assassination on November 22, 1963 as a prerequisite for entering the
position of the authorial audience of Demby’s The Catacombs. But instead of
stressing that readers must have or gain specific historical knowledge in order
to enter the authorial audience (in which case the historical requirement will be
foregrounded), Booth assumes that readers all know this historical event “before
[they] start to read.” Booth does not consider the fact that the authorial audience
the implied author has in mind is often situated in another historical context, and
that when reading the text in a later period or in a different sociocultural context,
readers may need to gain the relevant historical information so as to enter the
position of “the reader whom the implied author writes to.” Needless to say, this
18   Dan Shen

is a point that Booth would wholeheartedly endorse if there were such a lack of
historical knowledge in actual readers.
When Booth was writing the afterword to his book in 1983, what he called
“reader-critics” dominated the scene, and the concern with the textual tech-
niques the implied author uses to persuade the readers was under attack. Booth
took a defensive stance and stressed “the relatively stable audience postulated by
the implied author—the readers the text asks us to become” (1987 [1983/61], 420).
This stress also played a role in backgrounding the contextual requirement of the
“authorial audience.” Indeed, from the 1970s up to the present, many scholars
have believed that “all knowledge is relative to analytical frameworks, epistemo-
logical perspectives, subject positions,” and rhetorical critics have continued to
see the need to argue for the point that “knowledge and understanding can be
shared across frameworks, perspectives, and positions” (Phelan 2005, 183; see
also Phelan 2007). I fully subscribe to the rhetorical argument. But at the same
time, we need to emphasize that readers in different sociohistorical contexts may
not be able to enter the authorial audience and share reading as intended by the
implied author unless they have and consider the relevant historical information.
This contextual requirement is inherent in the notion of authorial audience, but it
has remained underdeveloped in rhetorical theory and largely unacknowledged
by scholars outside the rhetorical field.
With the realization of the contextual demand of the implied author and
“authorial audience,” we can see that the rhetorical narrative theory has a
stronger requirement of considering the historical context of literary creation
than the theories of what Booth calls “reader-critics.” If what matters are only
analytical frameworks, epistemological perspectives or the subject positions of
actual readers, we can ignore the implied author’s rhetorical purposes and the-
matic design in history and be satisfied to subject the text to present-day interpre-
tive frameworks or subject positions. Precisely because rhetorical theory requires
actual readers to find in the text “what the IA wanted them to find” (Booth 2005,
86), actual readers need to take into consideration the historical context in order
to enter the implied author’s like-minded audience in history; otherwise, there
cannot be successful communication between the implied author and us readers.
Interestingly, some challenges to the rhetorical approach from a historical or
cultural perspective implicitly affirm and consolidate the rhetorical contextual
requirement. Within the Chicago School, Ralph Rader is a good case in point.
In his “Tom Jones: The Form in History,” Rader criticizes Crane, Sacks and also
Booth for failing to consider the historical context of literary creation and recep-
tion. Curiously, Rader does not perceive that by shifting from textual poetics to
author-audience rhetoric, Booth’s theory takes on a historicizing potential that
does not exist in the theory of Crane or Sacks. Nevertheless, his historicized
“Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   19

investigation of Fielding’s Tom Jones can be regarded as an explicit realization

of Booth’s implicit historicizing requirement. Although Rader does not use the
rhetorical concepts “implied author,” “authorial audience” or “textual norms,”
what he investigates and reveals is what the rhetorical concepts are (implic-
itly) concerned with, namely, that the implied Fielding wrote Tom Jones for an
authorial audience he considered would be knowledgeable of “the latitudinari-
ans and eighteenth-century thought”; this means that when reading the novel in
twentieth-century America, readers need to take this historical information into
account, if they are to become Fielding’s “like-minded” audience and gain a more
adequate understanding of the textual norms.
Outside the Chicago School, a case in a similar vein is Vera Nünning’s “Unre-
liable Narration and the Historical Variability of Values and Norms,” which is
regarded as a representative work of what Bruno Zerweck (2001, 151) calls the
“second fundamental paradigm shift, one toward greater historicity and cultural
awareness” in the discussion of narratorial unreliability.9 But Nünning’s essay is
in fact an affirmation or consolidation of the contextual requirement in rhetorical
theory. It begins with the following words,

“The history of unreliable narrators from Gargantua to Lolita is in fact full of traps for the
unsuspecting reader.” This statement by Booth has certainly proved to be an accurate pre-
diction. […] Booth’s statement is also relevant in another respect, because the history of the
reception of the individual unreliable narrator is not only a minefield for critics, but for the
unsuspecting reader as well. (2004 [1998], 236)

Unwittingly adopting a rhetorical yardstick, Nünning tries to reveal the various

traps of interpretation—how different historical contexts affect readers’ concep-
tual schema and consequently distort the original meaning. It is in essence a
matter of the failure of actual readers in different historical contexts to enter suc-
cessfully the position of the implied author’s like-minded “authorial audience.”
Nünning claims that “we can at least eliminate one possible trap of interpretation
by taking into consideration the values that were current during the period when
a specific text was written” (248). Formulated in rhetorical terms, the claim could
be expressed as follows: we can at least eliminate one possible trap of interpreta-
tion by trying to enter the position of the authorial audience the implied author

9 The so-called “first paradigm shift” in the discussion of narratorial “unreliability” is from the
rhetorical to the cognitive, which is regarded by many scholars as a desirable replacement of the
rhetorical with the cognitive (see my defense of the rhetorical approach below as well as my more
detailed discussion in Shen 2014 [2009]).
20   Dan Shen

had in mind, a reading position informed of the values that were current during
the period when the implied author made the textual choices.
If Vera Nünning’s historically-oriented investigation of unreliability serves
to consolidate rather than to subvert the rhetorical approach, many cognitive
approaches to unreliability also essentially operate in the same way. The cog-
nitive approach to unreliability arose as a reaction to the rhetorical approach.
Many narratologists take the two approaches to be in conflict with each other
and think that the cognitive should replace the rhetorical. Zerweck, for example,
says, “Within the theory of unreliable narration such a cognitive turn represents a
first paradigm shift. It allows a radical rethinking of the whole notion of narrative
unreliability. Instead of relying on the device of the implied author and a text-cen-
tered analysis of unreliable narration, narrative unreliability can be reconceptu-
alized in the context of frame theory and of readers’ cognitive strategies” (2001,
Significantly, when cognitive critics undertake analysis of narratorial unre-
liability, they themselves often have recourse to the methods of the rhetorical
approach. In Tamar Yacobi’s (1981) foundational essay for the reader-oriented
approach, for instance, we see an implicit shift from the cognitive to the rhetori-
cal stance. After expounding the five reader-oriented mechanisms of integration,
Yacobi narrows down to the issue of unreliability and observes as follows:

To construct a hypothesis as to the unreliability of the narrator is then necessarily to assume

the existence of an implied (and by definition reliable) author who manipulates his crea-
ture for his own purposes. However, the invariability of this rule must not blind us to the
wide variations, from work to work and from passage to passage within the same work, in
all that concerns the modalities of the unreliable source(s) of narration vis-à-vis authorial
communication. (1981, 123)

Although context is taken into account, it is only textual context, and the implied
author still functions as the criterion for measuring narratorial unreliability (“vis-
à-vis authorial communication”), just as in the rhetorical approach.
On the whole, when cognitive narratologists are concerned with actual
readers’ different interpretations of narratorial unreliability, they tend to give up
the authorial yardstick and claim that all interpretations are equally valid; but
when they are discussing unreliability from their own point of view, they often
explicitly or implicitly shift to the rhetorical stance. Ansgar Nünning, another
representative of the cognitive branch, has more recently adopted a synthetic
“cognitive-rhetorical” approach, and he asks the following questions: “What
textual and contextual signals suggest to the reader that the narrator’s reliability
may be suspect? How does an implied author (as redefined by Phelan) manage
to furnish the narrator’s discourse and the text with clues that allow the critic to
“Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   21

recognize an unreliable narrator when he or she sees one?” (A. Nünning 2005,
101; emphasis added). These questions consolidate the rhetorical concern with
unreliability as encoded by the implied author for the implied reader (or readers
who try to enter that position) to perceive.
On the rhetorical side, influenced by Rabinowitz’s distinction among differ-
ent types of audience, many rhetorical critics have paid attention to the different
interpretations of actual readers.10 Although the rhetorical critics have a differ-
ent purpose in mind, namely, to show how the personal experiences and social
positioning of actual readers stand in the way of their entering the position of the
authorial audience/implied reader, their concern with actual readers’ divergent
interpretations implicitly function to consolidate the cognitive concern with the
pragmatic effects of textual phenomena.
In Yacobi’s more recent essay, “Authorial Rhetoric, Narratorial (Un)reliabil-
ity, Divergent Readings” (2005), we see the consolidation of both the rhetorical
criterion of unreliability and the cognitive concern with actual readers. Yacobi’s
analysis shows that, in order to grasp the “authorial rhetoric,” we must try to
enter the position of the authorial audience/implied reader so as to arrive at the
authorial reading. By contrast, “divergent readings” are attributable to the differ-
ences among actual readers and various contexts. So long as a literary narrative
is regarded, in Yacobi’s words, as an act of communication that cannot be defined
without reference to “an implied (and by definition reliable) author who manip-
ulates his creature for his own purposes,” cognitive critics, like rhetorical critics,
still stick to the author rather than shift to actual readers as the yardstick of nar-
ratorial unreliability. But the two approaches do form a contrast with each other
in term of focus of attention: the rhetorical approach centers on communication
between the implied author and the authorial audience (readers who try to enter
that position) whereas the cognitive approach tends to concentrate on the dif-
ference in reading strategy, conceptual framework or cultural/historical context
that underlies the divergent readings of actual readers. The point is that there is
no real conflict, but only essential complementarity between the two approaches.

To conclude, the postclassical narratological period is marked by various attempts

to connect form with context. There is no doubt that we need to investigate the
contextualized significance of narrative form in texts as communicative acts and
that we need to trace the historical development of narrative structures. But in
terms of narrative poetics, there is virtually no room for considering specific

10 See e. g., Booth (1987 (1983/61): 420–423) and Phelan (1996: 100–102 and 167–170; 2005: 5;
2007: 183 and 129–131).
22   Dan Shen

contexts. On the surface, we have a rectifying evolution from formal narrative

poetics to contextualized feminist poetics, cognitive poetics or other contextu-
alized poetics. But as far as the investigation of generic structures is concerned,
we have to leave aside varied specific contexts and focus on the decontextualized
structural properties shared by specific uses in narrative texts of the same genre,
whether the genre is narrative or a type of narrative. As discussed above, feminist
poetics and cognitive poetics are as decontextualized as classical poetics, and
they function to consolidate and enrich, rather than to subvert, formal narrative
poetics. In the case of rhetorical narrative theory, although it has been criticized as
a decontextualized approach, contextualist challenges actually function to bring
into play the contextualizing potential of rhetorical narrative theory itself. More-
over, there is no real conflict between the rhetorical and the cognitive approaches
to unreliability. That is to say, the picture is not, finally, as controversial or diverse
as it has appeared. In various contextualist approaches, form and history can
exist harmoniously in a similar relation between decontextualized poetics and
contextualized criticism; and in rhetorical narrative theory, form and history can
in fact enjoy a balance, since the theory has in essence a textual emphasis and a
historical emphasis.

Works Cited
Booth, Wayne C. 1961. The Rhetoric of Fiction. Chicago: University of Chicago Press.
Booth, Wayne C. 1987 [1983/61]. “Afterword” to The Rhetoric of Fiction. 2nd edition, 401–457.
Harmondsworth: Penguin.
Booth, Wayne C. 2005. “Resurrection of the Implied Author: Why Bother?” In Phelan and
Rabinowitz, eds., 75–88.
Bortolussi, Marisa, and Peter Dixon. 2003. Psychonarratology. Cambridge: Cambridge
University Press.
Case, Alison. 2005. “Gender and History in Narrative Theory: The Problem of Retrospective
Distance in David Copperfield and Bleak House.” In Phelan and Rabinowitz, eds., 312–321.
Chatman, Seymour. 1990. “What Can We Learn from Contextualist Narratology?” Poetics Today
11 (2): 309–328.
Crane, R. S. 1952. “The Concept of Plot and the Plot of Tom Jones.” In Critics and Criticism,
edited by R. S. C., 62–93. Chicago: University of Chicago Press.
Culler, Jonathan. 1975. Structuralist Poetics: Structuralism, Linguistics, and the Study of
Literature. London: Routledge & Kegan Paul.
Currie, Mark. 1998. Postmodern Narrative Theory. New York: St. Martin.
Darby, David. 2001. “Form and Context: An Essay in the History of Narratology.” Poetics Today
22 (4): 829–852.
Fludernik, Monika. 2005. “Histories of Narrative Theory (II): From Structuralism to the Present.”
In Phelan and Rabinowitz, eds., 36–59.
“Contextualized Poetics” and Contextualized Rhetoric: Consolidation or Subversion?   23

Genette, Gérard. 1980. Narrative Discourse: An Essay in Method. Translated by Jane E. Lewin.
Foreword by Jonathan Culler. Ithaca, NY: Cornell University Press.
Halliday, M. A. K. 2004 [1994]. An Introduction to Functional Grammar. 3rd edition. Revised by
Christian Matthiessen. London: Routledge.
Herman, David. 1999. “Introduction: Narratologies.” In Narratologies: New Perspectives on
Narrative Analysis, edited by D. H., 1–30. Columbus: The Ohio State University Press.
Herman, David. 2002. Story Logic: Problems and Possibilities of Narrative. Lincoln and London:
University of Nebraska Press.
Kindt, Tom, and Hans-Harald Müller. 2011. “Six Ways Not to Save the Implied Author.” Style 45
(1): 67–79.
Lanser, Susan Sniader. 1981. The Narrative Act: Point of View in Prose Fiction. Princeton, NJ:
Princeton University Press.
Lanser, Susan Sniader. 1986. “Toward a Feminist Narratology.” Style 20 (3): 341–363.
Lanser, Susan Sniader. 1992. Fictions of Authority: Women Writers and Narrative Voice. Ithaca,
NY: Cornell University Press.
Lanser, Susan Sniader. 1995. “Sexing the Narrative: Propriety, Desire, and the Engendering of
Narratology.” Narrative 3 (1): 85–94.
Margolin, Uri. 1984. “Narrative and Indexicality: A Tentative Framework.” JNT: Journal of Literary
Semantics 13 (3): 181–204.
Nelles, William. 1993. “Historical and Implied Authors and Readers.” Comparative Literature 45
(1): 22–46.
Nünning, Ansgar. 2000. “Towards a Cultural and Historical Narratology: A Survey of Diachronic
Approaches, Concepts, and Research Projects.” Anglistentag 1999 Mainz Proceedings,
edited by Bernhard Reitz and Sigrid Rieuwerts, 345–373.Trier: WVT.
Nünning, Ansgar. 2005. “Reconceptualizing Unreliable Narration: Synthesizing Cognitive and
Rhetorical Approaches.” In Phelan and Rabinowitz, eds., 89–107.
Nünning, Vera. 2004 [1998]. “Unreliable Narration and the Historical Variability of Values and
Norms: The Vicar of Wakefield as a Test Case of a Cultural-Historical Narratology.” Style 38
(2): 236–252.
Onega, Susana and José Ángel García Landa. 1996. Narratology. London: Longman.
Phelan, James. 1996. Narrative as Rhetoric: Technique, Audiences, Ethics, Ideology. Columbus:
The Ohio State University Press.
Phelan, James. 2005. Living to Tell about It: A Rhetoric and Ethics of Character Narration. Ithaca,
NY: Cornell University Press.
Phelan, James. 2007. Experiencing Fiction: Judgments, Progressions, and the Rhetorical Theory
of Narrative. Columbus: The Ohio State University Press.
Phelan, James, and Peter J. Rabinowitz. 2005. “Introduction: Tradition and Innovation in
Contemporary Narrative Theory.” In Phelan and Rabinowitz, eds., 1–16.
Phelan, James, and Peter J. Rabinowitz, eds. 2005. A Companion to Narrative Theory. Malden,
MA: Blackwell.
Rabinowitz, Peter J. 1977. “Truth in Fiction: A Reexamination of Audiences.” Critical Inquiry 4
(1): 121–141.
Rabinowitz, Peter J. 1987. Before Reading: Narrative Conventions and the Politics of
Interpretation. Ithaca, NY: Cornell University Press.
Rader, Ralph W. 1999. “Tom Jones: The Form in History.” In Ideology and Form in Eighteenth-
Century Literature, edited by David H. Richter, 47–74. Lubbock: Texas Tech University
24   Dan Shen

Shen, Dan. 2005. “Why Contextual and Formal Narratologies Need Each Other.” JNT: Journal of
Narrative Theory 35 (2): 141–171.
Shen, Dan. 2011. “What is the Implied Author?” Style 45 (1): 80–98.
Shen, Dan. 2013. “Implied Author, Authorial Audience, and Context: Form and History in
Neo-Aristotelian Rhetorical Theory.” Narrative 21 (2): 140–158.
Shen, Dan. 2014. Style and Rhetoric of Short Narrative Fiction: Covert Progressions Behind
Overt Plots. New York and London: Routledge.
Shen, Dan. 2014 [2009]. “Unreliability.” In Handbook of Narratology, 2nd edition, edited by
Peter Hühn, Jan Christoph Meister, John Pier and Wolf Schmid, vol. 2: 896–909. Berlin and
Boston: De Gruyter. Also available in the living handbook of narratology at: http://www. (Revised 31 December 2013)
Tillotson, Kathleen Mary. 1959. The Tale and the Teller. London: Rupert Hart-Davis.
Warhol, Robyn. 1989. Gendered Interventions: Narrative Discourse in the Victorian Novel. New
Brunswick, NJ: Rutgers University Press.
Warhol, Robyn. 1996. “The Look, the Body, and the Heroine of Persuasion.” In Ambiguous
Discourse: Feminist Narratology and British Women Writers, edited by Kathy Mezei, 21–39.
Chapel Hill: University of North Carolina.
Yacobi, Tamar. 1981. “Fictional Reliability as a Communicative Problem.” Poetics Today 2 (1):
Yacobi, Tamar. 2005. “Authorial Rhetoric, Narratorial (Un)Reliability, Divergent Readings:
Tolstoy’s ‘Kreutzer Sonata’.” In Phelan and Rabinowitz, eds., 108–123.
Zerweck, Bruno. 2001. “Historicizing Unreliable Narration: Unreliability and Cultural Discourse
in Narrative Fiction.” Style 35 (1): 151–180.
Jannike Hegdal Nilssen (Oslo)
Rethinking the Unreliable Narrator:
Is the Demarcation Heterodiegetic/
Homodiegetic Necessary?

1 Introduction
In their article “The Fifth Mode of Representation: Ambiguous Voices in Unreli-
able Third-Person Narration” (2011), Behrendt and Hansen discuss unreliability
in third-person narration, and they begin by pointing out how almost all discus-
sions of unreliable narration over the last decades have focused on first-person
narrators (219). At first sight, it may seem that this is the path I will also follow, as
Linn Ullmann’s Before You Sleep (2000 [1998]), my tutor text, appears to engage
mainly a homodiegetic narrator. A first-time reader of the novel is likely to think
that the novel employs the plot of a Bildungsroman and that the protagonist
matures and changes throughout the novel. However, uncertainty as to whether
it is the voice of the narrated “I” we hear or that of the retrospective “I” leads to
the question as to whether the expected distance between the narrated “I” and
the retrospective “I” and their merging actually pertains here.
Behrendt and Hansen (2011) discuss unreliability in third-person narration
by exploring what they label ambiguous discourse. In addition to the four dis-
cursive modes of speech representation,1 a fifth mode of representation on the
discursive level is introduced. Ambiguous discourse is the mode of ambiguity and
unreliability resulting from the convergence of free indirect discourse (FID) and
character-(in)dependent discourse (CID). In light of that article, I will investigate
the ambiguity of free indirect discourse in a number of seemingly paraleptic pas-
sages from Ullmann’s novel. While Behrendt and Hansen address the topic of
first-person narration in disguise, this article will explore whether it is possible to
speak of third-person narration in disguise. This leads to asking why the case of
Before You Sleep is not just a regular unreliable character narrator. Could we not
simply speak of the author or implied author as the orchestrator of the construct-

1 The four modes are direct discourse (quoted monologue) (DD), free indirect discourse (nar-
rated monologue) (FID), indirect discourse (psychonarration) (ID) and character-(in)dependent
discourse (where consciousness is represented exclusively through behavior and speech) (CID)
(Behrendt and Hansen, 2011, 228).

DOI 10.1515/9783110555158-002
26   Jannike Hegdal Nilssen

edness of this story?2 My suggestion will involve rethinking the concept of the
traditional demarcation between the heterodiegetic and homodiegetic narrator
To demonstrate how a rethinking of this demarcation is relevant, I will
proceed by looking into three different ways or levels of understanding the novel
Before You Sleep. Each level delves a little deeper and adds meaning to the inter-
pretation of the novel. Briefly, they can be described as the level of the Bildungs-
roman, the level of the unnatural construction of a Bildungsroman and the level
of hidden unreliability and ironizing over the Bildungsroman genre. When dis-
cussing the third level, I will deal with passages that are seemingly paraleptic.
First, however, I will consider the concepts of implied author and Bildungsroman.

2 The Implied Author and the Bildungsroman

Since Wayne Booth coined the term implied author in 1961, there has been much
debate as to whether or not it is a necessary concept and what it entails. Wolf
Schmid (2014 [2009]) gives a detailed presentation of the discussions and what
he calls an impartial definition, a much-needed synthetic approach. Regarding
Gérard Genette’s ambivalent definition, according to which the implied author
cannot be a “narrative agent” (1988 [1983], 148), Schmid points out that Genette
was not really positioning himself far from some of the defenders of the concept.
Schmid’s concluding definition makes much sense to me in that he states the
purpose of the implied author is to help us “describe the layered process by which
meaning is generated” (2014 [2009], 296). According to him, “[t]he presence of
the implied author in the work, above the characters and the narrator and their
associated levels of meaning, establishes a new semantic level arching over the
whole work: the authorial level” (297).
In his call for a resurrection of the implied author in 2005, Booth embellishes
on his original definition of the implied author, earlier defined as “the core of
norms and choices” (Booth 1983 [1961], 74). He calls the flesh-and-blood-per-
son (FBP) author’s activity of masking, meaning that the implied author can be
regarded as how the work conveys the values and norms of a better self, as com-

2 See also Per Krogh Hansen’s contribution to this volume. He considers the unreliability of the
author-narrator in the genre of autofiction without seeking recourse to the implied author. As
my analysis will show, our approaches are quite different, but in my opinion, what they have in
common is that something other than the implied author is responsible for unreliability.
 Rethinking the Unreliable Narrator   27

pared to those of the FBP author (Booth 2005, 78).3 This is an intriguing idea that
might be relevant in terms of how much masking the reader is able to perceive
at each different level of understanding in Before You Sleep. As we will see when
delving deeper into the three levels, the most “unmasked” level is the least ideal
level in terms of possibilities for a positive outcome.
The implied author is a central concept to bear in mind when speaking of
the genre Bildungsroman. According to Mikhail M. Bakhtin, the Bildungsroman
is “the image of man in the process of becoming” (1986 [1936–38], 19). Theorists
have put forth a variety of criteria for what must be prevalent in a novel for it to
be a Bildungsroman, but it is commonly accepted that this process of becoming
involves the development or formation of the protagonist. This suggests that the
implied author’s norms are not consonant with the norms of the protagonist at
the beginning of the novel, but that this changes toward the end. Franco Moretti
even says that as soon as the Bildung has occurred, the story ends:

Unlike the usual nineteenth-century novel, in the classical Bildungsroman the ending and
the aim of narration coincide. The story ends as soon as an intentional design has been
realized: a design which involves the protagonist and determines the overall meaning of
events. (Moretti 2000, 55)

If there is a first-person narrator, and one that is the protagonist as well, the
Bildungsroman will have to include a merging of the two, since the protagonist
gradually becomes as wise and mature as the retrospective first-person narrator.
When mentioning the Bildungsroman, Genette notes that all autobiographical
narratives start out with a difference in age and experience and that this differ-
ence is “inevitably decreasing in proportion as the hero progresses in ‛apprentice-
ship’ to life, […]” (1980 [1972], 253). This means that the implied author’s norms
and those of the narrator will be consonant with those of the protagonist by the
end of the story.

3 For a detailed commentary on these issues, see Dan Shen’s contribution to this volume, espe-
cially section 2.
28   Jannike Hegdal Nilssen

3 The First Level of Understanding:

The Bildungsroman
Having looked at the concepts of implied author and Bildungsroman, I will now
explore what I call the first level of understanding the novel Before You Sleep.
This is the surface level, the easy-to-access one. Judging from a number of the
blurbs for the novel, it is clear that the publisher and the author are seeking to
promote the novel as, among other things, a Bildungsroman. Also, the reliability
of the narrator—explicitly marked as the protagonist Karin—is questioned. The
following is my translation of the Norwegian publisher’s original blurb:

Before You Sleep is the young Karin Blom’s narrative about herself and her large, strange
family. In Karin’s world, imagination and reality glide almost unnoticeably into each other.
She narrates, disturbs, and seduces—while everybody around her irrevocably moves toward
Before You Sleep is a novel that spans from New York in the 1930s to Oslo in the 1990s,
through a variety of fantastical stories, burlesque scenes and silent gravity. In that way, this
is both a novel of growing up and a family portrait through a century. But most of all, it is a
novel about endeavors of love—of marriage, infidelity, family connections, vanity and hope.
(Blurb by Ullmann 2000)

In the current English blurb from the author’s webpage, which is often found to
present the book in online bookstores, the authority of the narrator is explicitly
marked as unreliable:

Through the sublimely unreliable voice of its narrator Karin, Before You Sleep reaches back
from present-day Oslo to Brooklyn in the 1930s to relate the emotional legacies of the Blom
family. Karin is both playful and melancholy [sic]—a serial seductress who defines herself
in contrast to the women in her life: her mother, Anni, alluring, manipulative, and melodra-
matic; her sister Julie, a wife and mother undone by suspicions of her husband’s infidelity;
her aunt Selma, the world’s angriest old woman; and her soldier-grandmother June who
bets that Karin will be the best damn soldier of them all. (Ullmann 2001)

From these two blurbs, Karin is presented as the narrator, telling her and her fam-
ily’s story while also adding fantastical elements and exaggerating the truth. It
even says that she beguiles the reader or is an unreliable voice. Clearly, we should
be skeptical of what she says,4 since none of these claims is in any way a stretch
of interpretation. The novel even has explicit discussions about her habit of lying:

4 Interestingly, the novel’s German title is Die Lügnerin [the liar].

 Rethinking the Unreliable Narrator   29

I don’t mean to brag. Or make things up. Or bite off more than I can chew. But sometimes it’s
actually necessary—to tell it not just like it is—but to tell it a little more than it is.
When I was a child, Grandma and Anni worried that I didn’t tell the truth. They said that
liars would not be tolerated in a community. They said I exaggerated to make other people
like me. [author’s blank line] (132)

I would suggest that the understanding we gain from the blurbs is one of a protag-
onist who tells the story of herself and her family. The reader can perceive that the
novel is entertaining but that it gets more serious toward the end and that Karin’s
character develops as a result of the events that Karin and her family experience
throughout the novel.
A preliminary narratological analysis demonstrates that the first two pages of
the novel constitute the beginning of a frame narrative. It is narrated by a hetero-
diegetic narrator using internal focalization through the protagonist Karin. At the
beginning of the novel’s part one, “Wedding, August 1990” (5), there is a shift to
a homodiegetic narrator who comes across as Karin the character. It appears that
the narrator tells the story about her family to make time pass while—in the frame
narrative—she and her nephew Sander are waiting for his parents to call from
their vacation in Italy, where they are attempting to salvage their marriage. The
larger chunk of the novel contains a complex web of internal and external ana-
leptic as well as internal proleptic stories ranging from the 1930s to the “present”
1990s with various characters from Karin’s family. On the surface, the only devel-
opment in the novel is that the frame narrative starts seriously and that, after a
series of fantastical and seemingly superficial stories, things take a more serious
turn. Karin’s character begins to take form as she starts out by living a carefree
and careless existence until more serious events intervene, culminating when her
sister fails to call to say she is okay and apparently leaving Karin’s nephew to face
the prospect of orphanhood. Up to this point, this could be a classical Bildungs-
roman plot, constituting what I call the first level of understanding the novel. But
where then is the closer connection between the story Karin tells and the frame
narrative? What role do these events play in the formation of Karin’s character?

4 The Second Level of Understanding:

The Unnatural Construction of a Bildungsroman
The reviewers seemed to agree with the blurbs about the content, even though
not everybody was entirely happy with the novel. Some felt that this large variety
of fantastical stories does not add up properly toward the end of the story, so that
30   Jannike Hegdal Nilssen

the novel is mostly for entertainment (Wærp, 1998; Bergström-Edwards, 1999).

Hans H. Skei (1999) claims that there is a lack of focus in the novel while Henning
H. Wærp (1998, 170) criticizes the missing connection between the confident, des-
perate man hunter he sees in the protagonist Karin and the wise reasoner he sees
in the narrator who gives us the stories of her family. In effect, he is critical of
the failure of the acting “I” to merge with the narrating “I,” a merging we should
normally be able to find in a Bildungsroman, where the protagonist becomes con-
sonant with the implied author’s norms.
What in one way seems to be a development might not turn out to be exactly
that. Thus, we must delve into what I call the second level of understanding the
novel. Careful analysis will disclose a sjuzhet with a complex analeptic and pro-
leptic structure, as already mentioned. These disruptions in the chronology are
carefully designed to make it seem like we have a classical Bildungsroman plot.
By using the distinction fabula/sjuzhet rather than story/discourse, I intend to
emphasize the more plot-related connotations that hinge on the term sjuzhet (cf.
Herman 2005; Abbott 2007), as this will prove highly relevant to the constructed-
ness of this novel.5 As the following analysis will show, the sjuzhet is constructed
in such a way as to make it seem that the narrating “I” merges with the implied
author. This configuration can be characterized as “unnatural.”

4.1 An unreliable sjuzhet?

As visualized in table 1, the order of events is far from chronological. The text in
bold lists the few events we can follow in chronological order after the frame nar-
rative. In between these events are quite a few external analepses, internal ana-
lepses and internal prolepses. Also, to the right are events that it is not possible to
place in the fabula with any clear time markers. This massive shuffling of events,
particularly in parts two and four, is not easy to detect. Each of the five parts of
the novel has a name that to some degree signals that the sjuzhet resembles the
fabula, i. e., that the order of events in the two cases is quite similar.6

5 For a definition of the terms, I refer to James Phelan: sjuzhet is defined as “[t]he fabula ren-
dered in a specific narrative discourse; the synthesis of story and discourse” (2005, 218); fabula
is defined as “[t]he what of narrative before it is rendered in a discourse; the sequence of events
in chronological order” (215).
6 This is also part of the reason why it makes sense to speak of a first level of understanding.
 Rethinking the Unreliable Narrator   31

Table 1: Order in Before You Sleep

32   Jannike Hegdal Nilssen

The names in the translated version of Ullmann’s novel are (in this order):
“Wedding, August 1990” (3), “Days, 1990–1997” (91), “America” (165), “Faces”
(213) and “Sander, December 1998” (249).7 From these names, we might gather
that the “America” and “Faces” parts take place after the days of 1997; however, as
table 1 shows, this is not the case. The external analepsis relating what happens
in NYC in the 1930s is quite explicit, as specific dates and historical events are
mentioned several times in this analepsis. The fact that Karin’s mother Anni
travels to NYC the same year as Julie’s wedding is not equally obvious, as the date
is stated at the beginning of part one (5) but narrated in parts three and four. It is
also stated in part one that aunt Selma is 82 years old (22), so that when it is said
in part four that she died at 83, the reader needs to remember her age to know that
her funeral takes place in 1991.
As both of these events come after a largely shuffled account of the days of
1990–97, I will venture that it is unlikely that the casual reader will take note of
this intuitively, i. e., that Anni’s NYC trip and Selma’s funeral actually take place
in 1990 and 1991, respectively. Both parts three and four of the novel are more
serious and in a graver tone than parts one and two, and as I have already estab-
lished above, Karin the character changes from her carefree and careless ways
at the beginning of the novel to become a more responsible and serious adult
toward the end.
The result of this shuffling of the events in the fabula is that the sjuzhet
becomes unreliable. This is a different type of unreliability than that of ambig-
uous discourse, which Behrendt and Krogh Hansen explain this way: it “leads
the reader to overlook story world facts of decisive importance for understanding
the story” (2011, 230). Although you cannot apprehend this shuffling of events by
investigating ambiguity between CID and FID, the effect here is exactly the same.
The narrating instance has constructed the sjuzhet in such a way as to make the
reader overlook certain story world facts. The sjuzhet makes it seem like there is a
causal connection between events even when this is in fact not the case. Actually,
the ambiguities and discrepancies between the sjuzhet and the fabula cause the
reader to overlook the fact that in the story world there is no causal connection
between the events in part three and those portrayed in the following the two parts.
Another example of such misfits can be seen in the way the protagonist’s
encounter with five different men is presented in the sjuzhet. One of Karin’s aims
is to seduce men, but she does so rather half-heartedly, her intent being, appar-

7 In the original Norwegian version, these parts are also marked with roman numbers, I–V, but
in the English version, they are omitted. For simplicity’s sake, I will continue to refer to the parts
as part one, part two, etc.
 Rethinking the Unreliable Narrator   33

ently, only to engage in casual sexual encounters (21). For instance, she lures Billy
to bed by rolling apples after him in the supermarket:

It takes me four weeks to persuade Billy to sleep with me.

I never thought about giving up. I don’t give up once. I’ve made up my mind. That’s the
whole point. But Billy got on my nerves. Billy irritated me. I felt more like slapping him than
sleeping with him.
Regardless, I finally talk him into it.
This is how it happens: […] I take one apple after another out of the cart and roll them
across the floor. Over to Billy. Lots of apples. […]
[…] Suddenly he lets go of his cart and comes toward me. I hide behind the shelves. He
lets go of his cart and comes toward me. I hide behind the shelves. I run away. I hide behind
the shelves. Billy comes toward me and finds me and grabs my arm and says, Let’s go. (95;
original emphasis)

It is hard to believe that this is a realistic account of what happened. Billy is the
second man she seduces in the sjuzhet. The third, Carl, is even harder to believe.
Karin’s encounter with Carl is the most fantastical of the five, as this is the man
who transforms into a fish, a mackerel, the morning after because Karin removes
Carl’s magic cowboy boots against his will while he is sleeping (138–139).
If one looks at the first letters of the names of all five men, one will find that
the order in which they are presented is not random. The sjuzhet presents the
men alphabetically: Aaron in part one; Billy, Carl, and Dag in part two; and Edwin
in part four. This alphabetical order suggests that there is a connection between
these encounters: if we look for a causal connection in how Karin acts around
men, it can be found. For each man she wants to seduce, it seems to get harder,
and the accounts become more unrealistic. Aaron is quite easy (74); Billy takes
four weeks to seduce in the supermarket; and for Carl she puts on a big singing
and dancing show in order to sweep him off his feet: “[…] and I dance Carl like
Carl has never been danced before” (131). The next is even harder, and she needs
to get him drunk unawares before she can have her way: “I bend down, lift Dag
off the floor, sling him over my back, and take him home with me” (163). Doing so,
she takes on a stereotypical male part in terms of cultural gender roles, where the
man is the decisive, unstoppable hunter in the seduction game. Also, we do not
really believe that she would be able to do this, slinging him over her back. The
fifth encounter, with Edwin, is different and suggests that Karin is undergoing
change. At first, he will not notice her, and when he finally does, she walks away.

I’m going now.

Okay, Edwin?
I don’t want you after all.
I don’t want you. (241)
34   Jannike Hegdal Nilssen

The reason might be that this time the seduction got so difficult that she gave
up—a radical change of attitude from the time when she was pursuing Billy. The
reason Edwin will not notice her is that he is playing with his daughter and is
more engaged with his child than he is with Karin. The narrating “I” claims that
she goes away because she perceives that the child does not want her there and
because Karin seducing Edwin would mean forgetting about the child. This is the
message the narrator wants to convey:

The little girl notices it too. He’s no longer with her.

She turns around and sees me.
I don’t want to go home with you, do you understand? her eyes say.
That’s fine. I understand.
Could you leave now so my father can be my father again?
Yes, I’ll go. (240; original emphasis)

One or the other is true: either she does not succeed with the last seduction or she
changes to become less ruthless and thus more mature. So far, this is consistent
with the first level of understanding the novel and exemplifies the Bildungsroman
However, the encounters with these five men are shuffled in a very specific
order. The sjuzhet places them alphabetically, but this order results in an unre-
liably constructed causal plot that does not resemble the fabula. The fabula
order is Aaron, Edwin, Billy, Dag and Carl. On this basis, Karin’s suggested matu-
rity (visible in the Edwin encounter) should already have been apparent in her
encounter with Billy, Dag and the mighty and fantastical Carl. The effect of this
reshuffling of the order of events is that the reader will understand the encounter
with Edwin in light of the others that have already been narrated. This being the
case, the reader can detect a change of course from the narrated “I” and thus
conclude that the plot follows the pattern of the Bildungsroman. Even so, analysis
will reveal that the narrated “I” does not change in the course of the fabula.

4.2 A merging of the retrospective narrating “I” and the

implied author?

Instead of a plot where the character actually changes in the course of her expe-
riences, the point might be that we can only change by narrating our own story.
Through the act of reviewing or “undergoing” her story in the order of the sjuzhet,
it can be said that Bildung occurs. It would seem that this Bildung takes place
because of the order in which the events occur; through careful analysis of the
sjuzhet, however, we can see that it is the act of reviewing or “undergoing” the
 Rethinking the Unreliable Narrator   35

story in the order of the sjuzhet that brings Bildung about. This brings us to a
similar point made by Per Krogh Hansen (2011) in his study of “reversed film” nar-
ratives such as Gaspar Noé’s Irreversible and Christopher Nolan’s Memento. The
events in Before You Sleep, though, are more shuffled than they are directly back-
masked, i. e., episodically reversed. Hansen’s topic in this article is unnatural
narratives, and he concludes with an interpretation that could just as well refer
to Before You Sleep in the sense that it becomes clear that the way we perceive
the reordering of events in Before You Sleep as a coherent succession (first level
of understanding) could be a way of “naturalizing” the “unnatural.” If one were
to understand that Karin’s character does not really change, then the “natural”
would be meaningless and the second level of understanding could be a way of
making sense of the natural by “denaturalizing” it (Hansen 2011, 183). It is in this
sense, along with the more “conventionalized” (in the sense of Alber et al. 2012)
unnatural elements (i. e., fantastical stories that are not realistic), that I perceive
this novel as “unnatural.” Achronological order in itself might not make a dis-
course unnatural, but a sjuzhet that creates such an ambiguity due to the exten-
sive shuffling of events and other features is something I would call an unnatural
A character narrator who changes and matures in the course of telling the
story, and where this changing and maturing is not really visible after a careful
analysis, differs from the traditional idea in which the narrating “I” and the retro-
spective “I” merge. In the following, I will take this point further by trying to iden-
tify a point in the fabula where the retrospective “I” is narrating and has matured.
At what point does this character narrator become endowed with maturity by
telling the story?
The narrating time of the story is not marked very specifically. Part one opens
with the narrator starting the story: “Once upon a time, almost nine years ago”
(5). This suggests that the narrating time is almost nine years after August 27,
1990. It would be reasonable to start out assuming that the narrating time is the
night when Karin is waiting for Julie’s call, since, as I have already suggested,
Karin starts to tell the story to make time pass while waiting. The question is:
would the narrator say “almost nine years” if the narrating time was December
27, 1998? That would mean less than eight-and-a-half years. I think not. Adding
to this mention of “almost nine years” are several narrator comments suggest-
ing that Julie was away for a while at the time of narrating. The most explicit
comment is: “Only once, before she disappeared, did I ever see her cry” (15).8 This

8 Other quotes from Before You Sleep that suggest that time has passed and that Julie does not
exist any longer are the following: 1) “I don’t know whether Julie found Aleksander contemptible.
36   Jannike Hegdal Nilssen

comment requires more knowledge of the situation than what Karin knows at the
time while she is waiting for her sister’s call. In the frame narrative, it is implied
that Julie is missing because Karin and Sander are worried that she has failed to
call them, even though it is not established that she did in fact disappear.
Nevertheless, and although the passing of narrating time is not made
explicit, it can be determined that time has passed since the night of waiting for
Julie’s call. Throughout the entire novel we find ambiguity between the carefree,
beguiling liar and the more serious, responsible truth teller even though, toward
the end of the novel, this ambiguity does diminish. From the beginning, both
the narrating “I” and Karin the character’s actions encourage telling lies as a life
motto, while the implied author’s norms do not seem to concur. As we can see in
the second external analepsis of part one, the narrating “I” presents herself as a
quite un-sympathetic person at age eleven:

That day I understood the difference between a lie that paid (that Pete [Anni’s dog] had
done number one and number two) and a lie that didn’t pay (that I had killed Pete with a
rock and thrown her into Oslo Fjord). A lie that had to be taken back and replaced with the
truth was a useless lie that didn’t pay. But a lie that didn’t have to be taken back, that might,
with time, be replaced by a new lie, but never—never!—with the truth, was a lie that paid.
Since I had now learned the difference between a lie that paid and a lie that didn’t, I decided
to lie as much as possible for the rest of my life. And that’s how it turned out. (37)

Given the narrator’s comment in the last sentence, there is no reason to believe
that the narrator is simply telling about her previous self. It seems that the narra-
tor’s norms are still consistent with this episode, just as carelessness is character-
istic for other narrated events in the novel. This carelessness is counterbalanced
a little later when the narrator, in another external analepsis, tells of the time her
sister Julie attempted to commit suicide:

[Julie] got in Anni’s car and drove like mad toward the city—she’s never been a good driver,
anyone who longs for death should take a ride in the car with Julie, and back then she didn’t
even have a license. Thank God the cops stopped her before anyone got hurt. So, finally she
slit her wrists in the shower, and that’s when Anni found her. Luckily. Because this time she
hadn’t made a single mistake. (85)

I think she did. I did, at any rate” (108); 2) “I’m not going to ask you to believe that Carl had magic
cowboy boots. I might have said something like that to Julie, Torild, and Val Bryn one evening
when we were out on the town. There are lots of things I say after a few too many glasses of one
thing or another. (So if you run into Torild or Val Bryn sometime, and they happen to mention
[…]” (132); 3) “You were much lonelier than me, Julie” (132).
 Rethinking the Unreliable Narrator   37

Even though the narrator jokes about Julie’s driving skills, there is also a serious
concern at the end of this passage that is more consonant with the implied
author’s norms. This type of tension between serious truths and praising the
habit of lying produces a sense of ambiguity throughout the novel.
What tells us that narrating time is passing while Karin the narrator is nar-
rating is that the narrator’s norms seem to change and gradually merge with the
implied author’s norms. This can be seen in the last sentences of part four, con-
sisting of aunt Selma’s fantastical words of wisdom to Karin. I characterize this
speech as fantastical because it occurs at aunt Selma’s funeral and because the
reader is told the dead aunt sits down by Karin during the ceremony. One clearly
perceives that the narrator is imagining this:

You’re a smart girl, Karin, but sometimes you exaggerate a little too much for your own
good. One day something might happen in your life that is a lot more serious than what’s
come before. One day something might break inside you, and you won’t be able to play as
carefree as you have. One day life will demand that you take responsibility, little Karin. (247)

What is different here from the narrator’s other made-up stories is the serious
content, a content that is probably consonant with the implied author’s norms. As
mentioned above, the funeral takes place in 1991, but in this context right before
part five it functions as a prolepsis to what comes next, namely Julie’s disappear-
ance. Again, we can see that the sjuzhet provides the reader with a Bildungsroman
plot. Another sign of the retrospective “I” merging with the implied author’s
norms is the change in the way that the narrator tells the story. The fantastical
elements (which could be seen as synonymous with her habit of lying) become
less visible toward the end of the sjuzhet and are practically non-existent in part
five. Even though the funeral speech is imagined, it is a more solemn, serious lie
that points toward the seriousness ahead in part five. Because of the change in
the way of narrating from beginning to end, and because of the change in Karin
the narrator’s norms, it can be concluded that narrating time in this novel is not
a fixed point, but rather occurs over time.
To sum up this second level of understanding the novel: we have an unre-
liable sjuzhet that gives the impression of a Bildungsroman plot in the course of
which the character narrator undergoes changes. The traditional change from
distance between the narrated “I” and the retrospective narrating “I” to their
merging is not exactly what happens in this case. What occurs instead is that the
retrospective narrating “I” merges with the implied author’s norms, since these
norms are not consonant with those of the retrospective “I” at the beginning.
38   Jannike Hegdal Nilssen

5 The Third Level of Understanding: Hidden Unre-

liability and Ironizing over the Bildungsroman
What is problematic about the second level of understanding the novel is the way
it blurs the lines between the homodiegetic narrator as a psychological character
and a narrating instance employing internal focalization. For the third level of
understanding, I wish to suggest that in order to propose a narrator who changes
in the course of telling her story, an orchestrator is required who is above the
speaking instance of the sjuzhet. Some might say that this instance is the implied
author. As I have already stated, however, I do not regard the implied author as an
active agent. Also, as this orchestrator is clearly unreliable, I do not see how it can
be compared to something resembling an implied author or as something such as
the values and norms of a work.
My claim is that to explain how the sjuzhet manipulates the fabula to make it
seem as though Karin the character changes in the course of the novel—when it is
actually the retrospective narrating “I” who changes—speaking of an extradieget-
ic-heterodiegetic third-person narrator makes more sense. To employ Genette’s
terms (1980 [1972], 1988 [1983]), the narrating instance of the frame narrative is an
extradiegetic-heterodiegetic narrator of an intradiegetic frame narrative. While in
part one there is a shift in voice to a narrating “I,” it is reasonable to say that we
have a homodiegetic narrating instance and that this voice remains throughout
the novel. Karin is then both an intradiegetic narrator and a metadiegetic charac-
ter. I will now delve into the option that the extradiegetic-heterodiegetic narrator
also remains throughout the entire novel.

5.1 Some seemingly paraleptic passages

Some of the passages that it is not possible to place in the fabula with any clear
time markers (as visualized in table 1) are passages where we find alterations that
first seem to be paralepses.9 The “I” narrates events and other people’s thoughts
that she would not normally be able to know about. One paralepsis starts like

9 “Paralepsis can likewise consist, in internal focalization, of incidental information about the
thought of a character other than the focal character, or about a scene the latter is not able to
see.” (Genette 1980 [1972], 197)
 Rethinking the Unreliable Narrator   39

I am Karin.
Julie is my sister, married to Aleksander.
If anyone should ask me how things are going with them, I could tell one of many stories.
This is one of them: (96)

This is the first of four bedroom chapters. They seem to form a continuous scene
from the bedroom of her sister and her sister’s husband Aleksander even though
they are not consecutive chapters. This comment, prior to a seemingly paraleptic
passage, makes it seem like the narrator is simply making up the story that is
about to come. In regards to these four bedroom chapters, though, only the first
has an introduction like the one quoted above. An example from the second of
the bedroom scenes is:

He listens to her breathing, that regular deep breathing of hers. Her eyes are closed. What
does she think? That he doesn’t know she’s awake, that he doesn’t know she’s awake and
noticing every single movement he makes? She thinks that he thinks she’s asleep. (115)

If these sentences are the voice of the retrospective “I” of Karin, then this must
be a paralepsis, as Aleksander here is the focalizing instance. It could be argued
that there is zero focalization in the last sentence as well, and that the ambiguity
resulting from which mode this actually is creates what Behrendt and Hansen
(2011) call ambiguous discourse, the mode of ambiguity and unreliability result-
ing from the convergence of FID and CID.
However, the context of the sentences prior to and following this one sup-
ports the notion that we are dealing with FID and that Aleksander is the experi-
encing, focalizing instance. However, the ambiguity whereof we suspect narrative
unreliability does not come from the question of whether we have CID or FID here
alone. The ambiguity that causes a blur between the story world and discourse
world (cf. Behrendt and Hansen 2011, 223) stems from the above-quoted narrator
comment placed prior to these four seemingly paraleptic passages. The narrating
“I” sets us up to believe that it is she who is narrating the passages and that she
might be making them up. Along with this confession comes doubt because of the
length of the passages in question. This results in unreliability.
Throughout the novel, the narrating “I” tells numerous stories that are appar-
ently paraleptic, but that are also fantastical and that she quite clearly could not
have experienced herself. We have already seen the example of the imagined
speech aunt Selma gave at her own funeral. Other examples include the external
analepsis of what happened in NYC in the 1930s in part three and the internal
analepsis of what happened between Anni and Preston in NYC in part four. These
stories both have narrator comments that suggest that it is Karin the narrator who
is imagining things. For example, when Anni in NYC sees Preston:
40   Jannike Hegdal Nilssen

She wants him to talk to her. She doesn’t know who he is, but he has a nice smile, she
thinks, and it’s been such a long time since anyone smiled at her that way.
What can I do?
I can’t stop this. I’m in Oslo thinking that Anni’s in New York and behaving like an ordi-
nary mother on vacation. (211)

Passing from the two first lines to the following three will leave the reader con-
fused as to who is actually speaking. Taken together, the five sentences could
suggest FID and thus a paralepsis or CID and two different narrators. In neither
case can ambiguous discourse be ruled out.
Combined, these examples of seemingly paraleptic passages all give the
impression of being uttered by an unreliable narrator, similar to the one described
by the blurbs. Is this a case of moral unreliability, then, that the narrator is simply
creatively inventing, like a zero focalized homodiegetic narrator could? That is
one option.
Another option is that this passage may not be a paralepsis at all, but a
third-person narrating instance in disguise, orchestrating the whole narrative.
Behrendt and Hansen have pointed out how Isak Dinesen’s “The Sailor-Boy’s
Tale” is a first-person narrative in disguise, as the ending of the story reveals that
the sailor-boy “lived long enough to tell this story” (2011, 236). It is also observed
that this third-person narrative demonstrates that the “[…] distance between the
protagonist and the narrator (and their merging in the last sentence) is similar to
the distance (and merging) between the retrospective and the narrated ‘I’ in a first
person narration” (237).
If we explore the idea of a covert extradiegetic-heterodiegetic narrator
“breaking through” in the seemingly paraleptic passages as well as in the frame
narrative at the beginning of the novel, then it would be far less problematic to
claim that the sjuzhet creates a story that is quite different from the story that
emerges out of the fabula. As I have already pointed out, the sjuzhet manipulates
the fabula to make it seem as though Karin the character changes as the story
advances, when it is actually the retrospective narrating “I” who changes; thus,
the expected “from distance to merging between the narrated ‛I’ and retrospective
‛I’” does seem to occur. With this understanding, it is the extradiegetic-hetero-
diegetic narrator who manipulates the fabula and demonstrates how a character
can change by narrating her own story.
 Rethinking the Unreliable Narrator   41

5.2 Ironizing over the Bildungsroman

Arguing that there is actually a covert extradiegetic-heterodiegetic narrat-

ing instance throughout the novel explains why there is not any real merging
between the narrated “I” and the retrospective “I.” This covert project constitutes
the narrator’s unreliability, as in Behrendt and Hansen’s definition for unrelia-
bility in the fifth mode of ambiguous discourse: it “leads the reader to overlook
story world facts of decisive importance for understanding the story” (2011, 230).
In this case, however, it is not the convergence of CID and FID that is at issue, but
the convergence of the first-person narrator and the third-person narrator. On the
metadiegetic level, we have an intradiegetic-homodiegetic narrator who is also a
metadiegetic character. But could this covert extradiegetic-heterodiegetic narra-
tor also be present the entire time, visible only in occasional narrator comments
and in passages that seem to be paralepses?
If we go by this idea of an extradiegetic-heterodiegetic narrator in disguise in
the novel, a narrator who must thus be omniscient, then the thematic point will
necessarily change. The added element in relation to the second level of under-
standing the novel is that the covert omniscient extradiegetic-heterodiegetic nar-
rator—an instance positioned outside the story world—is the one who enables
the sjuzhet to communicate the plot of the Bildungsroman as opposed to the
character narrator relating her own development. This is a point that would be
hard to defend if we were to say that, except in the frame narrative, an intra-
diegetic-homodiegetic narrator prevails throughout novel. It is how we choose to
narrate and talk about our own lives—the way we present our personal fabulas
as a sjuzhet when we speak of ourselves—that decides the outcome of our story
or our identity. Of course, the irony with this novel is that given that there is a
covert extradiegetic-heterodiegetic narrator, Karin is not really telling her own
story. A traditional Bildungsroman always seems to show how a character narra-
tor testifies to have changed and matured through his or her experiences. What
the overarching narrating instance communicates is that what really happens,
ironically, is only a constructed change. This constructed change consists of how
the character chooses to see herself as having changed in the course of narrating
her experiences. Thematically, this point gets lost if we consider that Before You
Sleep has only an unreliable intradiegetic-homodiegetic narrator.
42   Jannike Hegdal Nilssen

6 Theoretical Implications: Rethinking the

Demarcation Heterodiegetic/Homodiegetic
The potential for richer interpretation of a narrative afforded by taking a critical
look at the heterodiegetic/homodiegetic demarcation invites further reflection.
Narratologically speaking, what gets lost if we consider this novel to have only
an unreliable intradiegetic-homodiegetic narrator is the possibility of explain-
ing a fictionalized narrator with particular psychological traits without having
to succumb to the mimetic bias (cf. Alber et al. 2012). How then do we explain the
unnatural fact that a heterodiegetic narrator might just as easily be homodiegetic
(as Behrendt and Hansen 2011 have shown) and that this could also apply the
other way around? And is it really possible that the demarcation between the two
is not necessarily as bulletproof as it is traditionally regarded?
I have already mentioned the Genettian narrative levels, which are
extradiegetic, intradiegetic or metadiegetic. Genette was emphatic in his Nar-
rative Discourse Revisited that we should not confuse the phenomena of person
and level: “[…] the confusion that develops between the attribute extradiegetic,
which is a phenomenon of level, and the attribute heterodiegetic, which is a phe-
nomenon of relation (of “person)” (Genette 1988 [1983], 84). However, I would
like to explore the possibility of introducing different degrees in the relation
between heterodiegetic and homodiegetic. What if the narrating instance could
be regarded along a scale that determines the narrator to be more or less het-
erodiegetic or more or less homodiegetic? This would simplify some of the dis-
cussions regarding natural and unnatural narratology. The whole question of
whether or not the narrator is a character in the story fails with the mimetic bias.
For what is the narrator that is not a character in the story: a character outside
the story?
Why does it have to be one way or the other? In Before You Sleep, I suggest,
the heterodiegetic narrator is at the level above the homodiegetic narrator, and
thus the latter can always be seen as embedded in a heterodiegetic narrator.
But why then is it not just a homodiegetic narrator using zero focalization? The
answer is that it is hard to explain the changes a narrator undergoes without
making her into a fictional psychological character. And once you have a charac-
ter it is hard to avoid trying to find mimetic explanations for unnatural or omnis-
cient knowledge. One explanation is to invent another level, so that the narra-
tive this intradiegetic narrator is narrating becomes a metadiegetic narrative and
what concerns the character narrator happens on the intradiegetic level. Since a
homodiegetic narrator cannot be reduced to not being a character, the relation
between the narrated “I” and the retrospective narrating “I” cannot be properly
 Rethinking the Unreliable Narrator   43

addressed without another instance on a level above the homodiegetic narrator.

This cannot be the implied author, since the implied author is not an agent.
It is possible to argue along the lines that to determine who the narrator is,
you need traces in the text. This also entails that any presumed heterodiegetic
narrator who makes metanarrative comments, attracting attention to writing the
story, for instance, is rather a character narrator, but on a different diegetic level.
My suggestion is that if the pronoun “I” is used in a narrator comment, then we
are in the presence of a character narrator. Booth was the first to observe that
as soon as the narrator has referred to an “I,” the narrator has been dramatized
(1983 [1961], 152). But what now if there is clearly only an “I” who is synonymous
with the protagonist who is visibly present? How could it be claimed that we are
also in the presence of a heterodiegetic narrator?
In a chapter on paralepsis, Paul Dawson investigates the omniscient
homodiegetic narrator in his recent The Return of the Omniscient Narrator (2013).
He discusses Rüdiger Heinze’s (2008) view that an omniscient first-person narra-
tor is not omniscient but rather that any omniscient features must be attributed
to paralepsis (Dawson 2013, 201). Dawson looks into Heinze’s five types of para-
lepsis, of which only two are actual paralepses, because they cannot be natural-
ized in the narrative. This is the case when, for instance, the character narrator is
just making things up (204). Dawson argues that one should see the omniscient
elements of these apparent paralepses as the narrator’s rhetorical strategy: “So
rather than charting different types of paralepses according to degrees of natu-
ralization, we might think about different rhetorical mobilizations of the conven-
tional performative authority of omniscience” (204; original emphasis).
Again, if we were to consider Karin the narrator as employing a rhetorical
strategy while telling about things we would regard as paraleptic or as signs of
omniscience, it still would not remove the reader’s need to explain this narrator
psychologically. For example, as soon as a narrator starts out with a narrative
comment referring to the writing situation of the discourse, then the reader will
automatically imagine a person with psychological traits. I am suggesting that
a less problematic way of approaching the dilemma is to not make a clear dis-
tinction between the heterodiegetic and the homodiegetic. If there is a seemingly
stable and reliable narrating “I,” then clearly there is little heterodiegetic pres-
ence. But as soon as there is ambiguity as to who speaks, it will be more difficult
to determine exactly which is what. I am proposing that to include both would
be more fruitful, even though this might be less necessary in cases where there
seems to be no doubt about what kind of narrating instance we are dealing with.
To be able to speak of a narrating instance, it is clearer to define this as a
heterodiegetic narrator who in fact is not a character in the story world or who is
outside the story world. As soon as you build a picture of a narrator whose voice
44   Jannike Hegdal Nilssen

you can gather to be some sort of person, I would claim that this is a character
narrator, whether it is a character inside or outside the story world, and that this
narrator always has an omniscient heterodiegetic narrator above, functioning
as a “chief” narrating instance. This is what I have tried to demonstrate in the
novel Before You Sleep, where the presence of an “omnipresent” heterodiegetic
type of narrator can explain alterations and paralepses. As for the apparent
paralepses discussed above, they can still be regarded as employing the mode of
ambiguous discourse; however, the Aleksander-focalized passages would then
be extradiegetic-heterodiegetic, and not paraleptic. The ambiguity would consist
of whether the passage is in CID or in FID.
The key to viewing this proposal as not being reductive is to see how the
intradiegetic-homodiegetic narrator and extradiegetic-heterodiegetic narrator
function together. They need to be considered more or less simultaneously. The
third level of understanding the novel Before You Sleep does not make much
sense without the presence of the first two levels: the extradiegetic-heterodiegetic
narrator covertly structures the sjuzhet to make it seem as though the character
Karin and the narrator Karin are both changing while at the same time the narra-
tor Karin is narrating. There is much more meaning to harvest from the discourse
when regarding all the options simultaneously. Combined, they contribute more
layers of meaning than when one feels constrained to decide between one or the

Works Cited
Abbott, Porter. 2007. “Story, plot and narration.” In The Cambridge Companion to Narrative,
edited by David Herman, 39–51. New York: Cambridge University Press.
Alber, Jan, Stefan Iversen, Henrik Skov Nielsen and Brian Richardson. 2012. “What Is Unnatural
about Unnatural Narratology?” Narrative 20 (3): 371–382.
Bakhtin, Mikhail M. 1986 [1936–38]. “The Bildungsroman and Its Significance in the History
of Realism (Toward a Historical Typology of the Novel).” In Speech Genres and Other Late
Essays, translated by Vern W. McGee, edited by Caryl Emerson and Michael Holquist,
10–59. Austin: University of Texas Press.
Behrendt, Poul, and Per Krogh Hansen. 2011. ”The Fifth Mode of Representation: Ambiguous
Voices in Unreliable Third-Person Narration.” In Strange Voices in Narrative Fiction, edited
by Per Krogh Hansen, Stefan Iversen, Henrik Skov Nielsen and Rolf Reitan, 219–251. Berlin:
De Gruyter.

10 I wish to thank Paul Dawson, Wolf Schmid, John Pier and Per Krogh Hansen for their very
helpful comments and suggestions for improvements to this article.
 Rethinking the Unreliable Narrator   45

Bergström-Edwards, Pia. 1999. “Suck!” In Aftonbladet. August 16.

vss/kultur/bokbanken/recension/0,2024,9113007475,00.html (Accessed 19 May 2005)
Booth, Wayne C. 1983 [1961]. The Rhetoric of Fiction. 2nd edition. London: Penguin Books.
Booth, Wayne C. 2005. “Resurrection of the Implied Author: Why Bother? ” In Phelan and
Rabinowitz, eds., 75–88.
Dawson, Paul. 2013. The Return of the Omniscient Narrator: Authorship and Authority in
Twenty-First Century Fiction. Columbus: The Ohio State University Press.
Genette, Gérard. 1980 [1972]. Narrative Discourse. An Essay in Method. Translated by Jane E.
Lewin. Foreword by Jonathan Culler. Ithaca, NY: Cornell University Press.
Genette, Gérard. 1988 [1983]. Narrative Discourse Revisited. Translated by Jane E. Lewin.
Ithaca, NY: Cornell University Press.
Hansen, Per Krogh. 2011. “Backmasked Messages: On the Fabula Construction in Episodically
Reversed Narratives.” In Unnatural Narratives—Unnatural Narratology, edited by Jan Alber
and Rüdiger Heinze, 162–185. Berlin and Boston: De Gruyter.
Heinze, Rüdiger. 2008. “Violations of Mimetic Epistemology in First-Person Narrative Fiction.”
Narrative 16 (3): 279–297.
Herman, David. 2005. “Histories of Narrative Theory (I): A Genealogy of Early Developments.” In
Phelan and Rabinowitz, eds., 19–35.
Moretti, Franco. 2000. The Way of the World: The Bildungsroman in European Culture. London:
Phelan, James. 2005. Living to Tell about It. A Rhetoric and Ethics of Character Narration. Ithaca,
NY and London: Cornell University Press.
Phelan, James, and Peter J. Rabinowitz, eds. 2005. A Companion to Narrative Theory. Malden,
MA and Oxford: Blackwell Publishing.
Schmid, Wolf. 2014 [2009]. “Implied Author.” In Handbook of Narratology, edited by Peter
Hühn, Jan Christoph Meister, John Pier and Wolf Schmid, 288–300. Hamburg: Hamburg
University Press. Also available in the living handbook of narratology at: http://www.
(Revised 16 May 2014)
Skei, Hans H. 1999. “Glimt frå bokhausten 1998.” In Norsk litterær årbok, 209–217. Oslo:
Ullmann, Linn. 2000 [1998]. Before You Sleep. Translated by Tina Nunnally. New York: Picador.
[Translation of Før du sovner. 1998. Oslo: Tiden Norsk Forlag A/S]
Ullmann, Linn. 2001. Before You Sleep. (Online novel blurb). Sonnet Media. http://www. (Accessed 3 October 2014)
Wærp, Henning H. 1998. “Litt fra bokhøsten 1998. II Familiefragmenter.” Nordlit 4: 197–180.
Per Krogh Hansen (Kolding)
Autofiction and Authorial Unreliable
In recent years, the concept of “the unreliable narrator” has been among the most
debated within narrative theory. In the wake of a series of provocative articles
from the late 1990s by Ansgar Nünning (Nünning 1997, 1998a, 1998b, 1999), ques-
tions have been asked again and again regarding on what basis we determine
whether a narrator is unreliable and how broad the scope of the concept is. Is
the presence of an unreliable narrator in a given text the result of an author’s
intentional decision, or is narratorial unreliability a historically variable reader
response to textual inconsistencies and/or changing cultural norms? Does the
concept belong exclusively to fiction, or does it make sense to approach factual
or “real” narrators with the same concepts we encounter in fictional narrators?
In this article I will address these questions with reference to the genre of
“autofiction” (a genre parallel to or a sub-genre of autobiography) with special
attention to the Norwegian author Karl Ove Knausgård’s six-volume novel Min
kamp (2009–11). I want, on the one hand, to discuss how autofiction makes itself
vulnerable to narratorial unreliability due to the complex truth status of the told
and, on the other hand, to use this genre to question the concept of unreliable
narrator in rhetorical criticism as dependent on an intentional act by an (implied)
author. This is not to say, however, that unreliable narration cannot be a narrative
technique deliberately used by an author in the creative act of novel writing, but
that by using the concept exclusively in relation to this phenomenon, we lose
sight of important alternative (but comparable) versions. Among these is the kind
of unreliability we can encounter in autofiction.

1 The Real Fiction of the Self

When Serge Doubrovsky coined the term “autofiction” in relation to his 1977
novel Fils, he defined it, rather paradoxically, as “Fiction, of strictly real events
and facts.”1

1 Doubrovsky coined the term on the back cover of the novel, where he wrote: “Autobiographie ?
Non, c’est un privilège réservé aux importants de ce monde, au soir de leur vie, et dans un beau
style. Fiction, d’évènements et de faits strictement réels ; si l’on veut autofiction, d’avoir confié

DOI 10.1515/9783110555158-003
48   Per Krogh Hansen

Doubrovsky’s work was inspiringly provoked by the fact that Philippe

Lejeune, in his influential 1975 study Le pacte autobiographique, failed to leave
room for blending novelistic fictional writing and factual autobiography. Lejeune
defined autobiography as

a retrospective prose narrative produced by a real person concerning his own existence,
focusing on his individual life, in particular on the development of his personality. (Lejeune
1989 [1975], 4)

Lejeune stated that if a reader was confronted with a text where author, narra-
tor and protagonist were the same, he or she would not accept counterfactual
events or incidents as a matter of fiction. “False” information in autobiograph-
ical writing, Lejeune claimed, would instead be related to “the order of lying”
(“l’ordre du mesonge”).
Doubrovsky broke this dogma two years after Lejeune’s study when he pub-
lished the novel Fils and proclaimed a new genre: ‘autofiction’. For Doubrovsky,
it was characteristic of this new genre that it required homonymy between its
author, narrator and character and that it played on the generic ambiguity of its
contradictory pact: on the one hand, the work is claimed to be absolutely referen-
tial and factual; on the other, it is claimed to be a novel, that is: fiction.
The result is not only a contradictory pact but what Poul Behrendt (2006) has
labeled a “double contract.” On the one hand, the text is subject to the rules gov-
erning our social and cultural interactions, that is, where people have real names
and responsibilities, where (at least in our part of the world) there is freedom of
speech, but also laws protecting us from defamation, etc. On the other hand, the
text also relates to the communicative system of fiction, where reader and author
have agreed that exceptions exist. Here, the distinction between lying and truth is
disregarded: ideas and beliefs can be expressed, even if they are counterfactual,

le langage d’une aventure à l’aventure d’un langage en liberté, hors sagesse et hors syntaxe du
roman, traditionnel ou nouveau. Rencontres, fils de mots, allitérations, assonances, dissonanc-
es, écriture d’avant ou d’après littérature, concrète, comme on dit musique.” (Doubrovsky 1977)
“Autobiography? No, that is a privilege reserved for the important persons of this world, in the
evening of their lives, and in a beautiful style. Fiction, of strictly real events and facts, autofiction
if you will; to have given the language of an adventure to the adventure of a language in freedom,
without wisdom and outside the syntax of the novel syntax, traditional or new. Interactions,
threads of words, alliterations, assonances, dissonances, writing before or after literature, con-
crete, as they say music.” (translation mine)
Today, the term is included in French dictionaries, and in Canada it is widely used as a genre
concept, printed on the cover of novels.
 Autofiction and Authorial Unreliable Narration   49

as contributions to the ongoing negotiation of the constitution of our culture and

Here is not the time and place to follow up on the development of the concept
suggested by Doubrovsky in detail, since it has been scrutinized and developed
by several since then—most importantly perhaps by Gérard Genette in his Fiction
and Diction (1993 [1991]). Genette suggested that all cases in which an author of
fiction includes his own person (or a character with the same name as the author)
in his fictional story should be considered autofiction. If so, works like Cervantes’
Don Quixote (1605–15) and Dante’s The Divine Comedy (1555) would be included
in the category. But as it has been noted, Genette thereby disregarded the second
part of Doubrovsky’s characterization, namely that the work has to play on the
generic ambiguity that comes out of the claim of absolute referentiality and abso-
lute fictionality at one and the same time. Illustrated with another example:
when in City of Glass (1985), the first part of Paul Auster’s New York trilogy, a
character with the name “Paul Auster” appears and is furthermore presented as
an author living in New York in an environment that reminds us of the real Paul
Auster’s surroundings, it is less an autofictional than a metafictional strategy that
is being implemented. The intention is not to tell the story of Paul Auster in fic-
tional terms, but rather to enter a mode comparable to romantic irony.
What autofiction does is quite radical in the sense that instead of demarcat-
ing fiction from reality it blurs the border. This can be accomplished by promoting
a picture of the authorial self which confirms, negates, transforms or plays with
the public understanding of this self. Such is the case of Bret Easton Ellis’s Lunar
Park (2005), where the author mixes known public events and incidents from his
real life with counterfactual and even paranormal events and incidents. The blur
between fiction and reality can also result from using the novel form and the nar-
rative techniques related to this form to investigate and depict the history of the
author’s own self. This was what J. M. Coetzee did in the three volumes of autofic-
tional memoirs—Boyhood (1997), Youth (2002) and Summertime (2009)—when he
chose to let the first two volumes be told heterodiegetically in third-person narra-
tion and the third to take the form of a fictitious biographer’s interviews with five
factual people from Coetzee’s past. Coetzee himself is claimed to be dead in the
novel, and his voice only appears through a number of third-person fragments of
the kind the reader already knows from the first two volumes. But even though
no one would question that the trilogy most certainly is written in fictional terms
and therefore cannot be understood literally, the books are also understood to be
about Coetzee’s own life. In this perspective, autofiction pushes what is a general
paradoxical characteristic of practically all storytelling, namely that stories tell
the (or some) truth, even though what they are telling might not have happened.
50   Per Krogh Hansen

It is therefore also evident that we cannot always rely on the factuality of the
story being told by the author. But can we approach this authorial unreliability in
terms of ‘unreliable narration’?

2 Factual Unreliable Narration?

It has been claimed that only in fictional narrative can we have true cases of unre-
liable narration. The argument goes that narrative unreliability depends on, if not
difference, then at least on distance between narrator and authorial agent. In her
essay on discordant narration, Dorrit Cohn claims

that the diagnosis of ‘discordance’ can apply only to a fictional narrative, not to the kind of
storytelling (oral or written) that presumes to refer to real facts: though we often apply the
term ‘unreliable’ to voices we regard as wrong-headed in non-fictional works (historical,
journalistic, biographical, or autobiographical), the narrator of such works is the author,
the author is the narrator, so that we cannot attribute to them a significance that differs
from the one they explicitly proclaim. (2000, 307)

This assumption was challenged by James Phelan in the chapter from Living to
Tell About It on Frank McCourt’s memoir novel Angela’s Ashes: A Memoir (1996).
Phelan demonstrates how McCourt has “built his narrative on the founda-
tion of unreliability and the virtual absence of reflection in his authorial voice.”
McCourt’s “trick” is, as Phelan writes, to conceive “memoir,” not as an art of
direct telling from author to audience, but as an

art of indirection. Rather than speaking in his own voice at the time of the telling, McCourt
uses the historical present and speaks in the voice of his former self […] at the time of the
action. (Phelan 2005, 67)

What makes the narratorial unreliability possible in McCourt’s case is that he

uses what Phelan recognizes as a “nonstandard technique” for his autobiograph-
ical purpose when he “re-invents” his childhood-persona’s perspective on the
incidents. This technique is, on the other hand, a standard novelistic, fiction-
alizing technique: McCourt-the-Author has given the voice and the perspective
to a less experienced narrator, Frankie (McCourt-the-boy-character), and even
lets him tell from the moment of the action that is in the first person and in the
present tense. Hereby the difference or distance claimed by Cohn as a necessity
for the formation of an unreliable narrator is reinstated.
We might therefore also conclude that the extended use of this fictionalizing
technique relates McCourt’s memoir-novel to the genre of autofiction more than
 Autofiction and Authorial Unreliable Narration   51

it does to autobiography proper. Angela’s Ashes is “[f]iction, of strictly real events

and facts,” and it suggests homonymy between author, narrator and character,
although this homonymy cannot be considered absolute: it is nominal and per-
sonal, but not temporal. Furthermore, it plays on the generic ambiguity of what
Doubrovsky called autofiction’s “contradictory pact”: on the one hand, the work
claims to be absolutely referential and factual; on the other, it makes extended
use of a standard novelistic technique and therefore relates to fiction, too, at
least in Richard Walsh’s understanding of fictionality as a rhetorical rather than
an ontological quality. Fictionality should not, in this perspective, “be equated
simply with ‘fiction,’ as a category or genre of narrative: it is a communicative
strategy, and as such it is apparent on some scale within many nonfictional nar-
ratives” (Walsh 2007, 7).
But what would the result be if one’s attention were directed toward a work
that is not composed with the same obvious distance between the authorial I,
the narratorial I and the character I? If we were to follow Phelan’s line of argu-
ment, where unreliable narration is defined as narration “in which the narrator’s
reporting, reading (or interpreting), and/or regarding (or evaluating) are not in
accord with the implied author’s” (Phelan 2005, 219), then it is quite unlikely
that the narrator would be unreliable in such a case, since the necessary distance
between the subject positions is eliminated from the very outset. The question,
then, is whether Phelan’s exclusion would be correct—or whether it is the result
of too narrow a conception of unreliable narration.

3 A Norwegian’s Struggle with his Past and

That it is the latter which is the case I will try to demonstrate by looking at an
example of autofiction which makes us consider the reliability of the narrator
without seeking recourse to the concept of the implied author. The example is
the Norwegian author Karl Ove Knausgård’s six-volume autofictional novel Min
kamp, meaning My Struggle.2

2 The first volumes have been published in several languages including German, English and
French. Due to the obvious allusion to Adolf Hitler’s Mein Kampf (an issue to which attention
is given in the sixth volume of the series), the books have appeared under other titles than in
Scandinavia. The first volume is called Sterben (2011) in German, La Mort d’un père (2012) in
52   Per Krogh Hansen

My Struggle is among the greatest literary sensations in Scandinavia in

decades. Not only because it has been a huge sales success, but also because the
publication of the six volumes has been accompanied by a heated debate about
the use of autobiographical elements in fiction, and vice versa. In the six books,
Knausgård gives a detailed description of his life from the day he was born until
the moment he types the final sentence of the manuscript of volume 6, taking into
account the reception of the first volumes of the project and the effect it has had
on himself and his relationships. Knausgård tells the story in the first person and
from the position of the writing situation, the first and the last volumes in particu-
lar containing long essayistic passages reflecting on life, death, art and literature.
In long sections he changes the focalization and even the narrative tense from
past to historical present, so that it isn’t the narrating I’s but the experiencing
I’s (Karl Ove’s) perspective we are confronted with. In these chapters there are
numerous examples of the kind of unreliability Phelan focused on in McCourt’s
case, with the difference, however, that unreliability is established here due to
the distance between a diegetic narrator and a fallible filter.
But these examples are less interesting in the present perspective. It is more
interesting that there are aspects of the work which raise doubts about and even
undermine the author-narrator’s reliability.
In volume two, in the midst of a very detailed retelling of a long conversation
during a dinner party at a restaurant, Karl Ove and his friend Geir agree upon the
fact that Karl Ove has always had an extremely bad memory. Due to the frame, the
statement becomes a variant of Epimenides’ paradox: “All Cretans are liars.” An
author giving a hyperdetailed account of a long-past situation cannot in the same
situation claim to have a bad memory. The impact of this inconsistency is quite
remarkable. The conversation between Geir and Karl Ove is on a very highbrow
intellectual level of almost Socratic-dialogic dimensions, focusing on, among
other things, the personal costs connected with “the authorial call,” etc. But the
remark concerning Karl Ove’s weak memory makes the reader speculate whether
the whole dialogue isn’t more a matter of the authorial I’s creative attempt to put
himself in a better light, due to the decisions regarding family life and relation-
ships he has had to make to pursue his career as an author. We begin speculating
whether Knausgård-the-author’s reporting is reliable.
Another example: over the course of the six volumes, Knausgård returns to
some of the significant incidents in his life. Among these is his complicated rela-
tionship to his tyrannical and choleric father who died of alcohol abuse ten years

French and A Death in the Family (2012) in English. In the latter case, the series has been called
My Struggle. In the following pages, I will refer to the series name.
 Autofiction and Authorial Unreliable Narration   53

prior to the time of writing. The last period of the father’s life, as it was experi-
enced by Karl Ove, is focused on in both the first and the last volumes. But in
the latter case, the description is much more downtoned than in the former. And
where Karl Ove and his older brother Yngve were pretty much alone in clean-
ing up the mess after the father in the first volume, they receive attention and
help from an uncle and aunt in volume 5. Both descriptions focalize the incidents
through the experiencing I, but it seems as if the narrating I—the author—through
the process of writing the novel, develops his understanding of the story as it is
told, a contradiction that exposes a dynamic unreliability regarding values and
judgments. In volume 6, Knausgård receives an angry letter from his uncle (his
father’s brother) in which the author’s version of the incidents in relation to the
father’s death is being questioned with reference to supposed hard evidence. And
even though Knausgård becomes aware that there might be flaws and misjudg-
ments in his former retelling of the incidents, he is not abandoning it. As he says
to a prosecutor in an imagined trial: “This is how I remember it” (volume 6, 303;
translation mine). But as we (and he himself) know: he has an extremely bad
The death of the father is not the only example of this kind of unreliability.
In volume 1 we are told that Knausgård spent four years with a girl he didn’t love;
but when the relationship is described for us in the fifth volume, it most certainly
seems characterized by love. Again: the authorial perspective on and understand-
ing of the incidents has taken over, and the result is that our natural urge to rely
on the authorial voice of the narrative is challenged. Where our expectation with
regard to the authorial narrative agent is that it is stable and sanctions the norms
and values of the storyworld, we are instead engaging with inconsistencies, and
the result for our reading is that we redirect our attention from the told to the
teller and thus see the misrepresentations and inconsistencies as an expression
of character traits and unreliable narration. In that sense, Knausgård is, either
intentionally or unintentionally, flouting the assumption of a stable author. In the
beginning, we greet the representation as reliable. But as the telling and retelling
progress, we start having second thoughts, just as Knausgård himself does in the
sixth volume, when (as pointed out above) he refers to memory instead of fact.
The examples commented on so far are all intra- or internarrational insofar
as the effect of unreliability is a result of contradictions within the narrator’s dis-
course or between his discourse and the differing perspective of others (e. g., his
uncle’s) perspective on the same incidents.
54   Per Krogh Hansen

But there are also examples of extratextual circumstances influencing the

author-narrator’s reliability.3 Even though the books were published as “novels,”
Knausgård claimed that everything he told was true. All material had been pre-
sented to the persons concerned, and only a few names were changed at their
Everything in this work is depicted with an attention to detail that marks
the work as a fictional recreation of the past: no one can remember their past
as clearly and elaborately as Knausgård does. And even though we as readers
might accept the level of detail, bearing in mind that this is also a work of fiction,
the authenticity is punctured from inside the storyworld by the aforementioned
unreliability signals.
Due to the work’s claim to factuality, moreover, it was also punctured from
the outside by the persons and family members depicted. Several felt exposed
and misrepresented and expressed their displeasure and disappointment through
the media. Fourteen family members even announced that they wanted to bring
Knausgård to trial for his “Judas literature.”4
Now, Lejeune claimed that when an author writes in his own name, he signs
an autobiographical pact in which everything stated is to be considered true.
If something turns out to be false, it is not a matter of fictionalization but, as
quoted earlier, relates to the “order of lying.” This rather strict understanding of
autobiography’s truth value has been revised in later conceptions of the genre.
As Phelan notes, autobiographical theory has repeatedly shown that “subjective
truth is far more important to memoir than literal truth […] because it is crucial to
the autobiographer’s ability to give shape and meaning to experience” (2005, 73).
But as he also remarks, subjective “truth must also be accountable to some extent
to facts, people, and events that have an existence independent of the autobiog-
rapher’s perception” (73).
With reference to Lejeune’s initial distinction, it can be observed that a radical
subjective recounting of the story jeopardizes the author’s reliability and makes
us, the readers, react in the same way as when we get suspicious about a narra-
tor’s account in fictional narrative: we read with precaution and look beyond the
authorial representation of the facts; we try to figure out the true facts of the case

3 The concepts intranarrational, internarrational and extratextual unreliability belong, together

with intertextual unreliability, to a taxonomy of different ways of signaling/detecting unreliable
narration. For further detail, see Hansen (2005, 2007, 2009).
4 Uncle Gunnar (which isn’t his real name) was among the most critical voices to the project,
notably in a commentary published in the newspaper Fædrelandsvennen (cf. Kristensen 2011).
Knausgård’s estranged ex-wife, Tonje Aursland, also retorted in October 2010 in a radio docu-
mentary broadcast on NRK.
 Autofiction and Authorial Unreliable Narration   55

and use our observations to construct a critical picture of the authorial self about
which we make ethical judgments. We might even press charges against this self
due to the fact that autobiography belongs to the system of factual communica-
tion that is restricted by responsibilities and protection against defamation.
It is also in this respect that we find the main source of the difference between
the everyday use of the concept of unreliability and its use within the context of
fictional narrative: where the two forms of unreliability both invoke the reader’s
ethical judgment, only the latter opens up for an esthetic judgment.
Autofiction blurs this distinction. Insofar as Knausgård’s work is also claimed
to be fiction, the system of factual cultural interaction and communication is
partly suspended, and instead the rules governing fiction takes over: whatever
is depicted is fictionalized, and thus a simulated reality is told from a given per-
spective. Considered an author of a work of fiction, Knausgård cannot be blamed
for the misrepresentation of persons and events or for his exaggerated remem-
brance of details. Under the auspices of novelistic fiction, this kind of re-crea-
tional depiction of the past is expected and allowed.
This, of course, also has significance for the reader’s approach to the autho-
rial self. Due to the many misrepresentations in Knausgård’s narrative, the ethical
judgments passed on both the actions of the author and the characters and the
critical and questionable aspects of the author’s telling are accompanied by
esthetic judgments. The narrative is not read as an author’s telling about his past,
but as a refiguration of that past.
By making himself the author, the narrator and the protagonist of his work
and at the same time claiming the storyworld to be in accordance with reality,
Knausgård’s result is more or less doomed to be an act of unreliable narration.
Reliability is dependent on a perspective that can tell the truth. This is why, on a
general level, it makes sense to speak of fictional truth, insofar as narrative fiction
is the creation of a storyworld from a given perspective. But reality as such does
not support one truth alone. Truth in reality is perpetually being renegotiated.
When Knausgård claims that he tells the truth, but does it within the framework
of fiction, he suspends the negotiable nature of factual truth. By doing so, he is
both hit and saved by what we could call the “kernel paradox” of autofiction: he
claims that what he tells is true, thus opening up the possibility of negotiation,
while at the same time claiming that what he says is fiction, authorizing the truth
value of the told. It is in this tension between author and narrator that Knaus-
gård’s unreliability comes into being. More generally, autofiction as a genre must
at the very least be suspected of extratextual unreliability.
56   Per Krogh Hansen

4 Auto-Mocking and Self-Deception

From the perspective of unreliable narration as narration “in which the narra-
tor’s reporting, reading (or interpreting), and/or regarding (or evaluating) are not
in accord with the implied author’s” (Phelan 2005, 219), autofictional authorial
unreliability of the kind we find in Knausgård’s novels will clearly fall outside
the definition. We could therefore choose to invent a new term to cope with these
deviant cases.5 But insofar as there are quite a few common aspects between the
two sorts of unreliability, I would like to avoid throwing yet another narrative
term on the heap of narratological concepts and models. Let me instead close
this article by considering whether room can be found within the standard con-
ception that will allow us to accommodate narrative works such as Knausgård’s.
The concepts that serve as the toolbox we label “narratology” come from
a great variety of disciplines and discourses and are brought to foreign areas
thanks to the fact that someone has found an item that lacks description but
shares significant aspects with another item that has already been described. The
reason for this travelling applicability is the fact that narrative is a transtextual,
transgeneric and transmedial phenomenon. Applying general narrative concepts
developed in one context to comparable items in another context gives us the
added advantage of providing our analysis with considering whether the concep-
tion has been biased by the fact that it is formulated within a framework of one
genre, mediality or communicational mode.
Scholars from literary studies in particular have provided narratology with
a nearly endless series of useful concepts for the transmedial and transdiscipli-
nary study of narrative. But they have also quite often made themselves guilty
of making general claims as to what narrative is and does, whereas what they
are actually describing are specific features for the prose fiction, not narrative in
The concepts of unreliable narrator and unreliable narration have travelled
between genres and modalities ever since Wayne Booth coined the term more
than fifty years ago in his study of the rhetoric of novelistic narrative fiction. Ini-
tially, at least, he did not consider whether the concept was of general relevance
or limited to literary fiction. Even so, it seems that Booth did have a vague idea
that there was more to the story than he told.
Booth has often been blamed for the inconsistency of his initial formulation:

5 Phelan (2011) has suggested that we should distinguish between unreliable and deficient nar-
ration, the latter covering an example such as the one I have discussed here.
 Autofiction and Authorial Unreliable Narration   57

For lack of better terms, I have called a narrator reliable when he speaks for or acts in
accordance with the norms of the work (which is to say, the implied author’s norms), unre-
liable when he does not. (Booth 1991 [1961/83]: 158–159)

All discussions of the concept since then have their basis in differences of accen-
tuation in this definition. Rhetoricians working on literary narrative have focused
on the parenthetically mentioned authorial agent, intentionality and normativ-
ity (understood as ethics) whereas more semiotically oriented approaches have
highlighted the lack of accordance between the narrator’s representation of the
storyworld and the actual constitution of it. One might claim that the former has
developed the Boothian concept with respect to the fact that he was working with
literary, fictional narrative, whereas the latter have tried to include it in more
transdisciplinary approaches.
If we can agree that the concept has proven its value as a traveling concept, I
think we owe it ourselves to reconsider Booth’s definition and liberate the “narra-
tive/narratological” part of the concept from the literary part.
By focusing on unreliability in Knausgård’s autofictional case, it is clear that
the implied author component is not a necessary part of the unreliable narrator
machinery. Rather, it is a concept that proves helpful in the distinct cases of unre-
liable narration in which it is used as technique for what Phelan calls “indirec-
tion.” This, we could claim, is a feature related to fiction, a matter of fictionality,
and thus important to study in the study of narrative fiction.
With regard to the reliability of the narrator, seeking explanatory recourse
in the implied author is a matter of framing. What relates unreliable narrators
across the fact/fiction distinction, across genres and modalities, is that they do
not speak or act in accordance with the norms, values or facts of the storyworld.
As discussed above, it is characteristic for fictional storyworlds that their consti-
tution can be authorized by an authorial agent. In factual narratives, by contrast,
author and narrator are often (as Cohn stated) the same. This, however, does not
rule out the possibility of unreliability. The storyworld is simply not governed by
an implied author in these cases, but rather by sensus communis to the extent that
readers have a stake in it.
Autofiction is a special case, since it blurs the borders between author, nar-
rator and character as well as between fact and fiction. In itself, it is an extremely
unreliable genre, and it can thus take either the form of intentional auto-mock-
ing, as in McCourt’s example, or that of seemingly unintentional self-deception,
58   Per Krogh Hansen

as in Knausgård’s. So perhaps, after all, there is still room and a need for new
conceptions and distinctions.6

Works Cited
Behrendt, Poul. 2006. Dobbeltkontrakten. En æstetisk nydannelse. København: Gyldendal.
Booth, Wayne C. 1991 [1983/61]. The Rhetoric of Fiction. London, etc.: Penguin Books.
Cohn, Dorrit. 2000. “Discordant Narration.” Style 34 (2): 307–316.
Doubrovsky, Serge. 1977. Fils. Paris: Galilée.
Genette, Gérard. 1993 [1991]. Fiction and Diction. Translated by Catherine Porter. Ithaca, NY:
Cornell University Press.
Hansen, Per Krogh. 2005. “When Facts Become Fiction: On Extra-Textual Unreliable Narration.”
In Fact and Fiction in Narrative: An Interdisciplinary Approach, edited by Lars-Åke Skalin,
227–307. Örebro: Örebro University.
Hansen, Per Krogh. 2007. “Reconsidering the Unreliable Narrator.” Semiotica 165 (1/4):
Hansen, Per Krogh. 2009. “Unreliable Narration in Cinema. Facing the Cognitive Challenge
Arising from Literary Studies.” Amsterdam International Electronic Journal for Cultural
Narratology (ACJN) 5. hansen.htm (Accessed 15
February 2016)
Knausgård, Karl Ove. 2009–2011. Min kamp 1–6. København: Lindhardt og Ringhof.
Kristensen, Eivind. 2011. “‘Onkel Gunnar’ tar knallhardt oppgjør med Knausgård.”
Aftenposten. 17 February 2011.
knallhardt-oppgjor-med-Knausgard-6699655.html (Accessed 15 February 2016)
Lejeune, Philippe. 1989 [1975]. On Autobiography. Translated by Katherine Leary. Minneapolis:
University of Minnesota Press.
McCourt, Frank. 1996. Angela’s Ashes: A Memoir. New York: Scribner.
Nünning, Ansgar. 1997. “‘But why will you say that I am mad?’ On the Theory, History, and
Signals of Unreliable Narration in British Fiction.” AAA. Arbeiten aus Anglistik und
Amerikanistik 22 (1): 83–105.
Nünning, Ansgar. 1998a. “Unreliable Narration zur Einführung. Grundzüge einer kognitiv-
narratologischen Theorie und Analyse unglaubwürdigen Erzählens.” In Unreliable
Narration: Studien zur Theorie und Praxis unglaubwürdigen Erzählens in der
englischsprachigen Erzähllitteratur, edited by Ansgar Nünning, Carola Surkamp and Bruno
Zerweck, 3–40. Trier: WVT.
Nünning, Ansgar. 1998b. “Unreliable Narrator.” In Encyclopedia of the Novel, edited by Paul E.
Schellinger, vol. 2: 1386–1388. Chicago and London: Fitzroy Dearborn Publishers.
Nünning, Ansgar. 1999. “Unreliable, compared to what? Towards a Cognitive Theory of
‘Unreliable Narration’: Prolegomena and Hypotheses.” In Grenzüberschreitungen:

6 I would like to thank Professor James Phelan (Ohio) for his valuable comments on an earlier
draft of this article.
 Autofiction and Authorial Unreliable Narration   59

Narratologien im Kontext/Transcending Boundaries: Narratology in Context, edited by

Walter Grünzweig and Andreas Solbach, 53–73. Tübingen: Gunter Narr Verlag.
Phelan, James. 2005. Living to Tell about It. A Rhetoric and Ethics of Character Narration. Ithaca,
NY: Cornell University Press.
Phelan, James. 2011. “The Implied Author, Deficient Narration, and Nonfiction Narrative: Or,
What’s Off-Kilter in The Year of Magical Thinking and The Diving Bell and the Butterfly?”
Style 45 (1): 127–145.
Walsh, Richard. 2007. The Rhetoric of Fictionality: Narrative Theory and the Idea of Fiction.
Columbus: The Ohio State University Press.
David Stromberg (Jerusalem)
Beyond Unreliability: Resisting
Naturalization of Normative Horizons
“Why is it that a novelist […] should go out of
his way to force inconsistencies on us?”
H. M. Daleski (1985, 16)

1 The Unreliability Debate

The term “unreliable” was introduced by Wayne Booth to describe narrators who,
he claimed, speak or act in a way that is not “in accordance with the norms of the
work (which is to say the implied author’s norms)” (Booth 1983 [1961], 158). Booth
further described “unreliability” as a distance “between the fallible or unreliable
narrator and the implied author who carries the reader with him in judging the
narrator” (158). However, by discussing such distance as separating a work’s nar-
rator from the thematic or authorial norms implied by that work, Booth isolated
an integral element of a work’s literary dynamic—its problematized narrator—
and normatively detached it from the rest of the work. He formulated a narra-
tive’s internal tension in terms of a normative separation between narrator and
an author’s work even though the narrator is deployed by the author as part of the
work. By framing the discussion in terms that privilege the position of the author
or work over that of the narrator, Booth’s approach resolves a tension that, when
preserved, may be full of significance.
Another confusion about the use of unreliability arises from Booth’s having
largely elided the nature of the norms in relation to which the narrator diverges
from the work—whether they are artistic, moral, cognitive, ideological, political,
cultural, social, etc.—or the dynamic relationship that exists between different
kinds of norms. The term “unreliable narrator” was used mainly to relate either to
the cognitive abilities or to the moral attitudes of a narrator. The two are distinct
but interconnected and have ethical implications, particularly when cognitive
inadequacy leads to moral errors. But as I shall suggest in the example of Isaac
Bashevis Singer’s “Gimpel the Fool” (1953), a story that raises this specific norma-
tive tension in the figure of its narrator, resisting a naturalizing judgment of a nar-
rator’s cognitive or moral reliability can open broader aspects of its significance.
Scholarly attempts to parse the meaning of unreliability have aimed to clarify
the different levels and kinds of distance that are most commonly encountered in
literary works. Many theorists attempt to discern between narrators who are unre-

DOI 10.1515/9783110555158-004
62   David Stromberg

liable because they do not know better and those who are unreliable by malice.1
The past decade, in particular, has seen repeated attempts by narratologists and
narrative theorists to synthesize positions on the subject.2 Several of these critical
works point to a need for further classification of the different “types” of unrelia-
bility; others extend the scope of meanings projected by narrative unreliability.3
As a result, scholars who seek to widen the discussion continue to use unreliabil-
ity as their starting point.
The language of unreliability introduces value judgment into the discussion
of a literary phenomenon that merely points to a distance engendering tension. In
trying to resolve narrative tension, readers and critics tend to judge its agent in a
critical equivalent of shooting the messenger. I would suggest that it may be more
helpful to think about the distance as being between the narrator and story—
pointing to a tension between the teller and the told—and that we can evaluate
the distance using a dynamic doubt-faith paradigm rather than a static natural-
izing principle like unreliability. This preserves awareness of a narrative tension
that, I believe, opens our interpretive horizons. Unlike the finalized nature of the

1 Martínez-Bonati notes three levels: “First, there are unreliable basic narrators whose mimetic
sentences are uncertain but not necessarily false […] Second, there are unreliable narrators that
speak some false and mostly true mimetic sentences […] Finally, there are unreliable narrators
whose mimetic sentences of necessity cannot all be taken as true, but the reader cannot decide
which are true and which false. This is the logical structure of unsurmountable contradiction”
(1981 [1960], 116–117).
2 Notable attempts to rethink and extend the notion of unreliability include Greta Olson (2003),
Bruno Zerweck (2001), Tamar Yacobi (1981, 2001), Ansgar Nünning (2008), Per Krogh Hansen
(2008) and a recent volume focusing on “trustworthiness” edited by Vera Nünning (2015). Many
articles build on or debate Nünning’s work on integrating narrative unreliability with cognitive
theory, Monika Fludernik’s work on “natural” narratology, and Dorrit Cohn’s on “discordant nar-
ration.” The present volume includes two contributions dealing with unreliability, one by Dan
Shen, the other by Per Krogh Hansen. I agree with Shen that rhetorical and cognitive approaches
in many ways converge rather than diverge although I would contend that “authorial audiences”
do not necessarily involve shared contextual or historical conditions, since values can have val-
iance across time and space. I also agree with Krogh Hansen that unreliability is not necessarily
intentional, but would argue that this ultimately relates to the “trustworthiness” of the narrator’s
character. My contribution adopts an analytical approach to describe narrators who generate
doubt so that narrative and character qualities can be discussed separately rather than lumped
together under “unreliability.”
3 “Narrative theorists […] debate whether unreliability is located in the reader, in the text, in
the author, or in some interrelation between them; whether the concept of the implied author is
more of a hindrance than a help in our understanding of unreliability; whether a naïve narrator’s
accurate but uncomprehending reports should be called unreliable narration, discordant narra-
tion, or something else” (Phelan 2007, 225).
 Beyond Unreliability: Resisting Naturalization of Normative Horizons   63

reliability judgment, phenomena such as doubt and faith are intertwined and
changeable. A declaration of faith or doubt is never final, and the debate about
tension always remains open. To discuss literary phenomena that affect a read-
er’s relation to a narrative’s significance, including problematic narrators, I wish
to introduce the notions of narrative faith and narrative doubt.

2 Narrative Faith
Maurice Merleau-Ponty described “perceptual faith” as a “conviction that there is
something, that there is the world, the idea of truth” (1968 [1964], 30). Whereas
Edmund Husserl began his phenomenological investigation by refusing to take
the “world” as a given, Merleau-Ponty reminded us that the mind’s reflection on
the world is preceded by a brute perception by the senses. That is, before philoso-
phers began questioning the notion of “world,” they functioned in the world with
the assumption that it exists, constituting a kind of natural faith. The “world,”
however, does not refer exclusively to nature and its objects. It also includes cre-
ative action that brings forth new objects that become part of the world. These
objects can be papyruses, scrolls, books or oral transmissions of stories that
record what we call a “narrative”—a telling that describes, remembers, reports,
suggests, comments on or otherwise refers to a sequence of events experienced
in the world.4 And if perceiving natural objects depends on perceptual faith, then
we might say that a measure of faith is involved in the apperception of narrative—
in the cognizing of its concretized form as speech or writing or image and then
turning it into a mental impression. In the simplest sense, “narrative faith” can
be understood as a parallel or complementary conviction to “perceptual faith”:
an assumption that narrative, regardless of its form, relates to something, to the
world, to some idea of truth.
This faith extends over both real and fictitious narratives and thus to the
realm of literature. In his work on the cognition of the literary work, Roman
Ingarden (1961 [1937], 290) suggested that when experienced aesthetically, objects
are reconstituted into “aesthetic objects” regardless of whether they are “real” or
“fictitious.” Later, Wolfgang Iser developed this line of thought by arguing that
the fictionalizing act actually connects the imaginary with the real by relating

4 Gerald Prince defines narrative as “a discourse representing one or more events” and “the
recounting of a series of situations and events” as well as “a context-bound exchange between
two parties” (2003 [1987], 58, 60).
64   David Stromberg

unverifiable facts and events as actual. Hence, he claimed, “fiction is a means

of telling us something about reality” (Iser 1980 [1978], 53). This conviction, I
believe, depends on an unarticulated faith: an unverifiable conviction that narra-
tive maintains a potential relevance to human life in the phenomenal world even
when it is the product of imagination.
Part of literature’s singular nature, then, is that it can communicate signifi-
cances that resonate in us as “true” in the world through what Félix Martínez-Bon-
ati (1992, 152) calls “quasipropositional assertions.” Such assertions are compa-
rable to what Ingarden (1973 [1931], 173) termed “quasi-judgments.” The “quasi”
nature of these assertions and judgments suggests that they do not fall neatly
either into the realm of the imaginary or the real. The telling of such events, char-
acters, circumstances or states of being can complicate narrative faith by leading
us to question the relevance of what is told to what we experience as reality in
the world.
This complication is often “patched over” by the effect of verisimilitude, that
is, plausibility or believability. For while readers of fiction do not assume that
reported events are strictly factual, they are nonetheless “interested in recogniz-
ing them as more or less ‘verisimilar’” (Eco 1984, 12). Structurally, verisimilitude
in literature rests not solely on a storyworld’s similarity to the experienced world,
since the world is experienced in a variety of ways, but on what Gérard Genette
(1980 [1972], 195) called a narrative’s “coherent context,” that is, a “code which
governs” the “course” of the narrative’s unfolding. We can accept the plausibility
of events and characters in a novel, no matter how “unrealistic” they may seem to
us based on our experience, if these events and characters develop in a way that
is coherent within the novel’s code. Beyond the structural code, verisimilitude in
novels also rests on the use of language, in other words, the linguistic coherence
of a narrative’s telling. In fiction, then, verisimilitude includes both the imagined
events, their order and form represented through language, and the discourse
and style used in the telling, including the linguistic expression and evocation of
internal processes such as cognition or emotion. Such use of language within a
coherent context, both structural and discursive, gives a literary work its overall
verisimilitude: the effect that allows us to relate to its imagined aspects in such
a way that we can glean from them significance that is “true” in what we call
This dynamic recalls the notion of the reader’s “willing suspension of dis-
belief” from Samuel Taylor Coleridge’s Biographia Literaria (1817). What is often
forgotten is that for Coleridge such “suspension” has a broader purpose: reaching
the “dramatic truth of […] emotions” through “poetic faith.” His description, often
shortened to mention only “willing suspension,” actually suggests that literature
does more than voluntarily tolerate the non-actuality of imagination: it uses this
 Beyond Unreliability: Resisting Naturalization of Normative Horizons   65

tolerance for the sake of truth of a different order. Such “poetic faith”—not to be
confused with trust—is comparable to narrative faith in that it points to the sig-
nificance of literature in the world. It expresses the overall dynamic of literature
as a process involving similitude, truth and faith.
Looking to move beyond what she saw as the strict limits of classical narra-
tology, Shlomith Rimmon-Kenan wrote of the possibility of moving beyond doubt
toward a faith that she described as a “spiraling movement that integrates skepti-
cism and yet glances beyond it,” thus “shifting the ground” and opening “access”
to reality “through narration” (1996, 128). This suggests that narrative gives us
access to reality through a faith that incorporates doubt while also “glancing”
beyond it toward the world. In a sense, narrative faith is a conviction that, regard-
less of its source in the imagination, literary narrative retains its relevance to both
the phenomenal world and to human life in that world.

3 Narrative Doubt
Opposing the effects of verisimilitude is the recurrent recognition of a literary nar-
rative’s illusory status. Narrative strategies that pose a challenge to readers (for
example, by emphasizing a distance between the narrator and the story) prob-
lematize verisimilitude by instigating questions about narration. They induce
hesitation between the believability of the illusion and the consciousness of
the story as invention. This hesitation is an important aspect of a literary work’s
relation to what we experience as reality.5 When specific narrative strategies are
used to instigate such hesitation, they can be considered a structural element. On
the one hand, since they problematize verisimilitude, such “narrational border
crossings” carry an “artistic cost” (Daleski 2009, 244). On the other hand, nar-
rational border crossings “demonstrat[e] the importance of the boundary they

5 Different kinds of audience hesitation have been registered as intrinsic to a variety of aesthetic
responses. Tzvetan Todorov (1975 [1970], 167) claims that hesitation is what distinguishes the
genre of the fantastic from the uncanny (étrange) and the marvelous; he goes on to note that
such hesitation questions “precisely the existence of an irreducible opposition between real and
unreal” (167). Leona Toker believes that hesitation about the factographic pact is a feature of the
poetics of documentary prose, adding that “[b]y placing us in the area of hesitation, the author
invites us to treat the stories as true […] and yet also to ask of them the kinds of aesthetic ques-
tions that are usually reserved for works of fiction” (1997, 200). In an essay titled “Music Discom-
posed,” Stanley Cavell suggests that “the dangers of fraudulence, and of trust, are essential to
the experience of art” (2002 [1969], 188). Similarly, the kind of hesitation I deal with takes shape
in interaction between narrative faith and narrative doubt.
66   David Stromberg

tax […] in defiance of verisimilitude” (Genette 1980 [1972], 236). By brushing up

against the border of the imagined and the real, such border crossings foreground
narrative tension.
Problematizing narrative techniques includes what Genette (195) calls “mod-
ulation.” This can refer to a shift either in focal position, which refers to the per-
spective from which the story is told, or in narrative level, which distinguishes
between a voice that is presumed to be inside or outside the story’s diegesis.
Such modulation can result from narrative “alterations” including paralepsis and
paralipsis (giving either more or less information, respectively) than is available
to the focal character at any given moment in the narrative. Or modulation can
result from metalepsis, a narrative “transgression” in which a voice from one
diegetic level crosses into another (as when a speaker understood to be outside
the storyworld engages directly with the story that is told; cf. Genette 1980 [1972],
234). Other problematizing narrative strategies can include “disnarrated” alterna-
tives (Prince 1992, chap. 3) included in the telling as possibilities but which do not
materialize; narrative “gaps” and “blanks” (Ingarden 1973 [1931]; Iser 1972) which
the reader is led to fill in either with information given by the text or by employ-
ing information from his or her own frame of reference; or “stumbling-blocks”
which create tension with the reader’s personal preformed expectations (Harri-
son 1993). Such narrative techniques, among others, often seem to deviate from a
narrative’s “coherent context,” thus problematizing verisimilitude and triggering
narrative doubt.
In literary criticism, doubt can lead to analytical activity that examines those
aspects of a narrative that were previously taken for granted. But doubt in a
novel’s narrative strategy can also seep into doubt about a narrative’s broader
significance. When narrative doubt extends beyond our experience of the con-
structed narrative agent—the narrator—into a precipitate judgment of the narra-
tive itself, this tends to undermine narrative faith. That is, doubt about the narra-
tor may spill over into doubt about the narrative. And since the narrative is said
to invoke the world, this doubt is liable to extend to the narrative’s interrelation
with experience in reality.6

6 Such attitudes may be behind criticism of Dostoevsky’s Demons as untidy (Magarshack 1959
[1953]) or as having glaring defects (Carr 1962 [1931]) as well as attacks on Camus’ The Plague
as “teach[ing] men to ignore history” (Barthes 1993 [1955]) and on Isaac Bashevis Singer’s The
Penitent as “lacking in […] invention” (Wisse 1996).
 Beyond Unreliability: Resisting Naturalization of Normative Horizons   67

4 Doubting Unreliability
Tamar Yacobi (1981, 113) questioned the notion of narrative unreliability by
asking a basic question: “Are reliability and unreliability value-judgments or
descriptions?”7 Twenty years later, she took this matter up again, arguing that
there is not necessarily “a linkage between the teller’s overall makeup […] and
the (un)reliability of the telling” (Yacobi 2001, 223). She thus made an initial dis-
tinction between the judgment of a narrator’s character and the judgment of a
narrative’s significance. She also pointed out that “the judgment of a narrative as
unreliable—or otherwise—is always [a] hypothetical move” (2001, 224). Calling it
an “unreliability hypothesis,” she framed the notion in terms of a question rather
than a solution.
In her analysis of Isaac Bashevis Singer’s story “Gimpel the Fool” (the story
of a village fool who all his life is deceived by his fellow villagers), Yacobi looks
beyond a single normative horizon to identify what she describes as a “hierarchy
of norms” (2001, 226): a cognitive level on which she identifies Gimpel as negative
(by dint of being deceivable), and a moral/ethical level on which she identifies
him as positive (first by not deceiving others and subsequently by refusing to seek
revenge on his neighbors). Hence, she argues that “Gimpel’s narration is better
integrated within a hypothesis of ultimate reliability—despite appearances—than
in terms of apparent folly” (2001, 226). So while Yacobi begins by questioning
how we think of unreliability, she returns to the concept of unreliability by “inte-
grating” narrative tension. Identifying a “hierarchy” of norms (rather than a field
of normative horizons, for example) comes from a lingering tendency to integrate
the various normative levels which she has identified. Faced with the doubt gen-
erated by Gimpel’s telling, she seeks to bring “discordant elements into pattern”
(2001, 224) by rehabilitating the narrator’s moral reliability.
Yacobi’s reading is complicated, however, by Gimpel’s own admissions of
his tacit better judgment: “To tell the truth, I knew very well that nothing of the
sort had happened” (Singer 1988 [1957], 5), he says of one of the lies. Elsewhere
he admits: “I realized I was going to be rooked” (7). And again, he problema-
tizes what is perceived as his gullibility: “To tell the plain truth, I didn’t believe
her” (9). Though he says he knows that Elka, the woman the town wants him to
marry, is a “whore,” he allows himself to fall under the town’s influence, placat-

7 While Yacobi uses this question as a springboard to further develop the notion of unreliability,
I contend that the very need to ask this question should make us reserve the term “unreliability”
only for those cases when we mean it as a judgment on the narrator’s norms (cognitive, moral,
ideological, etc.), and not as a description of narrative infractions.
68   David Stromberg

ing himself with such false notions as “when you’re married the husband’s the
master” and “you can’t pass through life unscathed” (6). His foolishness comes
less from others’ deception than from his refusal to take responsibility for what
he himself sees and understands—including twice finding other men in bed with
his wife. Gimpel is not merely being deceived by others. He is actually deceiving
We understand from Gimpel’s narrative that, after twenty years, he has actu-
ally resigned himself to a status quo of lies until, on her deathbed, his wife con-
fesses that his children are not his. This shocks him not because he did not know
this (he admits several times throughout that they were not his) but because she
shatters the lie with which he has been willing to live. The seeming cognitive
deficiencies that had relegated him to the status of “fool” come into question.
For the first time, Gimpel can no longer blame deception on someone else. It
is when Gimpel is faced with his own hypocrisy that his true moral makeup is
tested. His first reaction leads to potential revenge: he bakes his own urine into
the townspeople’s bread. But after he does this, his deceased wife comes to him in
a dream and exposes to him the most disturbing truth: “I never deceived anyone
but myself. I’m paying for it all” (19). It is at this point that Gimpel becomes aware
of his deepest crime: his own self-deception. There is no sense in taking revenge
on the townspeople because the guilty one is he himself. Gimpel decides to bury
the soiled bread.
The temptation of revenge which wells up in Gimpel when he is faced with
the truth, and his conscious choice to reject it, is what delineates the story’s
normative horizons. When the tension is preserved, we understand that Gimpel
must face the lies he has accepted under the guise of “fool”—including his own
responsibility in those circumstances—without projecting onto others the anger
this realization engenders. This is what turns the same behavior—tolerance of the
wickedness of others—from foolish to morally significant. The story, therefore,
only seems to privilege morality over cognitive competency. The challenge is to
be cognizant of one’s circumstances, even when injustice is perpetrated, and yet
continue to act ethically, taking responsibility for one’s own share in those cir-
cumstances. Gimpel’s success in rising to this challenge is underscored by the
fact that he cannot easily live down the full awareness of his self-deception: it
has such a profound effect on him that he is incapable of continuing to live in
the town. He leaves everything to the children and becomes a wandering beggar.
The ethical horizon that emerges when we refrain from a hierarchical “inte-
grated” analysis expands toward the difficulty of being both moral and cogni-
zant. Whereas Yacobi ends by endowing Gimpel with “ultimate reliability,” part
of the story’s power, I believe, is due to the tension that remains in the irreduci-
ble and irreconcilable distance between moral and cognitive understanding. The
 Beyond Unreliability: Resisting Naturalization of Normative Horizons   69

doubt raised by Gimpel’s narration prompts readers to ask these questions, and
by retaining this tension through narrative faith, we are able to move beyond reli-
ability to fathom the narrative’s broader significance. Hard as it is for us to accept
antithetical states, Gimpel is neither reliable nor unreliable. He remains flawed,
yet becomes enlightened. That his response to this epiphanic understanding is to
abandon everything further suggests the intensity of the moral-cognitive conflict
which cannot be easily integrated into a single interpretive solution. What makes
Gimpel no longer a fool is that he can no longer face the lie that is his life.

5 Beyond Unreliability
James Phelan, like Yacobi, betrays reluctance about reliability when introducing
the notion of “bonding unreliability,” which he describes as “unreliable narra-
tion that reduces the distance between the narrator and the authorial audience”
(2007, 223–224). He suggests that a single overall unreliable narrator can be alter-
nately “bonding” or “estranging,” depending on his or her distance at any given
moment from the implied author. But in principle, neither position—bonding or
estranging—need necessarily be “endorsed” by the implied author. That is, a nar-
rator can be both bonding and estranging without ever being aligned with the
implied author.8
Since unreliable narration is by nature estranging, Phelan seeks to deter-
mine whether a bond might be forged between the implied author and the autho-
rial audience, one that, for him, has a single “ideal” reaction.9 Alternatively, he
attributes what he calls “disparate responses” to the so-called flesh-and-blood
reader (2007, 223–225). But these varying responses can in fact be understood as
potentialities encoded in the narrative and thus as manifestations of a potential
“ideal” collective response.10 Each interpretation is an attempt to use the fecund
narrative to reach some sort of significance that is not concretely stated but that

8 Consider the first part of Hemingway’s To Have and Have Not (1937) in which Harry Morgan,
even in the moments he appears to bond with the reader (e. g., suspiciousness about the man
who chartered Morgan’s boat for three weeks without paying), does not represent sentiments
that the implied author can be imagined as sharing.
9 Phelan describes Rabinowitz’s notion of “authorial audience” as the “hypothetical, ideal au-
dience for whom the author constructs the text and who understands it perfectly” (1996, 215).
10 “To make a story good, it would seem, you must make it somewhat uncertain, somehow open
to variant readings, rather subject to the vagaries of intentional states, undetermined” (Bruner
1990, 54).
70   David Stromberg

nonetheless comes from within that same narrative. The inherent “alternatives
form a duality rather than an ambiguity” (Toker 1989, 218), and thus the authorial
audience can embrace these alternatives—the spectrum of meaning latent in the
narrative—no less than the flesh-and-blood reader.
Honing his definition of bonding unreliability, Phelan admits that this leads
to a paradox: “although the authorial audience recognizes the narrator’s unreli-
ability, that unreliability includes some communication that the implied author—
and thus the authorial audience—endorses” (2007, 225; emphasis added).11 Since
unreliability can hardly be a bonding force, however, what appeals to the reader
is rather this other unspecified aspect of communication which transpires via the
doubt-generating narrator.
Phelan’s answer in searching for the source of this bonding is to look to the
implied author. But this authorial entity, in addition to constituting the bonding
communication, has also conjured up the doubt-generating narrator who is no
less essential to the literary work in question than its fabula. That is, the author
is the source of this “bonding” communication no less than of the “estranging”
techniques that affect our attitude to the narrator. When we keep in mind that the
author, as the origin of aberrant textual signs, situates or retains them in the text
deliberately, we as readers and critics can question whether and how this formal
aberrance may be connected with a work’s thematic or ethical implications. The
estranging devices, then, remain an integral part of the narrative’s normative and
interpretive horizons. They are no less necessary to the dynamic of such a literary
work than the bonding devices, and no less promoted by the (implied) author.
What emerges is an authorial model with two poles, one bonding and the other
estranging, and an oscillation between them which provokes readers to ask ques-
Thus the bonding pole and the estranging pole are counterbalances, and
neither can necessarily stand alone as a normative measure. Together, they foster
a dynamic narrative that alternates between positions and opens upon a norma-
tive field with alternate horizons. Such a dynamic can overcome the stretching of
verisimilar boundaries, which might be read as unreliable, and foster a reclaimed
narrative faith which, I believe, is of a different nature from reliability. In that
sense, narrators who generate doubt are not reliable or unreliable: rather, being
reliable and unreliable in turn, they result in oscillation between bonding and
estrangement. Yet notions such as bonding and estrangement, like unreliability,
produce rhetorical effects with assumed normative values, a kind of built-in prej-

11 A similar construction is Yacobi’s view of Gimpel’s “ultimate reliability—despite appear­

ances” (2001, 226).
 Beyond Unreliability: Resisting Naturalization of Normative Horizons   71

udice which the doubt-faith paradigm aims to avoid. On recognizing this, the dis-
cussion turns from the narrator’s reliability vis-à-vis (implied) authorial norms to
the tension between the narrator (who is telling) and the narrative (what is told).
This tension is paramount to the significance of these kinds of literary works,
and any attempt to diffuse the tension through the use of naturalizing concepts
results in flattening the significance of the literary works. For if, despite the doubt
generated in and by a narrator, the reader nonetheless bonds to some aspect of
a narrative, this is for reasons other than reliability or unreliability—something
lying beyond the narrator’s reach. It is due to the literary dynamic of narrative.

6 Resisting Naturalization
The tendency to look to unreliability as a solution for narrational border crossings
came, in many ways, on the heels of research on narrative “naturalization,” a
principle that contributed to the cognitive turn in narrative theory which focuses
on how “readers relate what they read to ordinary human actions, motivations,
and behavioral scripts” (Olson 2003, 98).12 One form of such naturalization is
Yacobi’s five principles of “integration” which readers use to “resolv[e] textual
tensions […] referential difficulties, incongruities, or (self-) contradictions” (1981,
114).13 She singles out two principles in particular that are often used to integrate
such tension: the “genetic” principle, which refers to the author, and the “per-
spectival” principle, which refers to the narrator. If analysis deems such problems
accidental or as resulting from carelessness, the genetic principle is used, and
the author is regarded as lacking in artistry. If such tension is seen as a deliberate
formal aspect of the work, readers will seek to integrate it by assigning culpability
to the narrating agent, thus reconciling the story’s border crossings by attributing
them to a storyteller lacking one or another sort of competency. That is, in trying

12 Jonathan Culler wrote that “to naturalize a text is to bring it into relation with a type of disclo-
sure or model which is already, in some sense, natural and legible” (1992 [1975], 138).
13 The five principles of integration used by readers to “resolv[e] textual tensions […] referential
difficulties, incongruities, or (self-) contradictions” are as follow: genetic—causal factors that
produced the text without coming to form part of it (i. e., creative process, history of finished
product, situation of historical producer); generic—the genre of the text; existential—resolutions
in terms of the world; functional—the work’s aesthetic, thematic, and persuasive goals; and per-
spectival—the peculiarities and circumstances of the observer through whom the world is taken
to be refracted (i. e., narrated, experienced, evaluated) (Yacobi 1981, 114).
72   David Stromberg

to integrate the tension caused by literary phenomena such as narrational border

crossings, readers and scholars tend to judge the narrator as unreliable.
It is the push for naturalization that has, in my opinion, fueled over-depend-
ence on the notion of unreliability. In searching for an alternative to naturali-
zation, I propose complementing the concepts of narrative doubt and narrative
faith with a phenomenological attitude described by Husserl as “bracketing,”:
the attempt to “abstain from any […] opinions, judgments, and valuations” of
the narrator while reading without foregoing a narratological evaluation of its
strategies and techniques (1964 [1931], 7–8).14 This “abstention” is a temporary
suspension for the sake of a more careful perception of a work’s tension. In the
framework of narratological inquiry, such an attitude might be called structural
bracketing. This does not mean eschewing an evaluation of either the characters,
the narrator or the narrative. Rather, it aims to avoid judging narrational border
crossings prematurely by assigning blame either to a would-be unreliable narra-
tor or to an inept author and thus to take into consideration the complexities of
the relationship between structure and theme using a dynamic paradigm. The
variety of narrational border crossings already mentioned can be discussed not in
terms of a narrator’s unreliability but in those of the doubt generated by a given
narrator and the influence of this doubt on a reader’s narrative faith.
To some extent, structural bracketing is tantamount to a kind of “negative
capability” described by John Keats in which one tolerates “uncertainties, Myster-
ies, doubts, without an irritable reaching after facts & reason” (2006 [1817], 942).
Whereas Keats recommended this for poets, I am putting it forth as a strategy for
readers and critics. In exploring a variety of narrational border crossings, such
an attitude aims to purge narrative modulation of the normative, and particu-
larly of the moral or cognitive, presuppositions found in discourse on narrative
unreliability. This approach has certainly been recognized and applied in literary
criticism before, but I suggest that it be extended to assessing narrative modu-
lations without necessarily naturalizing them, thus retaining literary elements
in tension while restraining the impulse to resolve them. If we assess narrative
tension within a non-naturalized conception of narrative so as to behold unpat-
terned elements and restrain the impulse to integrate or resolve them, we might
find this departure from interpretive convention in some ways more rewarding,
even if disconcerting and destabilizing in other ways.

14 This term is adapted from Husserl’s “epoché,” which comes from “suspension” in the ancient
Greek and was also used by Descartes. Husserl also uses the term “reduction” to describe a sim-
ilar process. For a distinction between bracketing and the “natural attitude,” see Husserl (1964
[1929], 14–15).
 Beyond Unreliability: Resisting Naturalization of Normative Horizons   73

7 Preserving Tension
If, despite a narrator who generates doubt, we detect a different type of communi-
cation taking place through the same text, the work as a whole can retain our nar-
rative faith and thus its claim of relevance to reality. One already much discussed
model of this kind of process can be exemplified by Nabokov’s Lolita, where the
reader who carefully parses its textual signals may see not merely a compulsive
pedophile’s crafty and entrapping confession of his perversion15 but also “a sen-
sitive representation of the human price that such an obsession exacts” (Toker
1989, 207). If we are prepared to situate these textual signals within a larger nor-
mative context (aesthetic, cognitive, moral, ideological, etc.), we may be able to
retain the narrative faith which such a text evokes and also partly subverts.
The doubt generated by a narrator implies some preexistent faith—a belief,
without proof or justification, in the narrative at hand as working toward a truth
relevant to the phenomenal world. And in tracing how a reader might nonethe-
less believe in, rather than just believe, the narrative of a doubt-generating nar-
rator, a narrative can shake our faith and yet reclaim it. When we understand
that “[t]he same strategies, differently used, serve both the destabilization of rep-
resentation and subjectivity and their rehabilitation” (Rimmon-Kenan 1996, 130),
we can perhaps map the way in which “various inconsistencies of [a] method”
which we think “would fragment [a] novel” actually “help to bind it together”
(Daleski 1985, 16). Rather than create an “unnecessary problem for the reader”
by implying that the “only recourse, in face of the ‘host of different perspectives’
that confront him, is to try to ‘make them consistent’,” we come to understand
that sometimes “it is quite impossible to reconcile the contradictions thrown up
by [a] narrative” (16). Indeed, by “accept[ing] the blatant inconsistencies,” we
are led back to the question with which we started: “why [is it] that a novelist […]
should go out of his way to force inconsistencies on us?” (16). By preserving the
irreconcilabilities of a narrative, we may perceive more of the horizon of a literary
work’s norms along with the tensions created by their conflicting interplay—ten-
sions which, if naturalized, integrated or resolved, may limit a work’s interpretive
The motivation behind such a turn is to avoid the interpretive risk of reacting
to a narrative’s problematic textual signals without duly considering their various
implications and connections in respect to the literary work’s form and multitu-
dinous normative layers. Instead, it promotes the discernment of problematizing

15 “[T]he audience […] begins to derive a pleasure from the account of the pursuit of ecstasy and
to ignore the price of this pursuit.” (Toker 1989, 202)
74   David Stromberg

techniques without extending a narrator’s moral makeup to the significance of

the narrative, thereby facilitating the reader’s communication with the (implied)
author despite or through the doubt-generating narrator and reclaiming narrative
faith in the course of this dialogue.
With narrative faith, “obstacles to reliability in the classical sense [can]
become assets” (Rimmon-Kenan 1996, 25). My aim in suggesting this provision-
ally neutral method of formal analysis is to discuss the specificities of the relation-
ship between a narrative’s form and the worldview that this narrative constructs.
Analysis of this interrelation makes it possible for a narrative’s significance to
survive the doubt generated by narrational border crossings and to justify, in a
way that is no longer naïve but rather complex, the narrative faith that a literary
work may have initially engendered.16

Works Cited
Barthes, Roland. 1993 [1955]. “La Peste, annales d’une épidémie ou roman de la solitude.” In
R. B. Œuvres complètes, I, edited and introduced by Éric Marty, 540–545. Paris: Gallimard.
Booth, Wayne. 1983 [1961]. The Rhetoric of Fiction. 2nd edition. Chicago, IL: University of
Chicago Press.
Bruner, Jerome. 1990. Acts of Meaning. Cambridge, MA: Harvard University Press.
Carr, E. H. 1962 [1931]. Dostoevsky 1821–1881. London: Unwin Books.
Cavell, Stanley. 2002 [1969]. “Music Discomposed.” In S. C. Must We Mean What We Say?,
180–201. Cambridge: Cambridge University Press.
Coleridge, Samuel Taylor. 1817. “Chapter XIV.” Biographia Literaria. http://www.english.upenn.
Culler, Jonathan. 1992 [1975]. Structuralist Poetics: Structuralism, Linguistics, and the Study of
Literature. London: Routledge and Kegan Paul.
Daleski, H. M. 1985. Unities. Athens: University of Georgia Press.
Daleski, H. M. 2009. “Narrational Border Crossings in Major Early-Twentieth-Century English
Novels.” Poetics Today 30 (2): 237–255.
Eco, Umberto. 1984 [1979]. The Role of the Reader: Explorations in the Semiotics of the Text.
Bloomington: Indiana University Press.
Genette, Gérard. 1980 [1972]. Narrative Discourse: An Essay in Method. Translated by Jane E.
Lewin. Foreword by Jonathan Culler. Ithaca, NY: Cornell University Press.
Harrison, Bernard. 1993. “Gaps and Stumbling Blocks in Fielding: A Response to Černy,
Hammond and Hudson.” Connotations 3 (2): 147–172.
Husserl, Edmund. 1964 [1929]. The Paris Lectures. Translated by Peter Koestenbaum. The
Hague: Martinus Nijhoff.

16 The author wishes to acknowledge the guidance of Professor Leona Toker. This narratological
departure would not have reached the present destination without her knowledge and wisdom.
 Beyond Unreliability: Resisting Naturalization of Normative Horizons   75

Ingarden, Roman. 1961 [1937]. “Aesthetic Experience and Aesthetic Object.” Philosophy and
Phenomenological Research 21 (3): 289–313.
Ingarden, Roman. 1979 [1931]. The Literary Work of Art: An Investigation of the Borderlines
of Ontology, Logic and the Theory of Language. Translated by George G. Grabowicz.
Evanston, IL: Northwestern University Press.
Iser, Wolfgang. 1972. “The Reading Process: A Phenomenological Approach.” New Literary
History 3 (2): 279–299.
Iser, Wolfgang. 1980 [1978]. The Act of Reading. Baltimore, MD: The Johns Hopkins University
Keats, John. 2006 [1817]. “Letter to George and Thomas Keats.” In The Norton Anthology of
English Literature, Eight Edition, Vol. 2, edited by Stephen Greenblatt et al., 942. New York:
Magarshack, David. 1959 [1953]. “Translator’s Introduction.” In Fyodor Dostoevsky. The Devils,
vii–xvii. Harmondsworth: Penguin.
Martínez-Bonati, Félix. 1981 [1964]. Fictive Discourse and the Structures of Literature.
Translated by Philip W. Silver. Ithaca, NY: Cornell University Press.
Martínez-Bonati, Félix. 1992. Don Quixote and the Poetics of the Novel. Translated by Dian Fox.
Ithaca, NY: Cornell University Press.
Merleau-Ponty, Maurice. 1968 [1964]. The Visible and the Invisible. Edited by Claude Lefort,
translated by Alphonso Lingis. Evanston, IL: Northwestern University Press.
Nünning, Ansgar. 2008. “Reconceptualizing the Theory, History and Generic Scope of
Unreliable Narration: Towards a Synthesis of Cognitive and Rhetorical Approaches.” In
Narrative Reliability in the Twenty-First Century First-Person Novel, edited by Elke D’hoker
and Gunther Martens, 29–76. Berlin and New York: De Gruyter.
Nünning Vera, ed. 2015. Unreliable Narration and Trustworthiness: Intermedial and
Interdisciplinary Perspectives. Berlin and New York: De Gruyter.
Olson, Greta. 2003. “Reconsidering Unreliability: Fallible and Untrustworthy Narrators.”
Narrative 11 (1): 93–109.
Phelan, James. 1996. Narrative as Rhetoric: Technique, Audiences, Ethics, Ideology. Columbus:
The Ohio State University Press.
Phelan, James. 2007. “Estranging Unreliability, Bonding Unreliability, and the Ethics of Lolita.”
Narrative 15 (2): 222–238.
Prince, Gerald. 1992. Narrative as Theme. Lincoln: University of Nebraska Press.
Prince, Gerald. 2003 [1987]. Dictionary of Narratology. Revised edition. Lincoln and London:
University of Nebraska Press.
Rimmon-Kenan, Shlomith. 1996. A Glance Beyond Doubt. Columbus: The Ohio State University
Singer, Isaac Bashevis. 1988 [1953]. “Gimpel the Fool.” Translated by Saul Bellow. In I. B. S.
Gimpel the Fool and Other Stories, 3–22. New York: Noonday Press.
Todorov, Tzvetan. 1975 [1970]. The Fantastic: A Structural Approach to a Literary Genre.
Translated by Richard Howard. Ithaca, NY: Cornell University Press.
Toker, Leona. 1989. Nabokov: the Mystery of Literary Structures. Ithaca, NY: Cornell University
Toker, Leona. 1997. “Toward a Poetics of Documentary Prose—From the Perspective of Gulag
Testimonies.” Poetics Today 18 (2): 187–222.
Wisse, Ruth R. 1996. “Singer’s Paradoxical Progress.” In Critical Essays, edited by Grace Farrell,
102–112. New York: G. K. Hall.
76   David Stromberg

Yacobi, Tamar. 1981. “Fictional Reliability as a Communicative Problem.” Poetics Today 2 (2):
Yacobi, Tamar. 2001. “Package Deals in Fictional Narrative: The Case of the Narrator’s (Un)
Reliability.” Narrative 9 (2): 223–229.
Zerweck, Bruno. 2001. “Historicizing Unreliable Narration: Unreliability and Cultural Discourse
in Narrative Fiction.” Style 35 (1): 151–178.
Valery Timofeev (Saint Petersburg)
Nabokov’s “Ultima Thule”:
An Exercise in Generative Narratology

1 Introduction
The main part of my paper is a close reading of the first page of Vladimir Nabok-
ov’s “Ultima Thule,” published in Russian in 1942. This reading will at the same
time be an exercise in generative narratology as a framework for analyzing fiction
whose aim is to show fiction in its making. What underlies the Generative Narra-
tology Framework is not a single and unique theory, but rather a complex set of
various interdisciplinary approaches. The overall cohesiveness of this approach
is a result of my thirty years of research in literary theory, narratology, reader-ori-
ented criticism and cognitive studies implementing multiple approaches and
methodologies for analyzing English and Russian fictional texts.
I am well aware that any description of the authorial creative process inferred
from reading a text in which the text is the sole foundation for insights into its
author’s mind is nothing more than an example of critical speculation, being,
by definition, hypothetical. Understanding what others think involves complex
operations (cf. Dimaggio et al. 2008; Lane and Schwartz 1987). Attempts to infer
what the author was thinking while creating the text require comprehension of
even the tiniest signals the text might contain to indicate the author’s motiva-
The main challenge, as I see it, is the need to distinguish between our per-
spective as readers and the narrator’s perspective. Naturally, the narrator’s per-
spective is to be distanced from that of the author, as well. While the narrator’s
motivations, mental states, emotions, etc. might be quite explicit, describing
them as related to a distinct cognitive process may be quite erroneous, since a
narrator is but a phantasm created by the author.
Though my reading of Nabokov’s text is an exercise in metacognition, right-
fully “referred to as fuzzy by many scholars” in an informative literature review
on the subject (Akturk and Sahin 2011, 3731), it is meant to reconstruct, or rather
construct, a hypothetical version of the way the creative process is organized in
terms of monitoring, regulating and controlling narrative progression.

DOI 10.1515/9783110555158-005
78   Valery Timofeev

1.1 Authorial self-consciousness

Since reading as mindreading aims, figuratively speaking, at gaining access to

authorial self-consciousness, the latter needs to be defined, if it is to become an
operational concept. I suggest that a person who experiences authorial self-con-
sciousness needs specific abilities that go far beyond the phenomenal or first
level of consciousness, which is sufficient only for primary self-consciousness
(cf. Fletcher et al. 1995; Vogeley et al. 2001; Kircher and Leube 2003). One of these
abilities is ‘concept possession’, whereby the subject is aware of him/herself as an
author. The hypothesis that concept possession is essential for primary self-con-
sciousness, as advocated by Savanah (2012), provoked a lengthy discussion in
Consciousness and Cognition 21 (2012) as to whether this concept extends beyond
primary self-consciousness. The authorial notion is obviously far more complex
than a set of mineness features acquired by humans during their first few years of
life. Authorship is historically variable, especially in those elements of its content
structure that are shared by authors linked by similar aesthetic background
which, to begin with, includes the aesthetic, not ontological, parameters of the
author’s status (e. g., how authors understand their mission, how the fictional
world they create correlates with the world they live in, etc.1). The conceptual
facets related to authorial self-consciousness must be divided into two separate
sectors: self-reflexiveness and internal state awareness. This part of the overall
hypothesis complies with most of the studies that use the self-consciousness scale
(e. g., Fenigstein et al. 1975). Theories of the two sectors of conceptual facets prove
to be very helpful for explaining the forms and features of I-thoughts, as shown
by Kristina Musholt (2013, 653–657), although I consider the distinction between
self-reflexiveness and internal state awareness to be incomplete, especially when
dealing with authorial self-consciousness. We thus need to contrast two modes of
thought about thinking: the introspective mode and the reflexive mode. In what
follows, I will explain why I consider these two modes to be important, both for
the theory of authorial self-consciousness and its practical implementations in
the field of literary analysis.

1 A list of the elements contained in the concept of authorial self-consciousness is discussed in
detail elsewhere (Timofeev 1999).
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    79

1.2 Introspection vs. reflection

The basic idea that underlies the hypothesis of this section arises from what Witt-
genstein (1958) described as the necessity to distinguish between two different
types of I-thoughts: I-as-subject and I-as-object (1958, 66–67). I argue that the
introspective mode of thinking about one’s own thinking—no matter how we use
the first-person pronoun I-as-subject or I-as-object—does not allow the I we are
thinking about to be defamiliarized. The same is true in cases of mindreading
when we think about the thinking of another person, no matter which pronoun
we use: I or you. It might result from empathy, to use a psychological term, or from
some other factor such as “the paradox of self-consciousness” (Bermúdez 1998).
When we think introspectively, we do not, or cannot, use our ability to differenti-
ate between our own perspective and that of someone else we are thinking about.
That, on the one hand, often leads to misidentification errors while on the other,
it helps us to be emotionally engaged when reading fiction. The reflexive mode of
thinking, on the contrary, starts with self-estrangement or defamiliarization. No
matter whose thinking we are thinking about, no matter what pronoun (I or you)
we use when describing the process, we tend to treat both the thought and the
thinking subject as at least partly estranged from us. In other words, the intro-
spective mode would unite and homogenize whereas the reflexive mode tends
to decompose. Content and form would be dealt with separately by a reflexive
thought. This might be explained in terms of conceptual blending: when we use
the introspective mode of thinking, the elements at work are blended in a process
known as the basic blending network (Schneider 2012). When we apply the reflex-
ive mode to think about thinking, not only do we use the results of blending, but
we also observe the process of blending, paying special attention to the fact that
some elements belong to one input space while others belong to another. I will
provide numerous examples of the reflexive mode of thinking at work, or reflex-
ive narrative, later in this paper. The notion I use here is close to David Herman’s
approach (2004, 2006), where narrative reflexivity is treated as a cognitive instru-
ment used to “constitute, along with representational properties of the medium,
the interpretive activity of readers, and other factors, a gestalt or system for dis-
tributing intelligence that is ‘smarter’ than the sum of its parts” (2006, 374).
When writing fiction, authors adopt both modes of thinking. Most people
also use both modes although for some reason not everyone is capable of main-
taining the reflexive mode while engaged in the introspective mode. Moreo-
ver, some authors use the reflexive mode more extensively than others, with
Cervantes, Shakespeare, Sterne, Joyce, Pirandello and Nabokov figuring in the
reflexive mode’s Top-10 list. Reflexive authorial self-consciousness seems to
prosper more in some historical periods than in others. Some aesthetic systems
80   Valery Timofeev

or literary movements, such as Naturalism or Realism, tend not to tolerate reflex-

ivity whereas Romanticism, especially in Germany, cannot be imagined without
the prominent role granted to reflexivity. Postmodernism obviously thrives on
reflexivity, as well.
Alongside individual or historical choices in favor of reflexivity, as some
critics would put it, there are certain technical reasons for preferring the intro-
spective mode. Introspection is required to create characters, since this is a clear
case of mindreading. Here, reflexivity would be of no use: imagine Leo Tolstoy
paying a short but reflexive glance at himself in the middle of working through
an internal monologue of the young Natasha Rostova. I can well imagine Tolstoy
bursting out in laughter at the glimpse from aside of an overweight middle-aged
gentleman mentally armed and dressed as a girl in her teens!
An entirely new set of questions arises if the correlation of authorial self-con-
sciousness and the narrator’s self-reflexivity is to be studied. It seems quite
improbable for a self-reflecting narrator to be created by an author lacking a good
amount of experience of thinking in the reflexive mode. A self-reflecting narrator,
like the one who narrates in Nabokov’s “Ultima Thule,” should explicitly demon-
strate habitual self-estrangement. As it will be shown in the next section of the
paper, Sineusov controls every word of his narration and scrutinizes every idea
that comes to his mind, constantly looking at himself from aside and providing
us with perfect examples of the reflexive mode of thinking about thinking. We can
see this, for example, when Sineusov addresses his recently deceased wife. In the
one-page narration of the opening episode I have chosen for analysis, he manages
to ask her only a couple of questions. The rest, which is approximately seven-
ty-five per cent of the text, is a series of comments on his own comments, multiple
analyses of his own speech, its lexis and grammar, free association games, bitter
irony followed by regrets, etc. Nabokov himself is no less self-reflexive than the
Sineusov he creates. In no way, however, is the story a case of self-representation:
Nabokov shares no more of his own emotional life experience with his charac-
ter than Leo Tolstoy (Lev Tolstoj) did imagining Nataša Rostova getting ready for
her first ball. Nabokov might compete with the Almighty Creator as most other
self-reflexive authors do. James Joyce enjoyed the if-God-were-me-his-act-of-crea-
tion-would-result-in-a novel game as well as the if-Homer-were-me game. Authors
cannot create aux nihil: they need material. What Nabokov used as material to
design Sineusov and his narration were not specific memories or certain states of
mind, but mental experience. Patterns were distilled out of experience and pro-
vided to Sineusov—as well as to all other narrators of his fiction—to be exercised
as drills. That means that if we enter the “grammar” of the narrator’s thought, we
might believe we are observing the way the text was being generated. In that case,
though, the pleasurable mental state that emerges during the reception of fiction
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    81

known as aesthetic illusion might be slightly undermined. Readers will have to be

immersed in their own interpretive activity no less than in the represented world
of the story. The usual position of the aesthetic illusion lies between the two poles
“of total rational distance (disinterested ‘observation’ of an artifact as such) and
complete immersion in the presented world (‘psychological participation’) in
the represented world” (Wolf 2014 [2011], 270). When the Generative Narratology
Framework is used, the position of the aesthetic illusion on the scale between the
two poles will definitely move towards the pole of observation due to the special
attention paid to interpretive activity, the instruments we use for thinking. The
amount of pleasure will stay intact. We might lose something in terms of psycho-
logical participation while we will gain a lot intellectually.
Though my observations are made with a certain degree of approximation, I
believe that the following section will shed light on these issues.

2 “Ultima Thule”
The original Russian text was written by Vladimir Nabokov in 1940 and first pub-
lished in 1942. The story was translated into English by Dmitri Nabokov with the
author’s collaboration in 1973 and published in The New Yorker. “Ultima Thule”
and “Solus Rex” were supposed to be chapters of a novel, but the project was
never realized.
The narrator of “Ultima Thule” is Sineusov, an artist who is heart-broken by
the death of his wife. Sineusov imagines himself talking to his late beloved wife.
The story seems, at least at the outset, to be a set of “voluntary memories” as con-
trasted with Marcel’s involuntary memories in Proust’s Remembrance of Things
Passed. The first phrase of the story might be seen as a clear allusion to the “made-
leine episode” in Proust. “Ultima Thule,” though full of references and allusions
to texts from all over the world, is quite subtle in its treatment of Proust’s theme
of achieving eternal life by turning le temps perdu into art in a much more concise
yet more sophisticated way. The structure of the story can be described as a set
of invertible matrëškas (Russian dolls) playing strange games when the smallest
one turns inside out so as to envelope all the others, as in a number of key phrases
in the story: “the dream within a dream (when you dream you have awakened)”;
or “If you don’t remember, then I remember for you; the memory of you can pass
[…] for your memory” (Nabokov 1973 [1940], 38). The same trick occurs with the
idea of the Other World. Passing through several stages, it is transformed first
into the Christian concepts of good and evil, then into Dante’s version of hell
mixed with paradise only to turn into eternal aesthetic reality while remaining
82   Valery Timofeev

an ironic parody of Plato’s cave allegory. Falter, Sineusov’s former tutor, claims
that a momentous truth has been revealed to him, just as philosophers claimed
in Plato’s Republic, while Sineusov, an artist who is incapable of seeing anything
but the shadows of the essential Ideas, should be ridiculed and banned from the
ideal world. Plato’s concept of art’s inferiority to life, due to its distance from
Ideas, is clearly opposed by the Romantic concept according to which artists are
the only ones to have the “true” worldview. The romantic notion is referred to
several times in “Ultima Thule,” though somewhat ironically. Other motifs and
key ideas, including those of good and evil, to say nothing of other poetic com-
monplaces such as muses, mermaids, etc., reveal similar invertibility, playing a
cat-and-mouse game with the readers of the story.

Do you remember the day you and I were lunching (partaking of nourishment) a couple of
years before your death?2

[Помнишь, мы как-то завтракали (принимали пищу) года за два до твоей смерти?]3

In the opening lines, our narrator is appealing to his addressee, i. e., the narratee,
who is apparently extremely unreliable. Standard narratology has studied unre-
liable narrators in depth; here, however, we are confronted with an unreliable
addressee. There are serious doubts concerning her memory, so even zavtrakali
(lunching: the Russian original actually means ‘having breakfast’) needs to be
explained: prinimali pišču (partaking of nourishment). One cannot but be struck
by the addressee’s unreliability on reading the phrase “a couple of years before
your death.” This means that the addressee is not a living being! No sooner has
the reader started occupying the addressee position, which is “You,” than “your
death” halts the process. This is the first reason why it is a problem for the reader
to enter the narrated world.
The second barrier is space. The spatial coordinates remain uncertain for
quite a long time, thus keeping the reader in suspense. The temporal coordinates
remain unspecified, as well. The “now” in the opening phrase immediately turns
into the departing point of a dizzying process. The incident in question took place
“a couple of years before your death,” but nothing is stated about how long ago
the addressee died. The reader is informed of the fact, but not of the time, and, as
we are soon to learn, the addressee can remember nothing.

2 All English quotes of “Ultima Thule” are from Nabokov (1973 [1942], 38).
3 All Russian quotes of “Ultima Thule” are from Nabokov (2000 [1942], 113–115).
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    83

Assuming, of course, that memory can live without its headdress?

[Если, конечно, память может жить без головного убора.]

Suspicion about lack of memory is made explicit here, since memory may not
be able to survive without a head. Curiously, the word “head” is replaced with
“headdress” for a reason we are as yet unable to see.

Let us imagine—just an ‘apropositional’ thought—some totally new handbook of epistolary


[Кстатическая мысль: вообразим новейший письмовник.]

What is this ‘apropositional’ thought? Here is where the process of cognitive

blending enters the picture. There must be two related entities (or input spaces)
involved, two beings, neither of which is a “handbook of epistolary samples.”
What has one had to look at so that the handbook-of-epistolary-samples idea can
come to mind? All we have had so far is the memory that cannot live without its
headdress. What does “without its headdress” have to do with a handbook of
sample letters? What do they have in common? What does “without a headdress”
actually mean?
The idea of “some totally new handbook of epistolary samples” was intro-
duced as an “‘apropositional’ thought.” In the Russian original we find kstatičes-
kaja mysl’, an authorial neologism coined in violation of the morphological
norms of the Russian language, since adjectives are not formed in this way and
since the adjective in question does not exist. This is a perfect example of the
switching-on of the estrangement effect, a phenomenon defined by the Russian
formalists as a device used for shifting the reader’s attention from content to lan-
guage. By introducing a piece of ambiguous information, the author adds new
barriers to understanding that keep the narrated world detached, thereby forcing
the reader to perceive it rationally rather than emotionally. Each time we come
across such a coinage in the text, we face a flash of hypertrophied reflection, by
which I mean a thought directed towards itself, aiming at the very same thought.
The question is: what exactly is this thought targeted at if it triggers the idea of a
totally new handbook of epistolary samples?
The reader is prompted to follow the narrator’s reflection so as to find out the
origin of the ‘apropositional’ thought. In this process, the estrangement effect
shifts the reader’s attention from content to language. This means that the source
of the ‘apropositional’ thought should be looked for in the sphere of language.
This source may be a part of speech or a lexical unit. In terms of lexis, what is
84   Valery Timofeev

meant by “living without a headdress”? Headdress is a euphemism, because

what does a headdress have literally to do with memory? For memory, the head is
relevant: memory is “dressed in head,” so to speak, so that in fact bez golovnogo
ubora (without its headdress) is a noun phrase with “head” being an attribute to
the noun “dress” (the Russian golovnoj is an adjective): “head dress” is “a dress of
head,” “dress made of head.” This is what memory cannot live without. Instead,
we are told that memory cannot do without a headdress.
What, then, does this euphemism mean? How is it related to a handbook of
epistolary samples (pis’movnik)? Euphemisms are subject to two sets of rules. The
first consists of language rules, since a euphemism is a linguistic phenomenon,
a figure of speech. In Žit’ bez golovnogo ubora (to live without a headdress), as
in bez golovy (without a head), the correct case must be retained. The other set
of rules, which for euphemisms is much more important, is comprised of ethical
norms. Under certain circumstances, we are disinclined to pronounce a certain
word or phrase and say something else, particularly for ethical reasons or as a
matter of decorum.
The same goes for a “handbook of epistolary samples” (novejšij pis’movnik):
there are linguistic rules as well as ethical rules to be taken into account. In fact,
the purpose of such a handbook is to teach epistolary etiquette, to instruct readers
on how to express themselves properly. The examples given are quite monstrous.

To a lady who has lost her right hand: I kiss your ellipsis.

[К безрукому: крепко жму вашу (многоточие).]

Since there is no hand, there is an ellipsis instead.

To a deceased: Respecterfully yours.

[К покойнику: призрачно ваш.]

Quite elegant, isn’t it? That is also a kind of euphemism. Etiquette is followed and
at the same time morbidly violated.

But enough of these sheepish vignettes.

[Но оставим эти виноватые виньетки.]

Once again we face a reflexive mode of thinking. Being reversive by definition, it

works in two directions: it looks at what has just been said, drawing a conclusion:
“enough of these sheepish vignettes.” In the original, there is a pun here, with
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    85

repetition of the vin- element, which stands for “guilt”: vinovatye vin’etki. This
repetition shifts the meaning. Vignettes normally have nothing to do with guilt,
but accompanied by vinovatye (guilty) they acquire guilt. What are the vignettes
in question? According to the text, they stand for a surplus illustration, an extra
and unnecessary ornament which violates some ethical parameters and are thus
related to guilt. As a result, we have another euphemism, a version of how ethics
meets linguistics: vinovatye vin’etki. Social conventions that underlie the euphe-
mism and the epistolary samples are deautomatized and estranged, to use the
Russian formalist terms. Conventions that are to be obeyed automatically in our
case are exposed so as to attract attention. The conventions are still at work, but
they work unconventionally. Instead of camouflaging things for good reasons,
they reveal their covers and their dresses are stripped off, similarly to Hans Chris-
tian Andersen’s “the king’s new clothes.” The emotions hinted at here belong to
the narrator, or rather to one fraction of his consciousness, split due to the reflex-
ive mode of thinking. The narrated world is still kept detached, preventing the
reader from feeling involved or affected. Textually, the vignettes are unnecessary
extras, and guilt is a moral evaluation of such an ornament.
To see how the reflexive mode of thinking works, let us imagine a dragon with
three heads since, in terms of the blending theory, two channels or input spaces
are involved. While blending theory describes how two ideas merge, generative
narratology seeks to explain how two ideas meet in such a way that a third idea
emerges. One “head” says something while another looks on, assessing what has
been said and adding remarks, whereupon the third “head” takes its turn.

If you don’t remember, then I remember for you; the memory of you can pass—grammati-
cally speaking, at least—for your memory, and I am perfectly willing to grant for the sake of
an ornate phrase that if, after your death, I and the world still endure, it is only because you
recollect the world and me.

[Если ты не помнишь, то я за тебя помню: память о тебе может сойти, хотя бы

грамматически, за твою память и ради крашеного слова вполне могу допустить, что
если после твоей смерти я и мир еще существуем, то лишь благодаря тому, что ты
мир и меня вспоминаешь.]

This passage is quite complicated. Toward the end is an image that is almost a
parody of the Romantic type of worldview, with reality split in two. The first reality
is that of the primitive empirical world of trivial physical phenomena; the second
is represented by the magical, superior, true and invisible essence behind it, or
rather above it. Irony is another important feature of Romanticism we unmistak-
ably perceive in the Nabokov text: radi krašenogo slova (“for the sake of an ornate
phrase,” or literally: “for the sake of a painted word”) is a perfect example of this.
86   Valery Timofeev

The upshot of it is that the world exists only due to the fact that we live in a dead
person’s memory. This is the Romantic outlook, turned upside down!
Radi krašenogo slova (for the sake of a painted word) is the author’s neol-
ogism. It is formed out of the deformation of an idiom, as though a non-native
speaker of Russian were translating it backwards from memory. The original
Russian idiom is “ради красного словца,” radi krasnogo slovca (“for the sake
of a red little word” in which “red” archaically stands for “beautiful”). Slovco is a
diminutive, turning “a word” into “a small word,” as slovo (word) sounds much
more respectful. Krasnoe (red) is the archaic for krasivoe (beautiful), as diction-
aries of Russian normally inform us, although the modern meaning of the word
krasnoe (red) refers mainly to the color. In order to become red, the word needs to
be painted, so to speak. Here, the logic takes us in the opposite direction. As men-
tioned above, blending involves information going through two channels which
somehow merge to let a third meaning appear. In the “painted word” passage, by
contrast, it is the third (resulting) meaning that splits and runs along two chan-
nels, up-stream rather than down-stream. Thus, we see that the two currents of
information are not of the same origin.

If you don’t remember, then I remember for you; the memory of you can pass—grammati-
cally speaking, at least—for your memory.

[Если ты не помнишь, то я за тебя помню: память о тебе может сойти, хотя бы

грамматически, за твою память.]

Is it possible for “the memory of you” to pass grammatically for “your memory,”
the two wordings being evidently different grammatically, both in Russian and
in English? Their semantic difference is grammatically regulated by the morpho-
logical relations of cases in Russian and by the parts of speech in English. This
suggests that our narrator’s mind bears the notions of “memory of you” and “your
memory” in some language other than Russian or English. Latin will do perfectly:
memoria tui. It is in Latin that “the memory of you” and “your memory” share the
same form. The narrator is most likely to stand in front of the inscription in Latin,
or he imagines beholding the inscription. Since the addressee is a dead person,
we might suppose that our narrator is either imagining himself being at a ceme-
tery or he is actually there.
The memoria tui motif is also a complex case of simultaneous usage of two
modes of thinking: the self-reflexive and the introspective. Normally, we might
need two thinking subjects, two heads engaged in thinking at the same time. It is
stated in a pointed manner, through the use of a euphemism, that the addressee
has no head of her own. This means that we need something like a head-within-
a-head-within-a-head, a matrëška structure capable of changing places or invert-
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    87

ing so that the smaller structure can develop the larger one. Nabokov is known
for his Matrëška structures (cf. Davydov 2004), but, as we have suggested, these
matrëškas are able to turn inside out. This is the first time that we are warned that
the narrator and the narratee relate to each other not just like Matrëška dolls, but
rather as special invertible matrëškas ready to change places.
The Memoria tui motif as well as the three examples discussed above,
confirm that special tools are needed for singling out the input spaces that “are
not even explicitly mentioned,” as is the case in most blends (Schneider 2012,
8), but that have been intentionally masked. The narrator uses ambiguous infor-
mation intentionally to make the reader stop, shift attention and make a special
effort to comprehend the passage. To reconstruct the narrator’s self-reflexive and
elusive reasoning, the reader must turn to the self-reflexive mode of reading. The
Russian formalists laid the groundwork for studying the literary devices used by
authors to create poetic language. The Generative Narratology Framework goes
further, however, in that it helps us to observe the “how-to” of the process by
taking account of the cognitive mechanisms the reader employs when processing
ambiguous information.

I address you now for the following reason. I address you now on the following occasion. I
address you now simply to chat with you about Falter. What a fate! What a mystery! What
a handwriting!

[Сейчас обращаюсь к тебе вот по какому поводу. Сейчас обращаюсь к тебе вот по
какому случаю. Сейчас обращаюсь к тебе только затем, чтобы поговорить с тобой о
Фальтере. Вот судьба! Вот тайна! Вот почерк!]

The triple repetition evokes the story’s opening, with its repeated Pomniš … (Do
you remember…).
Triple repetitions are typical of fairy tales and other folklore genres. What
they mark is a shift of linguistic register to turn ordinary speech into the magical
incantation of a sorcerer’s spell. A cemetery is a perfect place for talking to the
dead. To do so, you just need to switch registers. Otherwise you will be talking to
the tombstone instead of communicating right through it!

When I tire of trying to persuade myself that he is a halfwit or a kvak (as you use to Russian-
ize the English synonym for ‘charlatan’).

[Вот судьба! Вот тайна! Вот почерк! Когда мне надоедает уверять себя, что он
полоумный или квак (как на английский лад ты звала шарлатанов)…]

‘Kvak’ stands for ‘Quack’ in the translation, adding yet another language into the
original Russian text.
88   Valery Timofeev

He strikes me as a person who… who, because he survived the bomb of truth that exploded
in him… became a god! Beside him, how paltry seem all the bygone clairvoyants: the dust
raised by the herd at sunset, the dream within a dream (when you dream you have awak-
ened), the crack students in this our institute of learning hermetically closed to outsiders;
for Falter stands outside our world, in the true reality. Reality!—that is the pouter-pigeon
throat of the snake that fascinates me.

[я вижу в нем человека, который… который… потому, что его не убила бомба
истины, разорвавшаяся в нем… вышел в боги!—и как же ничтожны перед ним все
прозорливцы прошлого: пыль, оставляемая стадом не вечерней заре, сон во сне
(когда снится, что проснулся), первые ученики в нашем герметически закрытом
учебном заведении: он-то вне нас, в яви,—вот раздутое голубиное горло змеи,
чарующей меня.]

The Russian passage is geometrically peculiar. If we regard the syntax of the above
passage, we will see a complex correlation which can be described as follows.
There is a certain picture involved here which denotes a kind of closed space.
Spatially, there are two directions, two vectors, one of which is centripetal, the
other centrifugal. The whole text is perfectly consistent—one may even say bor-
ingly so—in building up the same picture: Я вижу в нем (“He strikes me as”
or, more precisely: “I see in him”) is internal movement; человека, который…
который… потому, что его не убила бомба истины разорвавшаяся в нем…
(a person who… who, because he survived the bomb of truth that exploded in
him…). The bomb explodes externally. Вышел в боги (“became a god” or, liter-
ally: “went out into a god”) is directed, both inside and outside, with both vectors
involved (went out, but into a god).
The passage might be described as a set of invertible Matrëškas playing
strange games when the smallest one turns inside out to envelop all the others.
И как же ничтожны перед ним все прозорливцы прошлого (Beside him, how
paltry seem all the bygone clairvoyants). The geometry is quite complex, since,
etymologically, prozorlivtsy (clairvoyants) are supposed to look into the future,
which is not at all obvious in this sentence. Once again, the reader has to over-
come informational ambiguity. Russian prozorlivtsy proshlogo (literally “clairvoy-
ants of the past”) can have two meanings. The clairvoyants are either “bygones,”
so that they are looking from the past, or they see the past instead of the future.
The next phrase is no less peculiar, if not hilarious. Пыль, оставляемая
стадом на вечерней заре means “the dust raised by the herd at sunset,” or rather
“the dust left on the sunset,” as it literally says in Russian. Sunset is an indication
of time; but to “leave the dust on the sunset,” i. e., to leave it on time, one needs
to see time spatially. That suggests an allusion either to Mikhail Bakhtin (Mixail
Baxtin) or to Albert Einstein: “dust on the sunset” is an example of chronotope
or spacetime. Baxtin’s essay “Forms of Time and of the Chronotope in the Novel”
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    89

was first published in Russian in 1937, and Nabokov might well have been aware
of Einstein’s ideas of spacetime or of Ukhtomsky’s (Uxtomsky) interpretation of
the notion.

the crack students in this our institute of learning hermetically closed to outsiders

[первые ученики в нашем герметически закрытом учебном заведении]

This idea of hermetically closed space is the author’s neologism, since the dic-
tionary meaning of “hermetically closed” does not coincide with the usage of
the phrase applied to a type of educational institution. Once again, two chan-
nels are involved, channeling the same thing from bottom to top rather than vice
versa. Under natural circumstances, information travels down the channels from
Input Space 1 and Input Space 2 to meet at the Blended Space as a basic blending
network, as described by Schneider (2012, 4). In Nabokov’s text, meaning is gen-
erated in the opposite way: the reader has to unfold the blended meaning by fol-
lowing the information channels upwards to scrutinize the initial ideas that were
used for Cross-Space Mapping. In other words, the reader is to track the author’s
way of thinking step-by-step while at the same time taking the same route back-
wards to understand how such notions as hermetically closed schools come into

that is the pouter-pigeon throat of the snake that fascinates me

[вот раздутое голубиное горло змеи, чарующей меня]

This passage contains a literalized metaphor. In Antoine de Saint-Exupéry’s Le

petit prince (1943) is a famous image of a snake that has swallowed an elephant.
But already in Nabokov’s text, “the pouter-pigeon throat of the snake” stands for
a snake that has eaten a pigeon.
Up to this point, our locus has been limited to a cemetery as the place for
the narrator to address his deceased wife. Now, what kind of snake would you
expect to encounter at a cemetery? And what sort of pigeon (in Russian, golub’
is both pigeon and dove)? The Holy Spirit and the Serpent of the Judeo-Christian
tradition naturally come to mind; the Serpent has swallowed the Holy Spirit. The
Serpent and Eve motif seems no less appropriate here. Eve can easily stand for
the narrator’s wife. The Serpent fascinated (or, in Nabokov’s Russian original,
enchanted) Eve. Why “fascinates me?” To understand the passage, we should
go back to Memoria tui motif described as invertible matrëškas, a head-within-
a-head-within-a-head, i. e., the narrator and the narratee capable of changing
places. The serpent is fascinating, that is, it tempts the narrator with the enigma
90   Valery Timofeev

of the forbidden fruit from the tree of knowledge of good and evil. Thus the
central theme of the story, as it is commonly accepted by the critics (cf. Johnson
1985; Dolinin 1997), is introduced in the story.

Remember the we lunched at the hotel managed by Falter near the luxuriant, many-terraced
Italian border, where the asphalt is infinitely exalted by the wisteria, and the air smells of
rubber and paradise?

[Помнишь, мы как-то завтракали (Remember the time we lunched) в ему

принадлежавшей гостинице, на роскошной, многоярусной границе Италии, где
асфальт без конца умножается на глицинии, и воздух пахнет резиной и раем?]

We have entered paradise at the luxuriant many-terraced Italian border. Most obviously,
this alludes to Dante. And the many-terraced border is something that must smell, if not just
of paradise, at least its combination with rubber.

Adam Falter was still one of us then, an, if nothing about him presaged—what shall I call it?

[Адам Фальтер тогда был еще наш, и если ничто в нем не предвещало—как это

With a textual and spatial reality constituted in this way, as pointed out earlier, several
dragon heads are involved. One of them says, “What shall I call it?” [kak ėto skazat’?—liter-
ally: “How to say it?”], and another replies, skažu (I’ll say it):

[…] say, seerhood, nevertheless his whole strong cast (the caromlike coordination of his
bodily movements, as though he had ball bearings for cartilages, his precision, his aquiline
aloofness), now, in retrospect, explains why he survived the shock: the original figure was
large enough to withstand the subtraction.

[…скажу: прозрения,—зато весь его сильный склад (не хрящи, а подшипники,

карамбольная связность телодвижений, точность, орлиный холод) теперь, задним
числом, объясняет то, что он выжил: было из чего вычитать.]

This passage allows us to see perfectly well how the author’s thought is being
born. It is not just a chain of associations, but a chain moving in reverse. In sil’nyj
sklad, the reflexive mode of thinking is at work as if it had a whole thesaurus
at hand. A number of cognates of sklad, if not all, are activated at once, as well
as its different meanings. Sklad is a storing place, hence ne xrjašči, a podšipniki
(not cartilages but ball bearings), which are best stored in a warehouse. But “not
cartilages” means it is not only sklad as a warehouse, but also as a body-build.
“Ball bearings for cartilages” is all about bodily movements, derived from the
male physical build. The “caromlike coordination” comes from Carom, a type of
billiard game whose peculiarity is that the table is pocketless so that the balls go
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    91

from one side of the billiard board to the other and points are scored in one shot.
This makes Carom a very complicated game in terms of body movements as well
as in sight and thought. Therefore, we read točnost’, orlinyj xolod, where “pre-
cision” is meant literally as very precise movements of the body and “aquiline
aloofness” (literally “cold”) is a contamination of two phenomena: the eagle’s
keen eyesight and a cool head. Teper’, zadnim čislom means “now, in retrospect,”
or literally: “now, by back number.” The word čislo (number) actualizes still
another meaning of sklad as složenie (mathematical addition), adding a pun of
withstanding the subtraction: “now, in retrospect, explains why he survived the
shock: the original figure was large enough to withstand the subtraction.”

Oh, my love, how your presence smiles from that fabled bay

[О, моя милая, как улыбнулось тобой с того лукоморья]

It is now worth noticing that at first our addressee was headless, which was
hinted by her lack of headdress; now she gains a smile. There is no head, but
there is a smile! This is also how two cats appear in the text. One is from England:
Lewis Carroll’s Cheshire Cat, with the smile appearing independently of the rest
of the body; the other is from Alexander Pushkin. Nabokov’s Russian original
“fabled bay” is the Lukomor’e in the opening lines of Pushkin’s 1820 poem Ruslan
and Ljudmila:

There’s a green oak-tree by the shores

Of the blue bay; on a gold chain,
The cat, learned in the fable stories,
Walks round the tree in ceaseless strain:
Moves to the right—a song it groans,
Moves to the left—it tells a tale.

Thus, Pushkin’s cat, the narrator telling tales, is in Lukomor’e. By using a passive
structure (non-standard Russian toboj ulybnulos’: “it smiled by you” instead of
“you smiled”), the narrator declares that the addressee, who coincides with the
narratee, turns into the narrator (a smiling storytelling cat). “You” are the narrat-
ing cat. The narrator and the narratee are ready to change places again. First, the
Memoria tui motif, then ‘the pouter-pigeon throat of the snake” episode. All three
episodes contain a similarly working “reversing device”: a complex case of using
two modes of thinking simultaneously, one self-reflexive, the other introspective.

—and nevermore!—oh, I bite my knuckles so as not to start shaking with sobs, but there is
no holding them back; down I slide with locked brakes, making ‘hoo’ and ‘boohoo’ sounds,
and it is all such humiliating physical nonsense: the hot blinking, the feeling of suffocation,
92   Valery Timofeev

the dirty handkerchief, the convulsive yawning alternating with the tears—I just can’t, can’t
live without you. I blow my nose, swallow, and then all over again try to persuade the chair
which I clutch, the desk which I pound, that I can’t boohoo without you.

[—и никогда больше, и кусаю себе руки, чтобы не затрястись, и вот не могу, съезжаю,
плачу на тормозах, на б и на у, и все это такая унизительная физическая чушь:
горячее мигание, чувство удушья, грязный платок, судорожная, вперемежку со
слезами, зевота,—ах не могу без тебя… и высморкавшись, переглотнув, вот опять
начинаю доказывать стулу, хватая его, столу, стуча по нему, что без тебя не бобу.]

The passage is structured by a sequence of interchanging modes of perception. It

starts with a visual image of the narrator glancing at himself from aside. An audio
version of the same thing follows with a “boohoo” resulting from nasal conges-
tion, whereupon mogu (can) in “I can’t [do] without you” sounds like boboo (in
Russian). Thus, tactile perception is present, as well. Everything is compacted
into one episode, even though some thirty or forty years later psychologists would
try to convince us that every person has a dominant mode of perception. This may
be the case with people, but it is not with narrators in literary fiction. This type
of method in Nabokov has been described in detail by Marina Grishakova (2002).
“To persuade the chair which I clutch, the desk which I pound” is replete with
reminiscences of Russian literature, ranging from Chekhov’s (Čexov) character
in The Cherry Orchard addressing a piece of furniture as “My dear and honoured
case!” to Nikolaj Gogol’s The Inspector General where we read “Alexander of
Macedon was a hero, it is true. But that’s no reason for breaking chairs.” The last
phrase is well known in Russia and often used to ironically describe people who
tend to be too self-involved and taken away by their own stories to react politely
to others. The English version contains an allusion to “The Raven” by Edgar Allan
Poe, which is not so evident in the Russian original. The allusion to Poe might
well be a marker of Nabokov’s ironic self-reference to the American icon he was
to later use so extensively in Lolita.

That’s from a banal questionnaire, which ghosts do not answer…

[Слышишь ли меня? (Are you able to hear me?) Банальная анкета, на которую не
откликаются духи…]

Here, a sequence is at work which Bakhtin would characterize as dialogism. In

my terms, however, this is not an instance of dialogism but of the reflexive mode
of thinking. One asks, and the other responds. It is very peremptory: no sooner
has a smile, not even a head, appeared, than it is endowed with no less than the
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    93

authority of narration, whereas a second voice says that ghosts do not answer, so
that there is no point in asking silly questions.

but how willingly our death-cell mates respond for them: ”I know!” (pointing skyward at
random), “I’ll be glad to tell you!”

[но как охотно за них отвечают односмертники наши; я знаю! (пальцем в небо) вот
позвольте я вам скажу…]

The picture drawn with odnosmertniki (deathmates) is quite peculiar. The mor-
phological model according to which this neologism is formed is frequent: class-
mates, schoolmates, flatmates, cellmates, etc. A group of individuals having a
common attribute is identified, distinguishing that group from all others. Class-
mates are those who were in the same class, where others do not belong. In Nabok-
ov’s text, on the contrary, the attribute is non-distinctive: death is something that
envelops all, with no chance for anyone to escape. Here, blending is reversed
once again. The contradiction contained in the word “deathmates” results not
from the word itself, nor is contextual. To grasp the bitter irony of the word, one
has to go back to its origin, the way it was produced step-by-step.

the hollow of your temple, the forget-me-not gray of an eye squinting at an incipient kiss,
the placid expression of your ears when you would lift up your hair

[Милая твоя голова (Your darling head), ручеек виска, незабудочная серость косящего
на поцелуй глаза, тихое выражение ушей, когда поднимала волосы]

Here comes the head. Ručeëk viska (literally, “the streamlet of your temple”) sug-
gests staring closely at something from an extremely short distance. “The forget-
me-not gray” is meant to be misleading, since forget-me-nots are not gray at all.
One might wonder why the color of the flower is changed. The author, of course,
is free to choose any attribute to match any color. Naïve realism is being playfully
attacked in this passage. What was the color of the real eyes? Were there any real
eyes at all? Whose eyes? As if authors should always go by some reality… And of
course, each character must have a prototype for eye color, inherited rather than
invented. Such naïve realism was often played with during the twentieth century,
rebuking nineteenth-century realism, when the aim of literature was to portray
reality. So, how might “the forget-me-not grey of an eye squinting at an incipient
kiss” be understood? Let us count the kissing heads. There is a kiss, and an eye
squinting at it, and another one looking at the temple… So there are at least three
people involved. We confront a monstrous three-headed dragon: two heads are
kissing, the third is watching. Alternatively, the whole process may be a mere
figment of the imagination, as artists are free to imagine whatever they think
94   Valery Timofeev

appropriate. “The placid expression of your ears when you would lift up your
hair”: why “placid,” or even “quiet”? Here, translation from the English is literal,
which is hard to trace in the backwards translation of an “I’m all ears” kind of
idiom. What is meant here is that the heroine is very attentive. This is not the first
occurrence of English in Nabokov’s Russian text: she spoke some English, and
the cat was partly English (from Cheshire), to say nothing of “nevermore” stand-
ing next to Pushkin and Carroll to refer to Edgar Allan Poe, whose cats (e. g., “The
Black Cat”) are quite famous, too. So why should English be left out if Latin has
been brought in?

how can I reconcile myself to your disappearance, to this gaping hole [in life] into which
slide everything—my whole life, wet gravel, objects and habits—and what tombal railings
can prevent me from tumbling, with silent relish, into this abyss? Vertigo of the soul.

[как мне примириться с исчезновением, с этой дырой в жизни, куда все теперь
осыпается, скользит, вся моя жизнь, мокрый гравий, предметы, привычки… и какая
могильная ограда может помешать мне тихо и сытно повалиться в эту пропасть.

Once again, we are confronted with a case of complex geometry. Imagine that
life can have a hole in it, like a bagel, and watch what happens next. Everything,
including life itself, slides into that bagel’s hole. This means the bagel has col-
lapsed into its own hole! That is another example of invertible Matrëškas.
“Tombal railings” helps us visualize a cemetery. And yet I am inclined to
think that the locus is highly imaginary and that the imaginary nature of the cem-
etery is emphasized. The reason is that a tombal railing is a distinctive feature
of Russian Orthodox graveyards, while this would be quite unusual in Italy or
France, where the story takes place. Dušekruženie (vertigo of the soul) is yet
another of the author’s numerous coinages, confirming that we are in this multi-
headed space of self-reflection. Moreover, the reader has to decide how to over-
come the obvious ambiguity at the end of passage. “Vertigo of the soul” might be
read as the answer to the “What tombal railings can prevent me from tumbling
[…] into this abyss?” Or it might be understood as a self-reflexive comment on the
passage we have read. It may also introduce the passage that follows as a shifting
device to keep the reader trapped in the eternal “Do you remember?” repetition.

Remember how, right after you died, I hurried out of the sanatorium, not walking but sort
of stamping and even dancing with pain (life having got jammed in the door like a finger)…

[Помнишь, как тотчас после твоей смерти я выбежал из санатория и не шел, а как-то
притоптывал и даже пританцовывал (прищемив не палец, а жизнь)…]
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    95

Jamming not a finger but a life is another bagel. Jamming of life has occurred,
but the picture goes on to show how a jammed life is an inverted literalized met-
aphor—like a finger (or a toe, as in Russian the same word, palec, means both).
For this reason, I went “not walking but dancing,” since my toe (life) hurts. This
is a reverse move, a metaphor broken down into its literal meaning, working
instead of or even as a metaphor. Normally we are not supposed to see the literal
meaning of metaphors, just the figurative ones. Nabokov often does this the other
way round. As it turns out, both meanings are at work in this passage: jamming
of life and jamming a toe.

alone on that winding road among the exaggeratedly scaly pines and the prickly shields of
agaves, in a green armored world that quietly drew in its feet so as not to catch my disease.

[…один на той витой дороге между чрезвычайно чешуйчатых сосен и колючих

щитов агав, в зеленом забронированном мире, тихонько подтягивавшем ноги,
чтобы от меня не заразиться.]

We have to diagnose a hypertrophy of the subject using a term taken from literary
theory—metaphor—so as to avoid adopting psychiatric language. The narrator
has such a peculiar relationship with the world that the world is afraid of catch-
ing his life jamming. In other words, the world can also see that life is “jammed”
and draws its feet back just in case, so as not to go limp, and it is awfully scared.
“[E]xaggeratedly scaly pines and the prickly shields of agaves” expresses the idea
of armor: the world is trying to defend itself.

Ah, yes—everything around me kept warily, attentively silent, and only when I looked at
something did that something give a start and begin ostentatiously to move, rustle, or buzz,
pretending not to notice me

[О да, все кругом опасливо и внимательно молчало, и только когда я смотрел на

что-нибудь, это что-нибудь, спохватившись, принималось деланно двигаться или
шелестеть или жужжать, словно не замечая меня]

This is how children behave in a nursery school. When the teacher enters, they
pretend to be asleep. In Nabokov’s text, this is reversed. When the narrator enters,
Nature pretends not to notice him. What a wonderful world!

“Indifferent nature,” says Pushkin. Nonsense! A continuous shying away would be a more
accurate description.

[«Равнодушная природа»—какой вздор! Сплошное чурание, вот это вернее.]

96   Valery Timofeev

The allusion to Pushkin is left for the reader to grasp in the original.

What a shame, though. You were such a darling.

[Жалко же. Такая была дорогая.]

The juxtaposition of žalko and dorogaja makes up a double beat of lexical units,
as in a dictionary. Žalko has two meanings: the emotion of pity and the condition
of greediness. Similarly, dorogaja can mean dear to me emotionally or expen-
sive, pricey. Following is another case of informational ambiguity for the reader
to overcome:

And, holding on to you from within by a little button, our child went with you. But, my poor
sir, one does not make a child to a woman when she has tuberculosis of the throat. Involun-
tary translation from French into Hadean.

[И держась снутри за тебя, за пуговку, наш ребенок за тобой последовал. Но, мой
бедный господин, не делают женщине брюха, когда у нее горловая чахотка.
Невольный перевод с французского на адский.]

Here, the whole idea is structured backwards.

You died in your sixth month and took the remaining twelve weeks with you, not paying off
your debt in full, as it were.

[Умерла ты на шестом своем месяце и унесла остальные, как бы не погасив полностью


This is where žalko and dorogaja, the words that have put both the emotion of
pity and the condition of greediness into play, are fully literalized. The guy was
robbed of three months.

How much I wanted her to bear me a child, the red-nosed widower informed the walls): the
red-nosed widower is the narrator, so he has switched over to third-person narration.

[А как мне хотелось, сообщил красноносый вдовец стенам, иметь от нее ребеночка.]

Êtes-vous tout à fait certain, docteur, que la science ne connaît pas de ces cas exceptionnels
où l’enfant naît dans la tombe?
“And the dream I had” that garlicky doctor [who was at the same time Falter] replying with
exceptional readiness that yes, of course it sometimes did happen, and that such children
[i. e., the posthumously born] were known as cadaverkins.
 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    97

[И сон, который я видел: будто этот чесночный доктор (он же не то Фальтер, не то

Александр Васильевич) необыкновенно охотно отвечал, что да, как же, это бывает,
и таких (то есть посмертно рожденных) зовут трупсиками.]

This last neologism, trupsik (cadaverkin), is coined in accordance with a pattern

that did not exist in the Russian language at the time of Nabokov’s writing, even
though it is typical for English. One word slips into another like a banknote goes
into a wallet. Such is the case with Nabokov’s trupsiki: the word trup (cadaver,
dead body) is full, and the pupsik part is just guessable. Pupsik is a very small
baby doll, so the word is used tenderly, similar to English “baby,” “sweetie,” etc.
The cadaverkin image uses the same geometrical pattern as a number of other
episodes we have already discussed.
The cadaverkin image, along with an ‘Indifferent nature’ citation, is part of
a lengthy allusion to Pushkin’s poem “Wondering the noisy streets” (“Brožu li
ja vdol’ ulits šumnyx”). If we take the poem as a whole, we will find all of its
themes in the opening episode of Nabokov’s “Ultima Thule”: death, temple, oak-
tree, infant, grave. In Nabokov’s text, everything is turned inside out, though,
and inverted. In Pushkin’s poem we read: “When I caress a dear child, I’m already
thinking: goodbye!”

3 Conclusion
Pushkin’s allusions, and the numerous invertible Matryoshkas and cases of infor-
mational ambiguity we have observed in the opening page of Nabokov’s story,
show how the basic structure and general mechanisms of blending theory might
be used to observe narratives in their making. When used with a reflexive mode
of thinking as an additional tool, it reveals the two or more input spaces several
times in a row. Once blended, the elements are then separated only to be blended
again. We need to go from blended space upward, checking the input spaces to
see where the elements at work belong and how both the ideas and their form
develop in interaction with each other. I call this method of analysis Generative
Narratology Framework and submit that it might be used to construct a hypothet-
ical version of the way the creative process is organized in terms of monitoring,
regulating and controlling narrative progression.4

4 I would like to thank my friends and colleagues Olga Voronia, Boris Averin and Ivan Dela-
zari for their help and encouragement. The paper was made possible by the support of a Saint
98   Valery Timofeev

Works Cited
Akturk, Ahmet Oguz, and Ismail Sahin. 2011. “Literature Review on Metacognition and its
Measurement.” Social and Behavioral Studies 15: 3731–3736.
Bakhtin, M. M. 1981 [1937]. “Forms of Time and of the Chronotope in the Novel.” In M. M. B.
The Dialogic Imagination: Four Essays, edited by Michael Holquist, translated by Caryl
Emerson and Michael Holquist, 41–83. Austin: University of Texas Press.
Bermúdez, José Luis. 1998. The Paradox of Self-Consciousness. Cambridge, MA: The MIT Press.
Dimaggio, Giancarlo et al. 2008. “Know Yourself and You Shall Know the Other… to a Certain
Extent: Multiple Paths of Influence of Self-Reflection on Mindreading.” Consciousness and
Cognition 17: 778–789.
Davydov, Sergey. 2004. Teksty-matryoshky Vladimira Nabokova [Matryoshkky Style Texts by
Vladimir Nabokov]. Saint Petersburg: Kirtsideli.
Dolinin, Aleksandr. 1997. “Zagadka nedopisannogo romana” [The Enigma of the Unfinished
Novel]. Zvezda 12: 215–224.
Fenigstein, Allan, Michael F. Scheier and Arnold H. Buss. 1975. “Public and private
self-consciousness: Assessment and theory.” Journal of Consulting and Clinical
Psychology 43 (4): 522–527.
Fletcher, Paul C. et al. 1995. “Other minds in brain: a functional imaging study of ‘theory of
mind’ in story comprehension.” Cognition 57: 109–128.
Grishakova, Marina. 2002. “Vizual’naia poetika V. Nabokova” [Nabokov’s Visual Poetics]. Novoe
Literaturnoe Obozrenie 54 (2): 205–228.
Herman, David. 2004. “From Narrative Narcissism to Distributed Intelligence: Reflexivity as
Cognitive Instrument in Joyce’s Finnegans Wake.” Frame 17 (2/3): 27–43.
Herman, David. 2006. “Genette meets Vygotsky: Narrative Embedding and Distributed
Intelligence.” Language and Literature 15 (4): 69–79.
Johnson, Donald Barton. 1985. Worlds in Regression: Some Novels of Vladimir Nabokov. Ann
Arbor: Ardis.
Kircher, Tilo, and Dirk Leube. 2003. “Self-consciousness, self-agency, and schizophrenia.”
Consciousness and Cognition 12: 656–669.
Lane, R. D., and G. E. Schwartz, 1987. “Levels of Emotional Awareness: A Cognitive
Developmental Theory and its Application to Psychopathology.” American Journal of
Psychiatry. 144: 133–143.
Musholt, Kristina. 2013. “Self-Consciousness and Nonconceptual Content.” Springer
Philosophical Studies 163: 649–672.
Nabokov, Vladimir. 1973 [1942]. “Ultima Thule.” Translated by Vladimir Nabokov and Dmitri
Nabokov. The New Yorker April 7: 38–54.
Nabokov, Vladimir. 2000 [1942]. “Ultima Thule.” In V. N. Sobranie sočinenii russkogo perioda
v piati tomax [Collected Russian Language Works in Five Volumes], edited by M. Kozikova.
113–139. Saint Petersburg: Simpozium.
Pushkin, Alexander. 2004a.[1820]. Ruslan and Ludmila. Translated by Yevgeny Bonver. http://

Petersburg State University Grant (Cognitive mechanisms overcoming information ambiguity

 Nabokov’s “Ultima Thule”: An Exercise in Generative Narratology    99

Pushkin, Alexander.1999 [1829]. “Wondering the noisy streets.” Translated by Babette Deutsch.
Savanah, Stephane. 2012. “The Concept Possession Hypothesis of Self-Consciousness.”
Consciousness and Cognition 21: 713–720.
Schneider, Ralf. 2012. “Blending and the Study of Narrative.” In Blending and the Study of
Narrative: Approaches and Applications, edited by R. S. and Marcus Hartner, 1–26. Berlin
and Boston: De Gruyter.
Timofeev, Valery. 1983. “O metodah realizacii filosofskoi koncepcii v khudozhestvennom
proizvedenii” [Methods of Manifestation of a Philosophical Concept in a Work of
Literature]. Tret’ja vsesojuznaja konferencija molodyh uchenyh-filologov, edited by G. M.
Čoxonelidze, 171–172. Tbilisi: Mincniereba.
Timofeev, Valery. 1999. “O jazykah avangarda” [Languages of the Avant-Garde]. Vestnik
fililogicheskogo fakul’teta Instituta Inostrannyh Jazykov v Sankt-Peterburge 2/3: 59–62.
Vogeley, K. et al. 2001. “Mind Reading: Neural Mechanisms of Mind and Self-perspective.”
Neuroimage 14: 170–181.
Wittgenstein, Ludwig. 1958. The Blue and Brown Books: Preliminary Studies for the
Philosophical Investigations. New York: Harper & Row.
Wolf, Werner. 2014 [2011]. “Illusion (Aesthetic).” In Handbook of Narratology, edited by Peter
Hühn, Jan Christoph Meister, John Pier and Wolf Schmid, vol. 1: 270–287. Berlin and
Boston: De Gruyter. Also available at: living handbook of narratology: http://www.lhn. (Revised 17 January 2014)
Natalya Bekhta (Giessen)
Emerging Narrative Situations:
A Definition of We-Narratives Proper
Recent years have seen a considerable rise in interest in fiction that challenges
existing narratological paradigms. Untypical strategies of narration give rise to
new narrative forms. Since the end of the twentieth century, there have been an
increasing number of second-person texts, narratives in the plural and other
experiments with narrative voice. At present, second-person texts have been the-
orized to a considerable degree, but narration in the plural still remains under-re-
searched. This is especially obvious from articles on we-narratives in which
extremely diverse texts are grouped under a single rubric. Defining we-narrative
is difficult for a number of reasons, but doing so is necessary, I claim, for the
phenomenon to be accessible to systematic investigation. In order to delimit
we-narrative proper, I propose to combine the existing structuralist and contex-
tual approaches. This definition requires a re-conceptualization of the narrator
category and thus of the narrator’s identity, knowledge, scope of focalization
and narrative levels. Furthermore, such a definition will contribute to current
attempts to revise classical theories of narrative.
On the one hand, the suggested definition of we-narrative proper relies on
the criteria used for identifying ‘narrative situations’ (cf. Stanzel 1984 [1979]) and
can be considered as a diversification of the classical typology of narrative forms.
On the other hand, when defined as a separate narrative situation, we-narratives
undermine the very paradigm used for their description and call for a re-concep-
tualization of the defining categories.
In Stanzel’s terms, narrative situation offers a cluster of concepts for describing
forms of narration based on mode (Who is narrating? Is there a narrator?), person
(What is the relationship between the narrator and the narrated?) and perspective
(How is ‘the fictional reality’ perceived? Whose perception mediates the access?).1

1 Here I follow Dorrit Cohn in my observations that Stanzel’s ‘nuclear typology’ of mode, person
and perspective anticipates many subsequent discussions such as the division of voice and vi-
sion or the fact that there is a close affinity between Stanzel’s and Genette’s principal categories
in spite of what is usually acknowledged (Cohn 1981, 159–160). Thus, in what follows I shall
use Genette’s hetero- and homodiegesis together with Stanzel’s ‘first person’ and ‘authorial’
narrators whenever the two systems are complementary. Otherwise, I give preference to Stan-
zel’s terminology for describing the narrator, as it does not automatically relate to categories of
narrative embedding (extra- and intradiegesis) which make Genette’s four-part model internally

DOI 10.1515/9783110555158-006
102   Natalya Bekhta

When combined, these categories of narrative transmission2 also serve as defining

features for we-narratives, although in a modified form. My reason for drawing on
the idea of narrative situation is that it is by treating these features in their totality
that one can establish the structurally significant and thus defining qualities of
narrative form. The analysis of structural features also requires that contextual
considerations be taken into account. In this way, it becomes possible to delimit
we-narrative proper, which I call the ‘performative we’, from other uses of we-dis-
courses in fiction, or the ‘indicative we’.3

1 An Emerging Narrative Situation: We-Narration

versus We-Narrative
When it comes to emerging modes of narration, pronominal experimentation is,
arguably, one of the major developments in narrative form since the second half
of the twentieth century. Let us consider, for example, alternations between first-
and third-person accounts in texts like Alberto Moravia’s novel Io e lui (1971) or
Christa Wolf’s Nachdenken über Christa T. (1968); use of two second-person pro-
nouns to designate protagonists in Julio Cortázar’s short story “Usted se tendió a
tu lado” (1985); a venture beyond first-person in Monique Wittig’s Le corps lesbien

contradictory (cf. Walsh 2010, 37–41). Furthermore, the hybrid nature of narrative situation can
be considered salutary in so far as it highlights a “disregard for the communicative model’s com-
mitment to a literal mode of transmission” (Walsh 2010, 44)—a destabilizing aspect of Genette’s
model that has led to problems with classifying plural narrators and otherwise untypical strat-
egies of narration.
2 The notion of narrative transmission (or mediacy in Stanzel’s use) calls for a caveat: it under-
lies the idea of narrative situation but at the same time cannot be extended to all narrative fic-
tion. It applies well to we-narratives, as such narratives have a manifest narrator with a distinct
voice and produce the effect of ‘re-telling’ of a story. However, ‘transmission’ is not a universal
quality of narratives. I elaborate on this point is Section 5 below.
3 It must be clarified from the outset that in this article my goal is to delimit a new narrative
form with the help of the notion of narrative situation and not to revise the Typological Circle
as such which has proved a daunting and unsuccessful task, pointing out many contradictions
inherent to this form of typology. Besides an attempt to update the Typological Circle with the
second-person narrative undertaken by Fludernik (see Reitan [2011] for a summary and critique),
I have in mind Richardson’s (1994, 77) attempt to place we-narrative on the Circle which, while
schematizing the relationship between ‘we’ and other narrative modes metaphorically, is not
compatible with the classificatory principles of the Circle and cannot be regarded as a revision.
For a critique of the term narrative situation itself (especially its hybrid aspect), see Wolf Schmid
(2010 [2005], 90–91).
 Emerging Narrative Situations: A Definition of We-Narratives Proper    103

(1973). More recently, narrative theorists have drawn attention to an abundance

of second-person narratives such as Lorrie Moore’s short stories (1985), Jay McIn-
erney’s Bright Lights, Big City (1984) or Édouard Levé’s Suicide (2008).4 Clearly,
the Genettian choice between two grammatical forms (first or third person) no
longer covers the gamut of the contemporary novel. The time has thus come to
act on one of the possibilities for revision of Stanzel’s typical narrative situations:
the three most frequently developed types “in the history of the novel”5 which,
he noted, “can be revised at any time, should the future development of the novel
demand it” (Stanzel 1984 [1979], 61). Two new narrative situations that emerge
in contemporary fiction are both primarily modifications of the distinction of
person: second-person narrative and first-person plural narrative.
While there have already been several publications on the topic, namely
by Susan Lanser (1992), Uri Margolin (1996, 2000), Brian Richardson (2006),
Amit Marcus (2008b), Monika Fludernik (2011), to name the key ones, the term
“we-narrative” in narratology is currently highly unconsolidated, and definitions
of this phenomenon are sketchy at best. This becomes obvious when one studies
a bibliography of we-narratives compiled by Fludernik (2011). Apart from her own
examples, Fludernik includes texts analyzed in other articles on the topic. The
bibliography contains a mixture of examples such as William Faulkner’s short
stories “A Rose for Emily” (1995 [1930]) and “This Will Be Fine” (1995 [1935]), the
former being a paradigmatic example of we-narrative proper and the latter, of a
first-person narrative (sensu Stanzel), or novels such as The Virgin Suicides (1993)
by Jeffrey Eugenides and Michael Dorris’s A Yellow Raft in Blue Water (1987).
Again, The Virgin Suicides is a good example of we-narrative, whereas A Yellow
Raft in Blue Water is what Richardson might call a multiperson novel, composed
of three sequential accounts by first-person narrators that contribute to a com-
munal story.
In this article, I suggest that we-narrative should be viewed in its own right,
distinct from first-person singular. Yet it should also be distinguished from other

4 For further examples, see an extensive overview in Richardson (1994, 2006) and in Fludernik
(1993). See Rolf Reitan’s (2011) insightful article for possibly the only attempt so far to bring to-
gether “to a real confrontation with one another” (148) the three perspectives on second-person
narratives: Brian Richardson’s, Irene Kacandes’ and Monika Fludernik’s. The impetus behind
Reitan’s article is similar to my present attempt to set properly we-narratives off from other uses
of ‘we’ in fiction.
5 The three typical situations are first-person narrative (e. g., David Copperfield and Huckleberry
Finn), authorial narrative (e. g., Vanity Fair and Sons and Lovers) and figural narrative (e. g., A
Portrait of the Artist as a Young Man and Mrs Dalloway) (cf. Stanzel 1971 [1955]).
104   Natalya Bekhta

uses of we-discourses in fiction.6 In fiction ‘we’ is regularly found in non-narrative

literary genres: lyrical poetry, hymns and songs (e. g., of solidarity and protest) or
Greek tragedy (the chorus’ speeches).7 Non-fictional narratives that employ the
‘we’ form represent a significant part of everyday storytelling and are, in most
cases, reports of past activities of a collective agency (e. g., stories from a group
of friends who have experienced an adventure or a couple that has just returned
from holiday). So one may speak of the relatively few literary narratives that are
written in the ‘we’ form. Instances of we-narration consist predominantly of short
stories and individual chapters or passages of plural narration in otherwise multi-
person novels.8 Entire novels that use we-narration throughout are relatively rare.
Nevertheless, it has been noted that this form has started to gain in popularity in
fiction as well as in theoretical investigations (Richardson 2009, 150–151).
One of the broadest uses of the phrase “‘we’ fictional narrative” is proposed
by Marcus, who subsumes under this heading texts with “a single narrator speak-
ing on behalf of others (e. g., Franz Kafka’s ‘Josephine the Singer, or the Mouse
Folk’)” and those “co-narrated by two or more narrators (e. g., Ágota Kristóf’s
The Notebook)” as well as “narratives in which there are thematically significant
shifts from ‘we’ to other pronouns and vice versa” (Marcus 2008a, 135). While
acknowledging that there is a significant structural difference between these
cases, Marcus nonetheless avoids “a more rigorous definition,” because of “the
scarcity of such narratives” (2008a, 135) and because “some of the most note-
worthy ways of employing the first-person plural are best illustrated in texts that
alternate between ‘we’ and other forms of narration” (Marcus 2008b, 2). In light of
this terminological confusion, I would like to propose precisely the kind of more
rigorous definition that Marcus strives to avoid in the desire to cover as many
examples of narration in the plural as possible. This definition will be based on
maintaining the structural distinction between ‘narration’ and ‘narrative’ and on
the contextual difference between indicative and performative uses of ‘we’-nar-

6 Instances of ‘we’-narration often occur in otherwise first-person narratives and cannot func-
tion as a qualifier of we-narrative proper. Much in the same spirit, Reitan (2011, 152) analyses the
existing approaches to the definition of second-person narratives and concludes that some of
its cases, namely the occurrences of ‘you’ in otherwise third-person narratives (i. e., authorial or
figural narrative situations) does not create a second-person proper.
7 Cf. Margolin (1996, 116).
8 See bibliographies in Richardson (2006, 141; 2009, 158–159) and Fludernik (2011, 136–141).
 Emerging Narrative Situations: A Definition of We-Narratives Proper    105

2 Creation of a Plural Narrative Voice and its

Linguistic Peculiarities
The first basic distinction that becomes obscured in Marcus’s work is that between
‘narration’ and ‘narrative’.9 Narration in the plural is a narrative technique used
in various narrative situations: in first-person narratives (mostly indicatively), or
in authorial narratives (indicatively and in the form of pluralis auctoris), or in
what I propose to call we-narratives proper (mostly performatively). We-narration
thus should not be equated with a full-blown we-narrative. In other words, most
of the examples that Marcus covers as noteworthy instances of the first-person
plural are passages of we-narration. These passages offer a lot of interesting mate-
rial for analysis, but they do so within the dynamic of those narrative situations
in which they are used and thus work differently from we-narration in a fully
realized we-narrative.
Second, I propose that we distinguish between two uses of the first-person
plural pronoun: indicative and subjunctive, which might also be compared to con-
stative and performative.10 In what follows, I shall use the pair ‘indicative’/‘per-
formative’, as these terms have a wider currency beyond theoretical grammar.
Strictly speaking, these grammatical distinctions, adopted from the category of
the verb, do not describe the moods in which the we-pronoun is used, since pro-
nouns cannot acquire moods. They can, however, be metaphorically applied to
describe how the pronoun is used in a particular act and what its functions and
effects are.
For ‘we’, the prototypical function, as Johannes Helmbrecht’s suggests, is to
establish, around the ‘I’ of the speaker, groups of “various sets of addressees and
non-speech act participants” (Helmbrecht 2002, 44). This is what I would call a
straightforward indicative use—that is, an indication or statement that something
is the case. If ‘we’ does not do this, then one can single out two non-prototypical
uses of ‘we’: (a) ‘we’ is used to express an emotional or a social connection of the
speaker with a group but does not refer to the speaker herself, as she does not
belong to this group properly speaking (e. g., a football fan saying ‘we won the

9 Marcus seems to be using the two categories synonymously, for example: “Like first-person
singular narration, ‘we’ narration is based on personal experience […]. ‘We’ narratives lack the
objectivity, reliability, and veracity conventionally attributed to third-person narration” (2008b,
1–2; emphasis added).
10 I am indebted to Daniel Hartley for this observation. My use of verbal categories in this case
also alludes to Genette’s (1980 [1972], 161–162) approach to ‘narrative moods’ and his justification
of the applicability of the category of mood to narrative discourse.
106   Natalya Bekhta

game’); and (b) ‘we’ is used for singular reference, either referring solely to the
speaker or to the addressee but not to any group. It includes: pluralis majestatis,
or the royal ‘we’ (‘We are not amused’, said Queen Victoria) which, similarly to
the nursery ‘we’ (‘And how are we doing today, sweetie?’), invokes “institutional
power or superiority of the speaker” (Helmbrecht 2002, 35); pluralis modestatis,
which is the muting of “I”;11 pluralis auctoris, or authorial ‘we’, which is some-
times inclusive (I+you) and draws in the speaker/writer and the listener/reader,
often in order to invoke empathy (‘And so, dear reader, we shall come back to our
hero’). I would like to suggest that a fictional narrating ‘we’, in its performative
function, goes further than the non-prototypical ‘we’ and completely erases the
need for the ‘speaker’ in the singular, being a reference of a group to itself—a ref-
erence which, at the same time, calls this group into being.
So by the indicative use of ‘we’, I mean an individual speaker’s straightfor-
ward reference to herself and another person or group to which she belongs or
with which she associates herself situationally (i. e., at the moment of speaking)
or more generally. Typical examples of such usages of ‘we’ (e. g., ‘We need to
buy more milk’) come from non-fictional non-narrative discourses: many kinds
of speeches, reports, declarations, petitions and prayers—all designated to rep-
resent a collective agency of some sort, be it a political party or a community of
academics working on a common project. This use will rely on the prototypical
function of ‘we’. Crucially, the use of such indicative ‘we’ alone is not a sufficient
condition for the creation of a plural, collective voice and a collective narration
which is one of the defining features of we-narrative. For example, Faulkner’s
short story “That will be fine” (1935) opens as follows:

(1) We could hear the water running into the tub. We looked at the presents scattered over
the bed where mamma had wrapped them in the colored paper, with our names on them so
Grandpa could tell who they belonged to easy when he would take them off the tree.
“This one is yours,” I said.
“Sho now,” Rosie said. “You come and get in that tub like your mamma tell you.” (Faulkner
1995 [1976], 265)

The direct speech that comes immediately after the opening paragraph identifies
the I-speaker, who is a child, and his nanny Rosie, making it clear that the ‘we’
in the opening paragraph is an indicative situational reference. The story itself is
a typical first-person narrative where, unlike in we-narrative proper, the narra-

11 When one pronominal formula is used instead of another (‘we’ instead of ‘I’), the effect is
that “the sender puts himself in the background and gives a general opinion without referring
explicitly to himself” (Dieltjens and Heynderickx 2007, 237).
 Emerging Narrative Situations: A Definition of We-Narratives Proper    107

tor-character is or has been physically present during the situations he is speak-

ing about and is subject to the limitations of the first-person singular at all times,
even when we-reference is used.
The performative ‘we’, on the other hand, creates a more complex reference:
it expresses something that is imagined or wished, creating something that did
not exist before. The performative ‘we’ erases the single speaker who utters it and
constructs a plural storytelling voice and a plural narrator, communal or collec-
tive. In other words, the speech act of the we-narrator with regard to itself is per-
formative in that it creates this very narrator.12 As I shall discuss below, such per-
formative ‘we’-references create plural narrators in the fullest sense. They have
two salient features: first, they never imply the ‘I’ or even hint at the possibility
of a singular individual speaking for the group; second, they often contribute to
a narrative mode that James Phelan has termed “lyric progression” (Phelan 2005,
10). Importantly, however, these two features—as well as we-narrative proper as
such—do not occur in one line or even in a few sentences. They become detectable
over time, through repetition, and thus it is difficult to demonstrate this phenom-
enon in one quotation: it is the “[r]epeated use of the pronoun [that] reinforces the
idea of solidarity, as each verb, each sentence marked by the first person plural
suggests common thought and common action” (Fulton 2003, 1106). An example
of we-narrative proper with a performative ‘we’ that creates such a communal
narrator is Joshua Ferris’s novel Then We Came to the End (2007). Right from the
beginning, a situation of narration is set up which recurs throughout the novel.

(2) We were fractious and overpaid. Our mornings lacked promise. At least those of us who
smoked had something to look forward to at ten-fifteen. Most of us liked most everyone,
a few of us hated specific individuals, one or two people loved everyone and everything.
Those who loved everyone were unanimously reviled. We loved free bagels in the morning.
They happened all too infrequently. Our benefits were astonishing in comprehensiveness
and quality of care. Sometimes we questioned whether they were worth it. We thought
moving to India might be better, or going back to nursing school. (Ferris 2007, 3)

This passage introduces the novel’s plural narrator-protagonist: a community of

office co-workers. Rather than voicing a fully homogeneous group, ‘we’ is used
to move between general statements that are true for all group members such as
“We were fractious and overpaid” and particularizations true only for parts of
the group: “most of us,” “a few of us,” “one or two people”—the latter being the
most group-detached reference. Moreover, the communal narrator acquires an

12 Here I directly contradict Margolin’s (2001, 244) observation that speech acts of such narra-
tors regarding ‘we’ are primarly constative.
108   Natalya Bekhta

individualized voice by expressing thoughts or dreams in what can be termed a

free indirect we-mode, e. g., “Sometimes we questioned,” “We thought moving
to India might be better.” No ‘I’-reference is conceivable—nor, in fact, needed.
And the novel never provides one. Unlike Faulkner’s ‘we’, the office ‘we’ exhibits
self-awareness: this community is capable of commenting on ‘ourselves’ being
a community. Unlike Faulkner’s ‘we’, the office ‘we’ functions as a fully auton-
omous character, not only with collective actions but with thoughts, emotions,
reflections on ‘our’ narration and comments to the narratee: “Is this boring you
yet? It bored us every day. Our boredom was ongoing, a collective boredom, and it
would never die because we would never die” (Ferris 2007, 4). Such we-reference
creates a communal character-narrator, capable of speaking as one, and estab-
lishes a peculiar narrative situation in which the authority of the narrative voice
is at once personal and authorial, subjectively individual and generally true, all
at the same time.
The distinction between the uses of we-discourse is thus not grammatical but
contextual. The opening sentences of Faulkner’s story and of Ferris’s novel can
both be read as indicative. Yet, as the text progresses, the latter acquires a per-
formative mode. So the first example (1) is a typical first-person narrative whereas
the second (2) is what I propose to call we-narrative proper.
As I have mentioned above, in fiction ‘we’ figures most prominently in
non-narrative genres. In non-fictional, everyday usage of ‘we’, this pronominal
form is familiar. In narrative fiction, however, ‘we’ can break its automatized per-
ception: as a collective voice starts to take form, we-narration stands out as some-
thing distinct from its non-fictional applications.13 However, both in its indicative
and performative roles, the we-reference remains “the most complex category of
all person categories” (Helmbrecht 2002, 33), as has frequently been observed in
linguistic investigations.
In fiction and non-fiction alike, the major linguistic peculiarity of the
first-person plural pronoun is its ambiguity of reference. Personal pronouns are
semantically empty: they cannot have a meaning until assigned to a context. In
other words, the reference ‘I’ is not clear until somebody utters it in a particu-
lar situation. With ‘we’, however, the situation is more complicated. Who utters
‘we’? From the viewpoint of enunciative linguistics, ‘we’ cannot have a voice, as
a voice is an exclusive privilege of the first-person singular pronoun only. Often,
it is not possible to identify all of the singularities that compose a ‘we’ even if the

13 Cf. Richardson (2006, 37) on a comparison between the “‘unnatural’ from the outset” sec-
ond-person voice and a less odd we-narration, and Marcus (2008b, 2–3) on the semantic instabil-
ity of the we-pronoun concealed by its everyday use.
 Emerging Narrative Situations: A Definition of We-Narratives Proper    109

context is given. The first-person plural pronoun “annexes an indistinct mass of

other persons to ‘I’” (Benveniste 1971 [1966], 203), thus producing a vast variety
of possible combinations. In enunciation theory it is a given that the ‘I’ is one and
subjective and does not allow for pluralization:

If there cannot be several “I”s conceived of by an actual “I” who is speaking, it is because
“we” is not a multiplication of identical objects but a junction between “I” and the “non-I”,
no matter what the content of this “non-I” may be. This junction forms a new totality which
is of a very special type whose components are not equivalent: in “we” it is always “I” which
predominates since there cannot be “we” except by starting with “I”, and this “I” dominates
the “non-I” element by means of its transcendent quality. The presence of “I” is constitutive
of “we.” (Benveniste 1971 [1966], 202)

On the one hand, then, when somebody says ‘we’, it is a singular ‘I’ speaking for
or on behalf of a group. There are only very few cases when ‘we’ can have a voice
in the full sense of the word. That is, when all of the singularities constituent of
‘we’ can speak in unison: in hymns, collective prayers, etc. On the other hand, the
uttered ‘we’ transcends, in some sense, the mere ‘I’ and can create a collectivity,
a collective identity; and, at least in fiction, this collectivity can have a plural,
collective narrative voice. Although classical narratology follows step-in-step
enunciative linguistics and claims that the narrating voice is always an ‘I’, that
there is always one single we-sayer on the highest level of narrative embedding,14
I suggest a (fictional) possibility of a plurally collective voice: a voiced totality
whose components are equivalent, even if only in the moment of speaking.
The ambiguity of the we-form is further complicated by the ability of this
pronoun to bring about a change of referent within the same morphological form
which, in its turn, brings about a change of perspective. That is, a we-group might
change with respect to the number or type of its referents, but this will not be
grammatically marked, since, in the English language, ‘we’ does not change.15 In
literary narratives, I argue, the ‘we’-narration, if allowed to transcend the ‘I’, is a
performative act of the creation of the collective and, in narratological terms, thus
offers a technique that expands the epistemological and cognitive possibilities of
first-person narration.

14 Cf. Genette (1988 [1983], 97), Prince (1982, 7–15) and Margolin (1996, 123), among others.
15 Cf. analysis of referential ambiguity in Joyce Carol Oates (1999) by Marcus (2008b, 5–6). In
other languages the referential ambiguity of the ‘we’-pronoun might be a lot more limited: e. g.,
in Slovenian the forms of the verb signal the composition of the ‘we’ (as ‘you and me’ or ‘we all’).
I am grateful to Wolf Schmid for this observation.
110   Natalya Bekhta

3 Definition of We-Narrative
One of the first attempts to address the phenomenon of narration in the plural
was Susan Lanser’s book Fictions of Authority (1992). Lanser’s investigation
focused on the relationship between narrative form and form-dependant con-
struction of the authority of narrative voice. She introduces the notion of ‘commu-
nal voice’ to complement the categories of authorial and personal voices evoked
by authorial and first-person narrative situations. Communal narrative voice is
“either a collective voice or a collective of voices that share narrative authority”
(Lanser 1992, 21)—authority that is invested in a definable community. This voice
can be produced in “a singular form in which one narrator speaks for a collective,
a simultaneous form in which a plural ‘we’ narrates, and a sequential form in
which individual members of a group narrate in turn”—that is, in a collaborative
effort (Lanser 1992, 21).
Lanser’s choice of the adjective ‘communal’ to refer to this new type of nar-
rative voice—as opposed, for example, to Margolin’s (2000) more general ‘col-
lective’—has important implications. Based on Raymond Williams’ (1985 [1983])
observations, ‘collective’ seems to be the most encompassing and general desig-
nation. It was originally used as an adjective “to describe people acting together”
(Williams 1985 [1983], 69). As a subject, it came to cover both a physical descrip-
tion of a group of people and the social and political sense of a specific unit.
‘Community’, on the other hand, as a form of social organization, refers to “the
more direct, more total and therefore more significant relationships” as opposed
to “the more formal, more abstract and more instrumental relationships of state,
or of society” (Williams 1985 [1983], 76). Its distinctive meanings are those of
direct relationships, immediacy or locality. Therefore, in the context of we-narra-
tion, I treat ‘communal’ and ‘collective’ as specific thematic types of plural narra-
tion. In some cases, however, it will be appropriate to use ‘collective’ narrator—
in its first, simpler sense of people acting together—as a synonym for ‘plural’ or
‘we-narrator’. ‘Communal’, however, remains reserved to a we-narrator who, as
a character, is manifest in a ‘more direct’ and local collective of people, as in the
village of Jefferson in “A Rose for Emily,” the community of colleagues in Then We
Came to the End, the group of boys and the suburban neighborhood in The Virgin
Suicides, and so on.
As I have suggested earlier, ‘we’ in narrative fiction is used in two contexts:
in the indicative and in the performative. The performative ‘we’ is the core of the
plural, collective narrator which is the key characteristic of we-narrative proper.
Plural narrator can be described as the collective narrative agent, in Margolin’s
sense: it is “a group of two or more individuals represented as a singular higher
order entity or agent, a collective individual so to speak, with global properties
 Emerging Narrative Situations: A Definition of We-Narratives Proper    111

or actions” (2000, 592). The ‘we’ in example (1) is a group of two individuals but
does not compose a “singular higher order agent,” being, rather, a situational
combination. The ‘we’ in example (2) is already such “collective individual.”16
For Lanser, on the other hand, communal narration is not only a we-narra-
tion and not just a multiperson narration, but a narration in which “narrative
authority” belongs to a community and is expressed through singular or plural
narrative voice, authorized to speak by that community. In recent usage, the
term “‘we’-narrative,” it seems, has come to be employed in all three senses, and
also in senses that Lanser (1992, 21) did not consider as instances of communal
voicing: inclusive ‘we’, often used by George Eliot, Faulkner’s multiple narrators
and in cases of several narrations of one and the same event, etc.17
The definition I propose limits we-narrative to Lanser’s simultaneous com-
munal narration only, i. e., a form of collective narrative in which a plural ‘we’
narrates. Why such a limitation? The purpose of the narratological taxonomy of
narrative situations is to give a rough description of a particular narrative form,
its prototype, which will combine both structural and rhetorical-performative
aspects. My purpose here is to define a new narrative form rather than to describe
instances of peculiar we-narration. Single narrators speaking on behalf of a com-
munity, multiple narrators who narrate in turns and ‘I’-narrators, who indica-
tively use the pronoun ‘we’ to refer to themselves and other characters—these do
not qualify as new narrative forms.
In order to be classified as a narrative situation, a narrative form must exhibit
a combination of certain formal characteristics. According to Stanzel, these
are person (homo- or heterodiegetic, using Genette’s terminology), perspective
(internal or external) and mode (teller or reflector). With we-narratives, the dom-
inant characteristic is that of person. The narrator designated by the personal
pronoun ‘we’ is both the teller and the character, i. e., a homodiegetic narrator.

16 Note that Margolin’s “collective narrative agent” is a broad designation and encompasses a
variety of plural subjects: those covered by plural pronouns (we, they), definite articles/num-
bers/quantifiers + plural nouns (the girls, a hundred men, some of the students) and collective
nouns (the gang) (cf. Margolin 2000, 592–594). Here it is a first-person plural that is understood
under a collective narrative agent.
17 Most recently, Patrick Colm Hogan (2013, 248) has taken up the issues of plural narrating
voice under the heading of “group narration,” repeating Susan Lanser’s classification: “instan-
tiated group narration,” (an individual speaking for the entire group), “distributed group narra-
tion” (where group members speak “as distinct parts of the group”) and “collective voicing” (i. e.,
we-narration). However, he excludes any performative aspect (e. g., communal or collective) and
thus the distinctions very easily become designations of any reference to a group rather than a
specific reference to narrative forms.
112   Natalya Bekhta

This excludes more straightforward cases of authorial inclusive ‘we’ (when ‘we’
is used only on the level of discourse) and indicative uses of ‘we’ by ‘I’-narra-
tors (only on the level of story). Furthermore, we-narrative is characterized by a
shifting perspective (of which more below) and teller mode, even though such a
narrative often acquires the dynamic of the mode of lyric progression.
In light of these considerations, we-narrative proper, even though definable
in terms of the classical categories of narrative situation, calls for a modification
of these categories. To start with the latter, teller mode in the case of plural nar-
ration acquires new characteristics which can best be described with the help of
Phelan’s term ‘lyric progression’ (2005, 158–159). Phelan uses the phrase ‘narra-
tive progression’ to refer to “the synthesis of the narrative’s internal logic, as it
unfolds from beginning through middle to end, with the developing interests and
responses of the audience to that unfolding” (19). Typically, narratives proceed
by the introduction, complication and (partial) resolution of instabilities involv-
ing matters of story (characters and their situations) and/or discourse (e. g., in
the case of unreliability). Lyric narratives, however, combine features of narra-
tive and lyric poetry and are characterized by, firstly, the lack of change in the
character narrator. Second, they focus on present situations: “even if […] most
of the sentences of the text focus on past events, the narrative still directs our
primary interest to the present situation of the character narrator” (Phelan 2005,
158). And, finally, the storyworld plays a secondary role for the reader’s ethical
engagement as she engages with the “underlying value structure of the lyric nar-
rative” (159).
Typically, we-narratives proper, to a greater or lesser extent, approach this
mode of lyric progression. For example, Julie Otsuka’s we-novel The Buddha in
the Attic (2011) describes the general situation of immigrant Japanese women in
the U. S. in the early 1900s. At first sight, Otsuka tells a tragic life-story of Japa-
nese women who travel to the US as ‘picture brides’ in the search of a better life.
To do this, she uses the plural narrator but also adds a particularization of the
collective voice which is similar to Ferris’s (‘one of us’, ‘some of us’). However,
despite the particularizations, individual stories of these women yield themselves
to a generalization as the narrative progresses. The story ceases to develop in the
typical sense of ‘what happens next’ and lingers more and more on the present
situation of the immigrant women. Despite their differences, women’s stories are
alike in one crucial factor: they are all in the same social position. Being sent off
to marry men they have never met before, these women find themselves under
oppression and exploitation on both the domestic and public levels:

(3) One of us blamed them [husbands] for everything and wished they were dead. One of
us blamed them for everything and wished that she were dead. Others of us learned to live
 Emerging Narrative Situations: A Definition of We-Narratives Proper    113

without thinking of them at all. We threw ourselves into our work and became obsessed
with the thought of pulling one more weed. We put away our mirrors. We stopped combing
our hair. We forgot about makeup. Whenever I powder my nose it just looks like frost on a
mountain. […] We cooked for them. We cleaned for them. We helped them chop wood. But it
was not we who were cooking and cleaning and chopping, it was somebody else. And often
our husbands did not even notice we’d disappeared. (Otsuka 2011, 36–37)

The unifying and homogenizing quality of the we-voice thematizes the disappear-
ance of the women’s individuality. Yet, Otsuka avoids the potential emotional
dryness of a uniform voice by retaining the particularizations and by inserting
quoted speech or thought of individual women, usually limited to one sentence
or line. Everything individual, however, always feeds back into the collective.
Very soon the focus shifts yet again—from the common misfortunes of women to
those of the ethnic minority in general. Due to the morphological and referential
nature of the ‘we’ pronoun, this transition to a different type of we-voice is seam-
less: toward the end of the novel, it turns out that the story is about the insignif-
icance and effacement of the whole community—both women and men—under
the working conditions for immigrants and racial attitudes in American society
during the 1990s. The reader’s ethical engagement, as Phelan notices in lyrical
narratives, becomes more than engagement with the fates of individual charac-
ters (whose stories are discernable threads in the collective one) and concerns the
general ‘value structure’ of the novel.
We-narrative proper is thus a narrative in which the first-person plural
pronoun is used on both the level of discourse and on that of the story to des-
ignate the narrating instance(s) that are also the narrated entities.18 In the ideal
case, this means that there is no reference to the ‘I’19—indeed, no possibility of
such a reference—and the narrator is thus a fully-established community or col-
lectivity (cf. office co-workers and women-emigrants in the examples (2)-(3)).
Consequently, we-narrative proper is characterized by its peculiar epistemolog-
ical composition, as it expresses collective knowledge. Another factor is focaliza-

18 Cf. Fludernik (2011, 105), Margolin (1996, 115–116 and 122). Obviously, such a we-group may
be the protagonist of the story (as in Buddha in the Attic) or a witness (as in Zakes Mda’s Ways of
Dying, 1995) or it can play some other role a narrator-character can assume.
19 There is, however, a grey area on the scale from instances of we-narration in otherwise
first-person narratives and we-narratives proper. The main tension in Toby Litt’s novel deadkid-
songs (2001), for example, is created due to the strong sense that there is a reconstructible singu-
lar speaker behind the mask of the we-narrator. Technically, the novel is a first-person narrative
because of the framing one-paragraph story in which a son explains that what follows is his
father’s manuscript, a suicide note of sorts. The manuscript is a highly intricate combination of
different narrative situations, the most prominent of which is the we-narrative.
114   Natalya Bekhta

tion, which combines the possibilities of homo- and heterodiegetic narrator. More
generally, as example (3) demonstrates, a basic feature of collective we-narratives
is the alternation between the individual and the collective: “Both individual and
collective levels exist concurrently and are irreducible to each other, so that [there
is] an unresolved tension between the two” (Margolin 2000, 592). This tension
will become most visible in collective focalization.

4 Narrative Levels in We-Narration

Describing we-narratives in terms of narrative levels is known to be problematic.
Homodiegesis becomes an unstable qualifier, since a we-narrator exhibits fea-
tures of narrators from both authorial and first-person narratives, especially when
it comes to ‘our’ knowledge. Richardson, for example, observes how we-narra-
tion “curiously occupies” both hetero- and homodiegetic levels at once, disclos-
ing the contents of individual characters’ minds: a homodiegetic narrator dis-
closes “that which can only be known by an external heterodiegetic intelligence”
(2009, 154–155). In other words, as a character in the narrated story-world—as
a homodiegetic singular (undisclosed) member of a group—the ‘we’-narrator
cannot be simultaneously omniscient and act in a heterodiegetic authorial role
by telling stories of the group’s individual members or other characters. Richard-
son’s observation, however, is implicitly rooted in the classical conception of the
narrator.20 In the next section, I shall argue that this should not be the case with a
collective narrator, if this narrator can be understood as a plurality. A we-narrator
is a peculiar, amorphous one that spans across time and space and cuts across
the level distinction.
Fludernik, without relying on the limiting classical conception of the narra-
tor, nevertheless speaks of a “puzzling and disturbing” “ontological illogicality
of a crossing of borders” (2011, 120–121) in we-narration. She observes how, in
examples of pluralis auctoris, “the personae stationed on the extradiegetic level
of the narrative seem to be moving metaphorically into the fictional world” (2011,
121)—a move for which the communication model of narrative cannot account.
Extradiegesis thus becomes problematic, as the authorial and reader positions
coincide with the intradiegetic we-voice in the storyworld. Fludernik suggests
that, since the boundary between the extradiegetic and the diegetic levels is

20 Cf. his analysis of Conrad’s we-narrator (Richardson 2009, 146). For an overview of the so-
called pan-narrator theory of narrative and a convincing critique, see Köppe and Stühring (2011).
 Emerging Narrative Situations: A Definition of We-Narratives Proper    115

not watertight, there is a continuity between the fictional and factual realms: a
we-reference makes the real-world reader immerse into a fictional text (2011, 122).
While it might be argued that with real readers this kind of immersion may some-
times be the case, Fludernik’s approach also extends similar immersive abilities
to fictional narrators: for her, extradiegetic narrators are outside fiction, and in
we-narratives, “at key moments they step into the fiction and seem to become
part of it” (2011, 120). This, of course, is already a reference to real authors rather
than to fictional narrators. Narrators remain fictional no matter what the diegetic
level, except when they become comparable to real authors who may appear to
be talking in their own voices, even though this is under the regime of fictionality
(cf. Walsh 1997, 510–511).
Without going into a more extensive debate, it suffices to say that we-narra-
tives demonstrate the limitations of Genettian narrative levels in several ways.
What could then be a more suitable terminological language for a description
of the workings of we-narrative? One of the few radical critiques of the commu-
nicative model for fiction and Genettian level distinction has been launched by
Richard Walsh. As Walsh observes in his essay “Person, Level, Voice” (2010),
Genette’s model of levels and persons is self-contradictory, and the distinctions
cannot be sustained, even notwithstanding the complications of plural narra-
tion. Homo-/heterodiegesis, being “a matter of the status of narrative act,” is an
epistemological matter (Walsh 2010, 38): whether or not the narrator is involved
in the story has consequences for the types and scope of information she can be
privy to (e. g., other character’s thoughts). This distinction rests on the ontologi-
cal discontinuity between the levels.
Extra-/intradiegesis is an ontological matter of narrative embedding and
becomes problematic at the extradiegetic level, since it both demands and con-
tradicts ontological discontinuity. For Genette, the extradiegetic narrator has to
be both outside diegesis (extradiegetic) (as Fludernik also notes) and still part of
fiction, that is both representing (in terms of level) and represented (in terms of
person, e. g., homodiegetic). Extradiegetic homodiegetic narrators, which would
be a typical case of we-narrators, “are indeed characters, and if there is any
meaninglessness lurking in that formulation, it can be located in the concept of
the extradiegetic itself. Narrators are always outside the frame of the stories they
tell: ‘Extradiegetic’ appears to have an additional force of placing the narrator
outside representation” (Walsh 1997, 498).
What Walsh advocates, instead of the Genettian approach, is a return to
Plato’s distinction between diegesis (the poet speaking in her own voice) and
mimesis (the poet speaking in the voice of a character) with the acknowledgement
that any narrative may contain other narratives. The consequences this rhetorical
model has for the problem of the we-narrator is that it allows for a simple but
116   Natalya Bekhta

liberating recognition of “fiction’s imaginative freedom” (Walsh 2010, 45). Thus,

a narrator in Then We Came to the End (2007) can be a group of office co-workers,
and this group can tell stories which might contain other stories in other narrative
modes, as is the case in a chapter written in third-person figural narration of Lynn
Manson’s battle with cancer (Ferris 2007, 196–230). Authors may choose to create
(i. e., ‘imitate’, speak ‘as if’) collective and communal voices. By virtue of being a
collectivity and not due to belonging to a certain level, these narrators can make
use of particular forms of collective knowledge, spatial, temporal and other pos-
sibilities outside the scope of a typical first-person singular narrator.

5 Plural Narrator
We-narrative proper establishes the dominance of a collective narrative agent
and makes it analytically unproductive to insist on the singularity of the narrator
that would somehow frame such collective narration. A plural, collective narra-
tor is not ‘actually’ a covert singular one on a virtual extradiegetic level, nor is it
an anonymous member of the we-group who ‘actually’ does all the talking but
without referring to herself.
As an example of false interpretative problems that the latter approach
creates, one can refer to numerous attempts to establish the singular gendered
identity of the narrator behind a we-narration in Faulkner’s short story “A Rose
for Emily.” Without any evidence from the text, the narrating ‘we’ has neverthe-
less been referred to as an anonymous male narrator who is either Emily’s neigh-
bor, a former suitor or an old friend of her father’s.21 But it is precisely by refusing
to provide any textual evidence as to gender and singular identity of the narrator
that Faulkner constructs a townsfolk community of Jefferson, thus conveying an
image of the South and creating the ethical ambiguity of “A Rose for Emily.” The
reader of the story faces a moral dilemma and is unable to either condemn or
justify Emily for the murder she committed.
The need to hunt for evidence of a singular narrator has been motivated
partly by the classical definition of the narrator as an always-present singular
structuring authority and partly by the logic of the non-fictional communicative
act applied to fiction: since out of three pronominal persons it is only the ‘I’-sin-

21 The community that narrates has been treated, for example, as an anonymous narrator-neigh-
bour (Kempton 1947, 104), a male “spokesman for the community” (Brooks 1983, 8), “the elderly
narrator, possibly contemporary with Miss Emily” (Skinner 1985, 43) and “an unidentified citizen
of Jefferson” (Towner and Carothers 2006, 64).
 Emerging Narrative Situations: A Definition of We-Narratives Proper    117

gular that can speak, then the ‘we’, just like second and third persons, cannot
and does not have a voice. Telling a story in unison is not possible in non-fictional
However, fictional narrative is not a literal communicative act, and the
absence of the ‘I’ of the speaker is conceivable. For the possibility of such a
conception, one only has to take apart the communicative model of narrative.
Walsh’s work The Rhetoric of Fictionality (2007), for example, provides one such
a step towards a new conception of narrative and a new theory of fiction. The gist
of his critique is that a communicative model of narration rests on narrative trans-
mission taken quite literally. Transmission means a “reproductive mediation of
a prior discourse” (2007, 36) which, in the case of fiction, does not exist prior
to the narration, i. e., it cannot be transmitted. A fictional narrative, however,
can represent a narrative transmission as an imitation of such discursive form
and represent an ‘as-if’ re-telling of a previously known or experienced story. In
other words, narration primarily consists of narrative representation, and only
secondly, if at all, of narrative transmission.22
What this approach to narrative means for a definition of the narrator is that,
firstly, narrators are represented characters and thus that there can be narratives
without such narrators. For the classical approach, the text has to have a fictional
narrator, even if only implicitly, because otherwise there is no way of distinguish-
ing between fiction (in the sense of a report of imaginary events by a fictional nar-
rator) and non-fiction (report of true events by an empirical person, an author).
However, the need for the ever-present narrator disappears if we take fictionality,
not the narrator, as the key distinction between fiction and non-fiction. Conse-
quently, when a narrative represents narrative transmission, as is the case with
we-narratives, narrators can come in all shapes and forms, including communi-
ties or other collectivities.

22 Denying the communicative model, Walsh contends nonetheless that fictional narrative is
indeed a communication. However, communication is not a structure “within which a commu-
nicative model of narrative acts is implied” (Walsh 2010, 35), but rather a real-world communica-
tive act, albeit an act under the regime of fictionality. For a further discussion of the peculiarity
of fictional narratives as real-world communicative gestures, see Walsh (2007). See also Phelan’s
(2011) critique of the communicative model from the standpoint of character narration.
118   Natalya Bekhta

6 Collective Knowledge and Focalization

The construction of a plural narrative agent is often a performative act that con-
stitutes a unified collective. The structural consequence of such a narrative agent
is the possibility of thematizing collective knowledge and employing a combi-
nation of closely interlinked external and internal focalizations. However, the
tendency has been to treat we-narratives (or instances of we-narration) from
the perspective of the first-person realist novel (variously called ‘conventional
fiction’, ‘mimetic narration’ or ‘natural narrative’; cf. Richardson 2009, 144), and
in particular from that of the limitations of a first-person singular narrator in such
novels. This seems to be the reason why the we-narrator’s knowledge has been
predominantly described as ‘transgressive’, ‘problematic’ and otherwise abnor-
mal: “knowledge that they cannot normally have acquired” (2009, 144). Similarly,
when ‘we’ narrate an individual character’s thoughts and feelings, for example,
or when focalization moves from the collective (i. e., the narrator’s) to individual
characters, this is “something of course a first-person narrator, singular or plural,
is not supposed to be able to do” (2009, 145).
The we-narrator’s knowledge spans individual members of the we-group
and makes interpreters wonder “How does the narrator know what he (mainly)
knows?” Hans Skei poses this question about the we-narrator in Faulkner’s “A
Rose for Emily”: how can ‘he’ “have access to information which does not come
from having watched, eavesdropped and listened to secrets and which simply
does not fit in with the implied restrictions of the first-person perspective”?
(Skei 1999, 155). Investigating collective ‘group narration’, Patrick Colm Hogan
observes that it projects uniformity that might be problematic because of political
implications and also because “in the real world, groups are not all so uniform
as collective […] voicing implies,” further raising “issues about narrator knowl-
edge and reliability” (2013, 236).23 Marcus equates we-narration with first-person
singular narration, as the former is also “based on personal experience and is
thus limited to the scope of human knowledge (in contrast with omniscient nar-
ration)” (2008b, 1).
Collective, however, cannot be compared to personal, since it transcends
personal experience. My contention regarding collective knowledge and focaliza-
tion is that these features should be treated under the conditions of we-narrative
proper specified above and not as extensions of a first-person narrative situation.

23 It is important to note that there is a difference between uniformity and collective properties
of the the macro-level that individuals acquire as part of a group. Cf. Margolin (2000, 595) who
relies on a sociological account of intra-group relations by Raimo Tuomela (1995).
 Emerging Narrative Situations: A Definition of We-Narratives Proper    119

The limitations of a first-person singular narrator cease to be those of a we-narra-

tor, if the broader possibilities of the plural narrator are taken into account.
The distinguishing features of the plural narrator are more comparable to
those of the authorial narrator (and to omniscient narration) than they are to
those of the first-person narrator. Hence, these ‘transgressions’ hardly go noticed
by the readers.24 Furthermore, mimetic narration is not a default mode of fiction,
and thus cannot be a measure of transgressions. Nor can mimetic narration in
fiction be equated with non-fictional narration (telling about something as it is). I
wish to avoid, then, any characterization of this narrative technique as problem-
atic even if it does not match the situation in the ‘real world’. I also refrain from
calling the technique of plural narration ‘unnatural’ since, in its current use, this
term remains inconsistent and self-contradictory. As an alternative, Richardson
proposes ‘anti-mimetic’ to speak of those narratives that “[traduce] the conven-
tions of nineteenth century realism” (2009, 150). ‘Anti-mimetic’ thus applies, for
example, to the twentieth-century fiction of postmodernism. Some we-narratives
fall under this criterion, but since my goal is not to describe a historical literary
movement but a narrative situation, I refrain from this term as well.
The question of whether or not one wants to talk about fictional groups in
comparison to real-world ones loses its significance when it comes to collec-
tive knowledge. Knowledge, in the philosophical sense as a justified true belief
that something is so, acquires the very status of knowledge by being collectively
accepted (Tuomela 2004, 112). In other words, knowledge is created institution-
ally and is social and collective. So, whereas collective story-telling voice might
be referred to as impossible and transgressive, when compared to its hypothetical
non-fictional counterpart, collective knowledge loses in this comparison in either
case. If one approaches it as a narrative technique that straddles realist distinc-
tions, a counterpoint can easily be made that collective knowledge is mimetically
justified (see the example of Faulkner’s short story “A Rose for Emily” below).
However, if one treats collective knowledge as such that subverts ‘what it is like
in the real world’, epistemology argues the opposite.
To state the obvious again, fiction can do more. A collective fictional narra-
tive can draw our attention to “the possibility of genuinely sharing ideas, under-
standings, and emotions,” as Hogan (2013, 249) aptly observes. “The cultivation
of such sharing—or at least an aspiration toward it—is important for a range of

24 See Phelan’s (2013, 174–175) analysis of breaks in what he calls the dominant mimetic code of
narration, and especially his examination of the reasons why epistemological ‘transgressions’ in
paraleptic narratives often go unnoticed by readers. I believe Phelan’s list applies very well also
to the case of we-narratives.
120   Natalya Bekhta

authors trying to create a sense of national or other social unity. The use of group
narration is clearly a possible discourse correlate of such shared mentalities”
(249). As a narrative technique, we-narration “glides between the individual sub-
jectivity and collective omniscience, between a strict and a more lax denotation,
and between mental experiences that are entirely, partially, or minimally shared”
(Richardson 2009, 152).
Obviously, authors can choose whether or not to make their plural narrators
comply with ‘real-world’ limitations and to what ends they wish to use a collective
character. Hazard Adams, for example, problematizes the group uniformity ques-
tioned by Hogan in a laborious way: he lists what seem to be all of the opinions
and views of his communal narrator (a group of students and faculty members)
on any topics they bring up. The only way the narration can progress is thanks to
‘our’ agreement on facts because ‘our’ opinions about them are, usually, too mul-
tiple to be shared, generalized and collectively told. Hence, they are presented in
a listing manner:

(4) We quarrel at once. Various groups among us claim that the author does not want
trouble, does want trouble, does not care one way or the other, wishes to trouble us, does
not wish to trouble us, is punishing us, testing us, playing with us or saving us, or is not
really involved one way or the other in the matter. Some claim it is not a matter at all. And
some claim there never is any matter. (Adams 1999, 15)

Adams continues in this manner for most of the page. Sometimes even such
meticulous listing cannot quite cover it: “It is virtually impossible for us to issue
a statement regarding taste, since we represent so many shades of it ourselves”
Then, again, writers need not resort to such extreme meticulousness: the use
of the collective narrator offers broader epistemological possibilities that can also
be justified under the conditions of realism. A good case in point here is “A Rose
for Emily,” where the information about Emily that circulates in the Jefferson com-
munity comes from observations, rumors, gossip and from individual accounts of
the ‘we’-group members that feed back into collective knowledge. What the com-
munity members did not have access to remains unknown: a Baptist minister was
asked to visit Emily but “[h]e would never divulge what happened during that
interview” (Faulkner 1995, 126). In this way the information ‘we’ have about Emily
fits in very well “with the implied restrictions” (Skei 1999, 155) of the first-person
narrator, but of the collective one, of the ‘we’-narrator rather than the ‘I’.
Finally, like collective knowledge, collective focalization can be justified real-
istically. As Margolin observes:
 Emerging Narrative Situations: A Definition of We-Narratives Proper    121

[A]ny ‘we’ sayer who makes assertions such as ‘we felt that…’ or, ‘our current state of mind
is…’ combines in fact immediate knowledge as regards his/her own internal states with
beliefs (second or third person attributions), based upon inference from external data, con-
cerning the internal states of the other individuals in the ‘we’ reference class. (Margolin
1996, 117)25

Thus, a collective statement of the kind ‘We felt that…’ usually does not draw
attention to itself. It is the collective narrator’s access to individual characters’
minds that has been troublesome.
However, the ‘we’-form can also expand cognitive possibilities connected
with the narration of individual minds. The we-narrative’s peculiar collective
focalization combines an external perspective on the narrated with a general-
ized internal one that characterizes the narrating group without giving access
to the group’s mind(s).26 In order to have this, more intimate, access, the two
perspectives can shift to those of individual characters, often in the free indirect
style. This dynamic is visible in the following example of we-narration from Toby
Litt’s novel deadkidsongs (2001). The novel opens similarly to examples (2) and
(3) with ‘us’, four boys, describing ‘ourselves’ as Gang. The ‘I’ of the narrator is
completely removed from the scene (as well as never disclosed openly in the nar-
rative). Located in the present, the ‘we’-voice narrates a story of ‘our’ long-gone
childhood adventures. Often collective perspective gives way to individual per-
spectives and the we-narrator retreats into the background:

25 Cf. Hogan (2013, 237–238) for a similar reliance on Theory of Mind and folk psychology to
explain the realist component in collective focalization.
26 I am aware that a mention of ‘group’s mind’ resonates with Alan Palmer’s (2005) notion of
‘intermental mind’. I do not adopt Palmer’s terminology for two reasons. First, I believe that
Palmer’s ‘literal’ use of ‘intermental mind’ as ‘actual’ social mind (2007, 218) is highly metaphor-
ical and thus that much of what he describes with this category can be handled more precisely
in terms of social conditions or ideology. As Manfred Jahn commented: “If ‘social mind’ were
literal, it should no longer be called ‘mind’, if only to avoid confusion” (Jahn 2011, 252). More-
over, a shared group mentality is a logical consequence of establishing a plural we-narrator as
a character capable of collective action, shared opinions, emotions, etc. Second, the usefulness
of Palmer’s terminology for literary narratives is rather vague: Palmer describes it as “increased
explanatory power” and an economic gain of “postulating one theoretical entity” instead of, for
example, “the two individuals” sharing a mentality (Palmer 2005, 430). Based on his examples,
the idea of intermental mind seems more useful for symptomatic cognitivist analyses which are
beyond the scope of this article, i. e., for the studies of narrative to investigate human cognition
in general and the ways “narratives participate in larger, socially embedded systems for think-
ing” (Herman 2003, 328).
122   Natalya Bekhta

(5) Matthew was having some trouble keeping pace with the rest of us. He was feeling dizzy,
and couldn’t see quite as well as usual. Whilst we walked, Paul tried to keep as much dis-
tance as possible between himself and his father. There was the terrible possibility that his
father might put his arm around him or even kiss him on the cheek. (Litt 2001, 28)

In this passage, ‘our’ external perspective on each of the boys gives way to Mat-
thew’s and Paul’s perception or thoughts. The first sentence is still focalized col-
lectively, whereas the second belongs to Matthew’s scope of perception. Then,
‘we’ take control again only to hand it over to Paul’s perception of the same sit-
uation. Over more extended passages of individual focalization, the collective
recedes into the background completely, while still containing the characters’
It can thus be seen that we-focalization always frames that of its individual
members and does not allow individual passages to be interpreted separately.
This phenomenon of embedded focalizations can be described with the help of
Marie-Laure Ryan’s ‘stack’ model of recursiveness as revised by Walsh (2010,
42–43). What happens when the reader attends to the current narration and focal-
ization is that all intermediate, embedding layers get occluded. In other words, in
the example (5) above the reader attends to the individual minds in the passage
as they come about without worrying about their correlation with the embedding
collective mind. If we treat the analysis of collective focalization not as “a compli-
cated matter of the relation between narrators’ vision and knowledge and char-
acters’ vision and knowledge but a straightforward question of who perceives”
(Phelan 2001, 63), then in the case of we-narrative, perception is oriented by the
collectivity and its individual members interchangeably.

7 Concluding Remarks
To recapitulate, in this article I have attempted to suggest a more rigorous defini-
tion of we-narrative proper as a separate narrative form, distinct from first-person
narrative. Rather than describing as many types of narration in the plural as pos-
sible, I have proposed to define a particular narrative form in its structural and
rhetorical-performative aspects. This, I believe, is the move that can make the
phenomenon of we-narration more accessible for systematic investigation.
The definition I have proposed is based on maintaining the structural distinc-
tion between ‘narration’ and ‘narrative’ and on the contextual difference between
indicative and performative uses of ‘we’-discourses in fiction. By the indicative
‘we’, I mean a prototypical function of the pronoun when an individual speaker
refers to herself and another person or group, indicating that there exists a col-
 Emerging Narrative Situations: A Definition of We-Narratives Proper    123

lectivity, either situationally or on a more general scale. The use of the indicative
‘we’ alone does not create a plural narrating voice that contributes to we-nar-
rative proper. The performative ‘we’, on the other hand, does: it erases the sin-
gular speaker and goes beyond the need for ‘I’. A performative we-reference of
the narrator creates a collectivity with a plural, collective voice that did not exist
before. In other words, I claim that there exists a plural storytelling voice. It has
two salient features: it does not imply or need a singular individual speaking for
the group, and it approaches the narrative mode of ‘lyric progression’. Crucially,
this performative creation of a plural voice comes about through repetition by
duration. That is to say, it does not occur within a single we-reference and can be
distinguished only in the context of a given narrative on the whole.
A we-narrator, whether collective or communal, is thus the first defining
element of we-narrative proper. Recognition of the plural nature of the narra-
tor leads to a more accurate analysis of the epistemological possibilities (rather
than limitations) of we-narratives and of their ability to shift effortlessly between
external and internal perspectives, both collective and individual. This, then,
calls for re-thinking the relationships between the narrated and the narrating and
focalization. Such re-thinking, I suggest, can be made possible within a re-con-
ceptualized theory of narrative.

Works Cited
Primary sources
Adams, Hazard. 1999. Many Pretty Toys. Albany: State University of New York Press.
Cortázar, Julio. 1985. “Usted se tendió a tu lado.” In J. C. Los relatos, 90–99. Madrid: Alianza.
Faulkner, William. 1995 [1976]. Collected Stories of William Faulkner. New York: Vintage
Ferris, Joshua. 2007. Then We Came to the End. New York: Little, Brown and Co.
Levé, Édouard. 2008. Suicide. Paris: P. O. L.
Litt, Toby. 2011. deadkidsongs. London: Penguin.
McInerney, Jay. 1984. Bright Lights, Big City. London: Bloomsbury.
Mda, Zakes. 1995. Ways of Dying. New York: Farrar, Straus and Giroux.
Moore, Lorrie. 1985. Self-Help. New York: Warner Books.
Moravia, Alberto. 1971. Io e lui. Milano: Bompiani.
Oates, Joyce Carol. 1999. Broke Heart Blues. New York: Dutton.
Otsuka, Julie. 2011. The Buddha in the Attic. London: Fig Tree.
124   Natalya Bekhta

Secondary sources
Benveniste, Émile. 1971. “Relationships of Person in the Verb.” In É. B. Problems in General
Linguistics, translated by Mary Elizabeth Meek, 195–204. Coral Gables, FL: University of
Miami Press.
Brooks, Cleanth. 1983. William Faulkner: First Encounters. New Haven, CN: Yale University
Cohn, Dorrit. 1981. “The Encirclement of Narrative: On Franz Stanzel’s Theorie des Erzählens.”
Poetics Today 2 (2): 157–182
Dieltjens, Sylvain, and Heynderickx Priscilla. 2007. “Strategic Uses of the Pronoun We in
Business Communication.” In Discourse, Ideology and Specialized Communication, edited
by Giuliana Garzone and Srikant Sarangi, 233–249. Bern: Peter Lang.
Dorris, Michael. 1987. A Yellow Raft in Blue Water. New York: H. Holt.
Fludernik, Monika. 1993. “Second-Person Narrative: A Bibliography.” Style 28 (4): 525–548.
Fludernik, Monika. 2011. “The Category of ‘Person’ in Fiction: You and We Narrative-Multiplicity
and Indeterminacy of Reference.” In Current Trends in Narratology, edited by Greta Olson,
100–141. Berlin and New York: De Gruyter.
Fulton, Dawn. 2003. “‘Romans des Nous’: The First Person Plural and Collective Identity in
Martinique.” The French Review 76 (6): 1104–1114.
Genette, Gérard. 1980 [1972]. Narrative Discourse: An Essay in Method. Translated by Jane E.
Lewin. Foreword by Jonathan Culler. Ithaca, NY: Cornell University Press.
Genette, Gérard. 1988 [1983]. Narrative Discourse Revisited. Translated by Jane E. Lewin.
Ithaca, NY: Cornell University Press.
Helmbrecht, Johannes. 2002. “Grammar and Function of We.” In Us and Others: Social
Identities across Languages, Discourses and Cultures, edited by Anna Duszak, 31–49.
Amsterdam: Benjamins.
Herman, David. 2003. “Regrounding Narratology: The Study of Narratively Organized Systems
for Thinking.” In What is Narratology? Questions and Answers Regarding the Status of a
Theory, edited by Tom Kindt and Hans-Harald Müller, 303–332. Berlin: Walter de Gruyter.
Hogan, Patrick Colm. 2013. Narrative Discourse. Authors and Narrators in Literature, Film, and
Art. Columbus: The Ohio State University Press.
Jahn, Manfred. 2011. “Mind = Mind + Social Mind? Response to Alan Palmer’s Target Essay.”
Style 45 (2): 249–254.
Kempton, Kenneth Payson. 1947. The Short Story. Cambridge, MA: Harvard University Press.
Köppe, Tilmann, and Jan Stühring. 2011. “Against pan-narrator theories.” Journal of Literary
Semantics 40 (1): 59–80.
Lanser, Susan Sniader. 1992. Fictions of Authority. Women Writers and Narrative Voice. Ithaca,
NY: Cornell University Press.
Marcus, Amit. 2008a. “Dialogue and Authoritativeness in ‘We’ Fictional Narratives: A
Bakhtinian Approach.” Partial Answers: Journal of Literature and the History of Ideas 6 (1):
Marcus, Amit. 2008b. “We are You: The Plural and the Dual in ‘We’ Fictional Narratives.” Journal
of Literary Semantics 37 (1): 1–21.
Margolin, Uri. 1996. “Telling Our Story: On ‘We’ Literary Narratives.” Language and Literature 5
(2): 115–133.
Margolin, Uri. 2000. “Telling in the Plural: From Grammar to Ideology.” Poetics Today 21 (3):
 Emerging Narrative Situations: A Definition of We-Narratives Proper    125

Margolin, Uri. 2001. “Collective Perspective, Individual Perspective, and the Speaker in
Between: On ‘We’ Literary Narratives.” In van Peer and Chatman, eds., 241–254.
Palmer, Alan. 2005. “Intermental Thought in the Novel: The Middlemarch Mind.” Style 39 (4):
Palmer, Alan. 2007. “Universal Minds.” Semiotica 2007 (165): 205–225.
Peer, Willie van, and Seymour Chatman, eds. 2001. New Perspectives on Narrative Perspective.
Albany: State University of New York Press.
Phelan, James. 2001. “Why Narrators Can Be Focalizers—and Why It Matters.” In van Peer and
Chatman, eds., 51–64.
Phelan, James. 2005. Living to Tell About It. A Rhetoric and Ethics of Character Narration. Ithaca,
NY and London: Cornell University Press.
Phelan, James. 2011. “Rhetoric, Ethics, And Narrative Communication: Or, from Story and
Discourse to Authors, Resources, and Audiences.” Soundings: An Interdisciplinary Journal
94 (1/2): 55–75.
Phelan, James. 2013. “Implausibilities, Crossovers, and Impossibilities: A Rhetorical Approach
to Breaks in the Code of Mimetic Character Narration.” In A Poetics of Unnatural Narrative,
edited by Jan Alber, Henrik Skov Nielsen and Brian Richardson, 167–184. Columbus: The
Ohio State University Press.
Prince, Gerald. 1982. Narratology. The Form and Functioning of Narrative. Berlin and New York:
Richardson, Brian. 1994. “I Etcetera: On the Poetics and Ideology of Multipersoned Narratives.”
Style 28 (3): 312–328.
Richardson, Brian. 2006. Unnatural Voices: Extreme Narration in Modern and Contemporary
Fiction. Columbus: The Ohio State University Press.
Richardson, Brian. 2009. “Plural Focalization, Singular Voices: Wandering Perspectives in
‘We’-Narration.” In Point of View, Perspective, and Focalization: Modeling Mediation in
Narrative, edited by Peter Hühn, Wolf Schmid and Jörg Schönert, 143–159. Berlin and New
York: Walter de Gruyter.
Reitan, Rolf. 2011. “Theorizing Second-Person Narratives: A Backwater Project?” In Strange
Voices in Narrative Fiction, edited by Per Krogh Hansen, Stefan Iversen, Henrik Skov
Nielsen and Rolf Reitan, 147–174. Berlin and Boston: De Gruyter.
Schmid, Wolf. 2010 [2005]. Narratology: An Introduction. Berlin and New York: De Gruyter.
Skei, Hans. 1999. Reading Faulkner’s Best Short Stories. Columbia: University of South Carolina
Skinner, John L. 1985. “‘A Rose for Emily’: Against Interpretation.” JNT 15 (1): 42–51.
Stanzel, Franz Karl. 1971 [1955]. Narrative Situations in the Novel: Tom Jones, Moby Dick, The
Ambassadors, Ulysses. Translated by James Pusack. Bloomington and London: Indiana
University Press.
Stanzel, Franz Karl. 1984 [1979]. A Theory of Narrative. Translated by Charlotte Goedsche.
Cambridge: Cambridge University Press.
Towner, Theresa M., and James B. Carothers. 2006. Reading Faulkner. Glossary and
Commentary. Jackson: University Press of Mississippi.
Tuomela, Raimo. 1995. The Importance of Us: A Philosophical Study of Basic Social Notions.
Stanford, CA: Stanford University Press.
Tuomela, Raimo. 2004. “Group Knowledge Analyzed.” Episteme 1 (2): 109–127.
Walsh, Richard. 1997. “Who Is the Narrator?” Poetics Today 18 (4): 495–513.
126   Natalya Bekhta

Walsh, Richard. 2007. The Rhetoric of Fictionality: Narrative Theory and the Idea of Fiction.
Columbus: The Ohio State University Press.
Walsh, Richard. 2010. “Person, Level, Voice: A Rhetorical Reconsideration.” In Postclassical
Narratology: Approaches and Analyses, edited by Jan Alber and Monika Fludernik, 35–57.
Columbus: The Ohio State University Press.
Williams, Raymond. 1985 [1983]. Keywords: A Vocabulary of Culture and Society. New York:
Oxford University Press.
Wittig, Monique. 1973. Le corps lesbien. Paris: Les Éditions de Minuit.
Wolf, Christa. 1968. Nachdenken über Christa T. Halle (Saale): Mitteldeutscher Verlag.
Nora Berning (Gießen)
Critical Ethical Narratology as an
Emerging Vector of Narrative Theory and
Autobiographical End-of-Life Stories
Ever since David Herman first made the distinction between classical and post-
classical narratology in his programmatic 1997 article “Scripts, Sequences,
and Stories: Elements of a Postclassical Narratology,” there has been a verita-
ble boom of “hyphenated and compound narratologies” (Nünning 2003, 258).
Among the most prominent postclassical approaches, Vera and Ansgar Nünning
(2002) have identified three productive branches that refer to the development of
transgeneric, intermedial and interdisciplinary narrative theories. Critical Ethical
Narratology (CEN), an analytical framework specifically designed for the analy-
sis of the representation, dissemination and construction of values and norms in
hybrid genres as one finds them in various types of media (Berning 2013), draws
on all of these border-crossings and thus bridges the gap between classical, struc-
turalist narratology and postclassical narrative theory.
Rooted in the disciplinary triangle of literary and cultural studies, media
studies and moral philosophy, CEN can be considered an emerging vector of nar-
ratology. While the focus of this article is on nonfiction in general and, more spe-
cifically, the appearance of nonfictional genres characterized by literary elements
on the Internet such as Weblogs, or blogs, points to narratology’s continued
trend toward diversification, the analytical framework of CEN reflects a phase
of consolidation insofar as the forms and functions of narrative are at the fore-
front of not just ethical but also other double-entry narratologies. Emerging lines
of research are currently one of the most reliable indicators of the fact that the
artificial divorce between narratology and various theories of interpretation has
not done much for either side. Instead of propagating a strict separation between
description and interpretation, a twenty-first century narrative theory should
combine systematic poetics and interpretive activities and be conceptualized in
such a way that transgeneric, intermedial and interdisciplinary expansions form
an integral part of narratological analyses.
One of the consequences of the paradigm shift in narratology from classical,
structuralist narratology to postclassical narrative theory has been a renewed
interest in politics and the ways in which narratives contribute to cultural
world-construction. In addition, the “narratological renaissance” (Herman 1999,
2) has led narratologists to reconsider Fredric Jameson’s concept of the ideology
of form (Jameson 1983 [1981], 141). However, postclassical narrative theory has

DOI 10.1515/9783110555158-007
128   Nora Berning

not yet produced a satisfactory account of the ways in which “narrative structures
are ideologically informed” (Helms 2003, 7). Hence, CEN is best understood as an
incentive for revisiting Jameson’s form-as-content argument with regard to the
rise of new genres and media formats.
Both genre-specific and media-specific structures shape a work’s dialectic
of form and content and thus inform processes of cultural world-construction.
Form, genre and mediality trouble not only our understanding of what narratives
are and what they can do, but they also serve as a forceful reminder that narra-
tives embody a tacit knowledge that cannot be reduced to a fixed moral message
(Eagleton 2012). Although critical appraisal of form, genre and mediality is a nec-
essary step in the direction of a twenty-first century ethical narratology, CEN is
more than the sum of its parts (form, genre and mediality). Rather, it is an ana-
lytical framework that merges thematic and formal narratologies (cultural exten-
sions of classical narratology), ethical narratology and transmedial narratology,
and in this way it narrows the gap between classical, structuralist narratology
and postclassical narrative theory.
CEN is not only a particularly fruitful framework for shedding light on the
ways in which narrative techniques and strategies are semanticized (Nünning
2013) to the extent that they convey culturally specific notions of storytelling
and ideologically charged values and norms. It is also a valuable framework for
attending to recent developments in genre theory and such important processes
as hybridization and medialization and the ways in which they function as cata-
lysts for generic evolution, change or innovation and for the development of new
thematic foci in emerging genres and media formats (Basseler et al. 2013).
Though not new in terms of their thematic focus, autobiographical end-of-life
stories published on the Internet are best understood as an emergent genre that
distinguishes itself from ‘traditional’ autobiographical writing through its inno-
vative techniques and strategies of narrating death and dying. For the purpose
of this article, the autobiographical end-of-life story will be defined as a factual
illness narrative (Hydén 2005) written by a terminally ill person who makes use
of fictionalizing strategies in the story. Autobiographical end-of-life stories pub-
lished on the Internet such as the famous Schlingenblog (2008–10) by Christoph
Schlingensief or Arbeit und Struktur (2010–13) by the now deceased best-selling
author Wolfgang Herrndorf pose a number of challenges to narrative theory.
It is worth mentioning that autobiographical end-of-life stories, which can be
defined as a hybrid genre that oscillates between the worlds of fact and fiction,
do not figure prominently in classical, structuralist narratology. If narratologists
want to cease neglecting generic hybrids, however, the concepts, models and
methods used for analyzing them have to be fine-tuned to the epistemological
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   129

underpinnings of hybrid genres that proliferate not just in the electronic environ-
ment but also in other media (Berning 2011).
Second, end-of-life stories present at the same time a great challenge to con-
temporary narrative theory and genre theory. Life-writing—which in the context
of autobiographical end-of-life stories means narrating about one’s own death as
implied by the notion of autothanatography (Derrida 1987 [1980], 393)—questions
narrative theory as regards the relationship between narrative and knowing and,
more specifically, the role of narrative form in the production of knowledge. Fur-
thermore, these particular kinds of stories challenge the static models of formal-
ism and structuralism which contributed to the idea that “[t]he human became
an effect of the system, rather than the reverse” (Gibson 1996, 245). Consequently,
the task for a CEN that concerns itself with autobiographical end-of-life stories is
to try to re-humanize the narratological toolkit so that it can effectively deal with
the human problem of suicide (Mandle 1984).
If one endorses, moreover, Anne Nesbet’s (1991) thesis that end-of-life stories
are a sign of genre in crisis, then this explains not just the ongoing fascination of
genre theorists with suicide and death, but it also invites narratologists to con-
ceive of such stories both as an aesthetic means and as fulfilling certain norma-
tive functions, besides the cognitive and emotional ones (Baumbach et al. 2009).
Suicide is simultaneously a literary-cultural subject with a long literary history
(extending from Cato, young Werther and Madame Bovary to Anna Karenina) and
a media phenomenon (e. g., Bronfen 1992; Skelton 2003). Besides the mediatized
suicides of the German soccer goalkeeper Robert Enke and the photographer and
self-proclaimed playboy Gunter Sachs, a number of suicides that gained traction
in the mainstream media were either made public via social networking web-
sites (e. g., Twitter, Facebook) or they were webcasted. For instance, John Patrick
Bedell left YouTube videos that could be read as a suicide note; Paul Zolezzi indi-
cated via Facebook his intention to commit suicide; and Kevin Whitrick as well as
Abraham K. Biggs webcasted their suicides. The tendency among so-called digital
natives to interneticize their end-of-life stories suggests that every crisis of genre
contributes to a new configuration of the cultural subsystem of literature. In the
case of autobiographical end-of-life stories, this new configuration is closely
linked to such processes as hybridization (Nünning and Schwanecke 2013) and
medialization (Nünning and Rupp 2013).
Since the present article deals with autobiographical end-of-life stories on
the Internet, specifically Herrndorf’s blog Arbeit und Struktur, the interpretation
and ‘negotiation of values’ (Korthals Altes 2014) poses an additional challenge
because “electronic literature is a ‘hopeful monster’ (as geneticists call adaptive
mutations) composed of parts taken from diverse traditions that may not always
fit neatly together” (Hayles 2008, 4). This is especially true for blogs whose origins
130   Nora Berning

can be described as “a hybrid of existing genres, rendered unique by the particu-

lar features of the source genres they adapt, and by their particular technological
affordances” (Herring et al. 2004, 10). Herrndorf’s suicide blog shares its episodic
structure with epistolary novels and diaries (Walker 2003) and its thematic prior-
ity with the suicide novel. His blog also has some characteristics in common with
the (Internet) diary (Kapp 1987), a genre that also hinges closely on the writer’s
place and circumstance of writing.
These “family resemblances” (Wittgenstein 1968 [1953], 32) with other con-
fessional genres that are currently on the rise (famous examples include, for
example, James Frey’s 2004 fictionalized memoir A Million Little Pieces or Del-
phine de Vigan’s Nothing Holds Back the Night 2013) make the blog a particu-
larly adequate means “to explore the emergent culture of the early 21st century”
(Miller and Shepherd 2004) with the help of the analytical framework of CEN.
CEN assumes that there is an anthropological, psychological and cultural neces-
sity for a non-essentialist ethical discourse for the formation of personal and col-
lective identities under late capital. Using the example of Herrndorf’s blog, it will
be shown in the analytical part of this article that blogs can assist terminally ill
people in mending the fragmented self. Considering that weblogs are not only
widely disseminated but also a particularly controversial site for reframing the
literary-cultural concept of suicide (Berning et al. 2014a), it is quite astonishing
that autobiographical end-of-life stories on the Internet have not yet been exam-
ined through the lens of narrative theory. Valuable though Lore Knapp’s (2012)
analysis of the Schlingenblog is, it says very little about thanatopoetics and the
negotiation of values surrounding suicide (Bronfen 1992; Hook 1927).
Suicidologists define suicide as a multidimensional illness that is inextrica-
bly linked to the Wish To Hasten Death (WTHD), which has a multi-factor basis
encompassing pain, physical suffering, psychiatric disorders and psychological
or existential distress (Monforte-Royo et al. 2011). A fluctuating and unstable
feeling, the WTHD can also be perceived in Herrndorf’s blog. One of the most
urgent needs in clinical studies is to arrive at “[a] better understanding of the
WTHD, one which clarifies its conceptual limits and distinguishes between dif-
ferent stages or situations” (Monforte-Royo et al. 2011, 802). In order to improve
our knowledge of suicide and develop adequate interventions, it is fruitful to
bring together clinical studies on suicide and insights from narrative theory. My
analysis of Herrndorf’s blog is therefore best understood as a contribution to the
interdisciplinary field of narrative medicine in general and narratives of illness in
particular (cf. Charon 2008).
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   131

1 Theorizing and Historicizing the Cultural

Dynamics of the Blogosphere
In a study conducted by the Pew Internet & American Life Project, bloggers are
described as “the internet’s new storytellers” (Lenhart and Fox 2006). The study
suggests that storytelling techniques have migrated to the blogosphere. Blogs, a
term coined by the American blogger Jorn Barger in 1997, are “frequently modified
web pages in which dated entries are listed in reverse chronological sequence”
(Herring et al. 2005, 143). The genre expectation of blog entries is nonfiction
(Walker 2003). The specificity of autobiographical entries is rooted in the reader’s
belief that the author, narrator and protagonist of the narrative are one and the
same, as implied by the autobiographical pact (Lejeune 1989 [1973]). Herrndorf,
to illustrate the complex relationship between genre expectations and autobio-
graphical end-of-life stories, is the chronicler of his own life and death: since it is
logically impossible to gain reliable knowledge about death or suicide, an expe-
rience that resists explanation from the inside (Brooks 1985), Arbeit und Struktur
provides an incentive to have a closer look at aesthetic communication and the
negotiation of values on the Internet.
Herrndorf’s blog is based on new ways of conceptualizing authenticity that
invite the reader to problematize and question the medium. It is a kind of authen-
ticity that takes shape in the form of a ‘metafictional maneuver[]’ (Hutchins 2012,
81), for it “does not aim at finding the actual self, but rather at recognition of
the subjective game being played” (81) by the medium in question. The blog’s
structure is determined by an innovative, relational mode of knowledge produc-
tion. The blog merges current thinking and empirical research on suicide with
introspection. Since the narrative oscillates between fact and fiction, the blog
raises intriguing questions about the reliability of the narration. Moreover, hybrid
genres such as autobiographical end-of-life stories serve as an indicator of new
forms of life that are linked to “medical advances, which have transformed dis-
eases that once led to a quick death into chronic illnesses” (Monforte-Royo et al.
2011, 802). Due to these medical advances, the WTHD is gaining more and more
currency in contemporary society.
According to Carolyn Miller and Dawn Shepherd (2009, 266), the history of blogs
can be divided into three phases: prior to 1999, “blogs were used primarily by web-
savvy individuals […] to share information with each other.” At a time when search
engines did not exist, blogs served as a means to exchange links to new websites.
They functioned as a “subjectively uninvolved source of information” (Shankar
2010, 529) that developed independently from the more subjective Internet diary,
which emerged in 1994 and developed out of personal home pages (McNeill 2009).
132   Nora Berning

In 1999, free fill-in-the-blank site-building tools and publishing software became

available. Knowledge of HTML was no longer required, and anyone with an Internet
connection could create a blog by using the new tools and software.

These changes in technology opened the way to the second phase of blogging, with a new
kind of user, younger and less technologically adept, and a new emphasis on personal
commentary rather than links, self-disclosure rather than information sharing. (Miller and
Shepherd 2009, 267)

During this second phase, blogs mushroomed and became less and less distinct
from the Internet diary (McNeill 2009; Herring et al. 2005). The advent of social net-
working websites in 2002, which marks the beginning of the third phase, changed
the way in which blogs were accessed, primarily via a link on the user’s profile.
From a formal point of view, blogs have evolved rather slowly over the last two
decades (Miller and Shepherd 2009). If one wants to provide a comprehensive
description of the blog and its generic variants, however, it is important to pay
attention to formal, functional, communicative and narrative as well as media-spe-
cific features. Though figure 1 does not shed light on the blogosphere from a dia-
chronic perspective, the typology by Ansgar Nünning and Jan Rupp (2012, 31) is
nevertheless useful for differentiating between different variants of the blog.

Figure 1: Classification of blogs according to formal, functional, communicative, narrative and

media-specific features (© 2012 Ansgar Nünning and Jan Rupp)
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   133

Whereas the x-axis represents a continuum between Individual I-blogs and

Community We-blogs, supplemented by a scale indicating an intensification of
interactivity, multiperspectivity and multimediality from left to right, the y-axis
covers a spectrum that ranges from personal modes of narration with a focus on
a limited audience to topical/corporate blogging targeted at a larger audience.
Herrndorf’s blog breaks new ground because it mixes so-called stream-of-con-
sciousness blogging with elements of literary blogs. It is important to point out,
though, that this classification is dependent on whether one conceives of an indi-
vidual blog entry or of the blog as a whole as a unit of analysis. In the former case,
Herrndorf’s blog presents itself as a narrative chameleon that is characterized
by the combination of different literary styles and sign systems. If one considers
the entire blog as a unit of analysis, Arbeit und Struktur closely resembles literary
blogs where multiperspectivity and interactivity as well as multimediality play a
subordinate role.
In both cases, however, the emphasis in terms of the functional aspect is on
individual identity construction (Tophinke 2009). The blog “is a contemporary
contribution to the art of the self” (Miller and Shepherd 2004). Moreover, blogs
give “credence to the commensurate realization […] that we are each of us, in a
global sense, bound inextricably to one another” (Shankar 2010, 540). The search
for empathy, friendship and even love that technological enthusiasts employ in
order to propel web-based tools and applications has become a prolific means
to capitalize on the ideological construct of global interconnectedness. The blog
is, of course, not immune from this logic of capital (Puschmann 2012). A dis-
tinction between static, conventionalized and flexible features shows that blogs
are always partly determined by the software code and partly by the user. The
formal characteristics (deictic markers, perspective, temporality, discursive rela-
tions) and the flexible characteristics (style, length of the entries, thematic foci,
frequency of publication, interactivity, design) that are set by the user have an
impact on the communicative situation and functions of blogs. In Herrndorf’s
blog, the communicative situation is extremely complex and involves high expec-
tations on the part of the blogger toward the reader. He or she is invited to act as a

Witness testifying to the experience; Therapist unconditionally supporting emotions;

Cultural theorist assessing the contestation of meanings, values, and identities in the per-
formance; Narrative analyst examining genre, truth or strategy; and Critic appraising the
display of performance, knowledge and skill. (Scheidt 2006, 202)

That Herrndorf’s communicative goals are closely linked to the imagination of a

diverse readership may seem speculative at first sight, but in Herrndorf’s case the
unseen audience does not operate completely in the shadows. In contrast to the
134   Nora Berning

Schlingenblog, which was accessible to a vast readership from day one, Arbeit und
Struktur did not at first function according to the principle of open access that is
typical of blogs, listservs and online journals (Shankar 2010). When the blog was
made public, however, it became immediately clear that entries are meant to be
read as a “purposive communicative act” (Phelan 2007, 203), that is, as a cry for
help, which serves both a personal and political purpose. In this context, it is
striking that Herrndorf does not make use of one of the principal characteristics
of digital media: interactivity. As a result, his blog remains a self-contained and
self-referential universe that people can comment on only retrospectively in other

2 Herrndorf’s Thanatopoetics, or:

The Negotiation of Values on the Internet
Though the Internet has expanded the opportunities for language use and textual
innovation, electronic literature is not fundamentally different from traditional
literature. Hence, it “can be analyzed well within the parameters of established
categories of literary theory” (Zenner 2005, 6). The reason why Herrndorf’s blog
Arbeit und Struktur (AuS) does not require new interpretive strategies on the part
of its readers is that, in contrast to complex electronic literature such as Michael
Joyce’s afternoon, a story (1987–90), Herrndorf’s blog contains only a few links.
Since there is no possibility for users to comment on the text, the commonly held
view of the death of the author does not apply to the structural level of Herrn-
dorf’s narrative, although there is a grain of truth in it insofar as the author’s
suicide “shapes a new life-narrative within the space defined by death” (Higon-
net 2000, 299).
Herrndorf’s blog, which was written over a period of approximately two-
and-a-half years (8 March 2010–20 August 2013), is divided into forty-two chap-
ters. In addition to the individual chapters, it contains a closing word by Sascha
Lobo, a friend of Herrndorf’s who set up the blog for him. In February 2010, a
few weeks before Herrndorf began to work on the blog, he was diagnosed with a
brain tumor at the age of forty-five. As the title of his blog indicates, the author of
three best-selling books—In Plüschgewittern (2002), Tschick (2010), Sand (2011)
and a collection of short stories titled Diesseits des Van-Allen-Gürtels (2007)—was
obsessed with work both before and after his diagnosis.

I feel the best, when I work. I work in the tram on printouts, I work in the waiting room for
radiotherapy, I work the minute that I have to stand in the changing room, with the paper on
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   135

the wall. I sink into the story, which I write there, as I sank at the age of twelve years when
I read books.1 (AuS, Chap. 4)

[Am besten geht’s mir, wenn ich arbeite. Ich arbeite in der Straßenbahn an den Ausdrucken,
ich arbeite im Wartezimmer zur Strahlentherapie, ich arbeite die Minute, die ich in der
Umkleidekabine stehen muß, mit dem Papier an der Wand. Ich versinke in der Geschichte,
die ich da schreibe, wie ich mit zwölf Jahren versunken bin, wenn ich Bücher las.]

As much as his work as an author and public figure shaped his personal life,
his personal diagnosis shaped Herrndorf’s late work insofar as writing came less
and less naturally to him. Not only does Herrndorf find it increasingly more dif-
ficult to express himself coherently, but he also has trouble connecting with the
world beyond the screen. As a remedy for loneliness and as an attempt to distract
himself from his panic attacks and anxieties, Herrndorf resorted to literature as
“relationship and human connectivity” (Newton 1995, 7).

I am a writer, and one will not believe that literature would otherwise have left me cold. But
what now returns when reading is the feeling that I last had during childhood and adoles-
cence regularly and thereafter only very sporadically and in a few books: that one partic-
ipates in an existence, in people and in the consciousness of people on something about
which you usually do not have much opportunity to learn something in life, even, to be
honest, only rarely, in conversations with friends and even more rarely in movies, and that
there is a difference between art and shit. […] One can of course also take a critical view: The
sliding into a fantasy world as an expression of perfect helplessness. (AuS, Flashback, Pt. 1)

[Ich bin Schriftsteller, und man wird nicht glauben, daß Literatur mich sonst kaltgelassen
hätte. Aber was jetzt zurückkehrt beim Lesen, ist das Gefühl, das ich zuletzt in der Kindheit
und Pubertät regelmäßig und danach nur noch sehr sporadisch und bei wenigen Büchern
hatte: daß man teilhat an einem Dasein und an Menschen und am Bewußtsein von Men-
schen, an etwas, worüber man sonst im Leben etwas zu erfahren nicht viel Gelegenheit hat,
selbst, um ehrlich zu sein, in Gesprächen mit Freunden nur selten und noch seltener in
Filmen, und daß es einen Unterschied gibt zwischen Kunst und Scheiße. […] Man kann das
natürlich auch kritisch sehen: Das Absacken in die Phantasiewelt als Ausdruck vollkomme-
ner Hilflosigkeit.]

In this introspective passage, the author puts emphasis on a relational mode of

knowledge production via narratives. The ethical status attributed to narrative
is based on what Adam Newton (1995) describes as an intersubjective chain of
narrative relations that bind the narrator and the reader in a dialogic relationship

1 Note that this article was composed before Herrndorf’s blog Arbeit und Struktur was turned
into a book of the same title, published by rowohlt in December 2013. All German quotes are from
Herrndorf’s blog and have been translated by Jill Grinager for the purpose of this article only.
136   Nora Berning

where authenticity and the construction of norms and values emerge from the
interaction of several agents involved in aesthetic communication. In this context,
ethics “signifies recursive, contingent, and interactive dramas of encounter and
recognition” (Newton’s 1995, 12), exactly the sort of dramas which Herrndorf’s
blog transmits. What can be observed in the above-quoted passage, then, is how
norms and values become objects of reflection in their own right. By attributing to
narrative a kind of ethical status, Herrndorf implicitly thematizes the ethics of his
own suicide blog. The narrative situation of Herrndorf’s blog is heavily shaped by
the specter of suicide, which serves both as an aesthetic device that punctuates
the autobiographical quest and as a contentious ethical issue that undermines
the widespread “assumption that life, however painful, is worth living” (K. Ryan
2004, 95).
Arbeit und Struktur can be regarded as a continuation and postmodern
update of a long philosophical tradition of discussions about suicide as well as
the tensions between freedom, determinism and morality in the works of David
Hume, Immanuel Kant, John Stuart Mill, Arthur Schopenhauer and Jean-Paul
Sartre. Through the personal union of the author, the narrator and the suicidal
protagonist named Wolfgang, Herrndorf seeks to establish suicide as “the essen-
tial autobiographical act” (Higonnet 2000, 229) and uses his blog as a medium of
critique in the Kantian sense of the term, that is, as a medium of reasoning and
also revision. Arbeit und Struktur can be read as a revisionist narrative that stands
in stark contrast to the tradition of theologians who purport to know suicide.

“The Lutheran Bishop of Hanover, Margot Käßmann, sees that it is a great danger to provide
patients with the availability of a fast, effective death. ‘It leads one to think one can even
decide about death.’” Piteous stupidity, over-qualified for the office of German President.
(AuS, Chap. 5)

[“Die Landesbischöfin von Hannover, Margot Käßmann, sieht eine große Gefahr darin,
Patienten einen schnellen, effektiven Tod zur Verfügung zu stellen. ‘Es führt dazu, zu
meinen, man könne mal eben über den Tod entscheiden.’” Mitleiderregende Dummheit,
für das Amt des Bundespräsidenten überqualifiziert.]

The narrator’s lack of understanding of Margot Käßmann’s opinion is based on

the belief that the feeling of having an terminal disease generates a kind of expert
knowledge that only those affected have access to. The narrative situation is
shaped, moreover, by the narrator’s view that suicide is sometimes ethically jus-
tifiable. For example, he criticizes the German advance health care directive for
not catering to a patient’s wish to control how and when he or she wants to die.
There are many other examples in the narrative that imply that the blog is geared
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   137

toward an ethical pluralism. The narrator constantly puts emphasis on the values
of living rather than the value of life (Hook 1927).
Herrndorf’s blog inevitably raises questions about the narrator’s reliability,
since suicide is something that lies beyond the borders of the known (Higonnet
2000). Arbeit und Struktur is a narrative that is told by an autodiegetic narrator
who is both suicidal and has a tendency to hallucinate. Considering that the nar-
rator emphatically tries to dismiss his conversations with physicians either as
rubbish or reframes them as if they were dialogues taken from a crime novel that
he is working on, doubts about his reliability are more than justified. The relia-
bility of the first-person narrator is undermined in two ways: on the one hand,
the fact that the blog merges different narrative levels (e. g., metalepsis occurs
between the intra-diegetic and the extra-diegetic level) makes it hard for the
reader to tell whether the narrative is an exercise in autosuggestion or a genuine
attempt at demystifying death. On the other hand, the question of whether the
reader can trust the narrator is linked to generic aspects, specifically the ways in
which the worlds of fact and fiction interpenetrate in the narrative.

Passig names what I write here Wikipedia literature. New meaningless genre overloaded
with reality that owes itself to the simplicity of the research. Defend me with the fact that
most of it is simply invented. Two, three years ago, sometimes even started to write stuff
in the Wikipedia that appeared in my novel. Either the fiction adapts to the reality or vice
versa. The accusation of slovenliness will not be liked. (AuS, Chap. 9)

[Passig nennt das, was ich da schreibe, Wikipedia-Literatur. Neues sinnlos mit Realien
überfrachtetes Genre, das sich der Einfachheit der Recherche verdankt. Rechtfertige mich
damit, daß das meiste ja doch erfunden ist. Vor zwei, drei Jahren auch schon mal angefan-
gen, Sachen in die Wikipedia reinzuschreiben, die in meinem Roman vorkamen. Entweder
die Fiktion paßt sich der Wirklichkeit an oder umgekehrt. Den Vorwurf der Schlampigkeit
will man sich ja nicht gefallen lassen.]

In addition to these verbal cues of unreliability, the multimodal and media-spe-

cific elements of Herrndorf’s blog deconstruct dominant notions of socially
accepted values and norms as well as discourses concerning normal psychologi-
cal behavior (Zerweck 2001). For instance, in the flashback entitled The Penguin
(Pt. 10), which brings the first one-third of the story to a close and is literally
linked to the introductory chapter in the context of which the narrator recounts
his admission to a psychiatric hospital, Wolfgang performs his illness in a theat-
rical fashion. Precisely because the dress of the penguin is beyond the socially
acceptable and serves as a metaphor of cultural alterity, it questions the very
authority of the narrator.
The authority of the narrator is, furthermore, challenged with the help of
a specific kind of psycho-narration (Cohn 1978) that has ethical implications.
138   Nora Berning

The persistent use of what Dorrit Cohn calls ‘fictional present’ is striking in this
context. Herrndorf makes use of techniques of fictionalization that occasionally
contribute to passages of local non-communication that do not appear to be
addressed to someone. In these passages, psycho-narration serves primarily as
a means of rendering transparent the consciousness of his Walther PPK and its
opponent and interferer Wilhelm, the tumor.

When trying to support the already successful Walther in its fight, I personify the disturbing
instance at first as Störer (spoiler), then Wilhelm Störer. I try to speak to him and record his
reactions. In contrast to the Walther he responds as well as not at all and likes to do the
opposite of what I want. So I call on him in quieter periods to show up again and taunt him:
Whether he no longer wanted or could interfere? Whether he was afraid of the Walther? And
then he does not show up. No balls, the man. He prefers to show himself in the protection of
other, positive thoughts. (AuS, Flashback, Pt. 5)

[Beim Versuch, die ohnehin schon erfolgreiche Walther in ihrem Kampf zu unterstützen,
personifiziere ich die störende Instanz zuerst als Störer, dann Wilhelm Störer. Ich versuche
ihn anzusprechen und notiere seine Reaktionen. Im Gegensatz zur Walther reagiert er so
gut wie gar nicht und macht gern das Gegenteil von dem, was ich will. Also fordere ich
ihn in ruhigeren Phasen auf, sich doch wieder einmal zu zeigen, und verhöhne ihn: Ob er
nicht mehr stören wolle oder könne? Ob er sich vor der Walther fürchte? Und dann zeigt
er sich nicht. Keine Eier, der Mann. Er zeigt sich am liebsten im Schutz anderer, positiver

The first-person narrator has a split personality and creates his own inter-tex-
tual universe. What happens is that the internal narrator undergoes a process of
externalization, as a result of which multiple storylines are interwoven. Besides
the experiential viewpoint of Wolfgang, the viewpoints of nonhuman objects
are foregrounded. This turn toward the nonhuman serves to show how objects
and other entities shape the construction of the self. That Wolfgang’s self is best
understood as a sort of inter-text becomes clear when the narrator announces
that he has found the world formula, that is, the key to his own text.

I have found the world formula, terrible, the world formula is a vicious circle, we revolve
forever in a loop, hell, and now the text, my text, the large text returns. But maybe it is a
literary text? Yes, of course, that is the rescue: I am in my own text, therefore, set pieces of
my other texts appear regularly, I can write it down, I have already written it down, it exists
in my head […]. (AuS, Flashback, Pt. 8)

[Ich habe die Weltformel gefunden, furchtbar, die Weltformel ist ein Zirkelschluß, wir
kreisen ewig in einer Schleife, Hölle, und jetzt kommt der Text schon wieder, mein Text,
der große Text. Aber vielleicht ist es ein literarischer Text? Ja, natürlich, das ist die Rettung:
Ich bin in meinem eigenen Text, deshalb tauchen auch dauernd Versatzstücke meiner
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   139

anderen Texte auf, ich kann ihn aufschreiben, ich habe ihn schon aufgeschrieben, er steht
in meinem Kopf […].]

The narrator’s self-fashioning as the Borgesian Man of the Book is a strong image
because it works not only as a mediating hinge between life and text—specifi-
cally the bringing of life into a text—but it also serves as a fruitful way to create
suspense. The Man of the Book is a God-like creature who knows the master nar-
rative. What Wolfgang means when he claims to have found the world formula is
that he knows how the action of the story will be resolved. He has the end in mind
because the narrative ends when his life ends. The problem with regard to the
reliability of the narrator in this context concerns the “impossibility of reaching a
‘correct’ reading” (Higonnet 1985, 103). This is due, firstly, to the epistemological
problem of gaining reliable knowledge about death and suicide and, secondly, to
the chicken-and-egg dilemma that concerns the question of whether the narrative
precedes the event (the suicide) or vice versa.
All formal elements in terms of the unreliability of the narration must be
interpreted as “sedimented content in their own right, as carrying ideological
messages of their own, distinct from the ostensible or manifest content […]”
(Jameson 1983 [1981], 99). In Herrndorf’s blog, unreliability does not only serve
as a critique of traditional autobiographies, in the context of which reliability is
the default option, but it also creates an interpretive void that is rooted in what
Nesbet (1991, 835) circumscribes as “an uncanny fulcrum between ‘meaningful’
life and ‘meaningless’ death.” This uncanny fulcrum shapes not only the narra-
tive situation, but also the temporal framework of Arbeit und Struktur, the con-
struction of space as well as the configuration of the human body.
Herrndorf’s blog is a temporal and spatial geography of death. The temporal
framework of the narrative is complexly structured: it is organized in such a way
that the past, present and future are woven into each other. The work on the blog
itself serves as an inducement to live in the present, but the uncanny specter of
suicide that haunts the author qua narrator makes him want to live in the past.

Supposedly, sentimentality grows with age, but this is nonsense. From the beginning my
gaze was fixed on the past. When in Garstedt the thatched roof house burned down, when
my mother taught me the alphabet, when I got crayons for my first day at school and when
I found the pheasant feathers in the aviary, I always thought back, and I always wanted
standstill, and almost every morning I hoped the beautiful twilight would be repeated
again. (AuS, Introduction)

[Angeblich wächst die Sentimentalität mit dem Alter, aber das ist Unsinn. Mein Blick
war von Anfang an auf die Vergangenheit gerichtet. Als in Garstedt das Strohdachhaus
abbrannte, als meine Mutter mir die Buchstaben erklärte, als ich Wachsmalstifte zur
140   Nora Berning

Einschulung bekam und als ich in der Voliere die Fasanenfedern fand, immer dachte ich
zurück, und immer wollte ich Stillstand, und fast jeden Morgen hoffte ich, die schöne Däm-
merung würde sich noch einmal wiederholen.]

Herrndorf published new blog entries on an almost daily basis. The entries are
interspersed with dream-like passages, extracts from his novel Tschick (2010),
scattered readings of other novels, medical reports and statistics. This mixture
of novelistic elements and empirical data conjures up the impression of a frag-
mented text that allows Herrndorf to convey the idea that suicide “does not fit
into conventional narrative” (K. Ryan 2004, 96). Whereas all entries in chapters
1–42 are preceded by a date and time, the flashbacks do not contain any tempo-
ral deictic markers. However, by carefully scrutinizing the photographs in Her-
rndorf’s notebook, which form an integral part of the retrospective sections, the
reader can extrapolate the dates of some flashbacks.
Chapters 1–8, together with the subsequent series of flashbacks, cover a
period of three weeks, that is, from mid-February 2010 when Herrndorf had his
first surgery until he was admitted to the psychiatric hospital on 8 March 2010. In
contrast to the thirty-four remaining chapters, the speed of the first third of the
story is quite fast. The narrative speed, to put it in Gérard Genette’s (1980 [1972])
terms, can be described as a summary: the narrator summarizes the main events
and creates the impression that he is still very much in control of his own life-nar-
rative. Only gradually does the speed of the narrative slow down and the entries
begin to read more and more like a farewell letter.

With the diagnosis, life goes on, life without hope does not. In the beginning, I could always
say to myself: You have still at least one year. A year is a long time. Even though I had
to ignore the physical and mental deterioration which should deteriorate further over the
anticipated 17 months still remaining. But after the greater part of the statistically expected
time is over, the view of the dwindling remainder is increasingly disturbing. Although I (in
a privately kept milkmaid-style bill) seem to have broken through after ten months without
relapse already to the right side of the bell curve. But the days are dwindling, and with them
hope. The work is getting continually heavier. The last few weeks frantically bolted together
chapters, the sense of futility overwhelms me. (AuS, Chap. 12)

[Mit der Diagnose leben geht, Leben ohne Hoffnung nicht. Am Anfang konnte ich mir
immer sagen: Ein Jahr hast du mindestens noch. Ein Jahr ist eine lange Zeit. Auch wenn ich
den körperlichen und geistigen Verfall, der von den avisierten 17 Monaten noch abgehen
sollte, dabei ausblenden mußte. Aber nachdem der größere Teil der statistisch erwartbaren
Zeit vorüber ist, ist der Blick auf den schwindenden Rest immer beunruhigender. Obwohl
ich mich (private Milchmädchenrechnung) nach zehn Monaten ohne Rezidiv bereits auf die
rechte Seite der Glockenkurve durchgeschlagen zu haben glaube. Aber die Tage schwinden
dahin, und mit ihnen die Hoffnung. Das Arbeiten wird immer schwerer. Die letzten Wochen
krampfhaft Kapitel zusammengeschraubt, das Gefühl der Sinnlosigkeit überrennt mich.]
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   141

The sense of futility is one of the factors that Cristina Monforte-Royo et al. (2011)
see as directly related to the WTHD. In Herrndorf’s case, psychological distress
is coupled with physical symptoms, psychiatric disorders, social factors such as
feeling like a burden to others and factors grouped under the label of psycho-ex-
istential suffering such as the loss of meaning of life. At several points in the
story, Wolfgang considers giving up the biographical quest and shutting down
the blog. The thought of stopping writing for good is particularly strong on those
days when he is essentially waiting for death to come. The hoped-for release does
not come, though. Rather, it seems as if time stands still and the narrator thinks
his whole world is expanding.
Wolfgang’s suicidal subjectivity has not only a huge impact on the order and
speed of the narrative but can also be seen as an aesthetic means, especially
when it comes to research. Research plays a significant role in the context of
Herrndorf’s blog, in general, and in terms of the configuration of narrative time,
more specifically. The narrator reads widely about glioblastoma. He googles, for
instance, the latest studies on residual tumor cells and consults Wikipedia to
interpret his symptoms. He also gains insights into his illness by reading relevant
journals such as the Annals of Neurology. Based on his research, Wolfgang calcu-
lates his day of death and uses the preterit (“overlooking the water, there where
I died” [“mit Blick aufs Wasser, dort, wo ich starb”]; AuS, Chap. 41) to refer to his
suicide as a completed action in the future.
Though Herrndorf does occasionally mention the sources of the studies that
he quotes from, scientific research is re-appropriated by the narrator for the
purpose of playgiarism, a neologism coined by the author Raymond Federman
which refers to the playful re-use of existing source material. In Herrndorf’s blog,
playgiarism takes place in an already barely regulated environment: the Inter-
net. Since the narrator uses the electronic environment to put forth a particu-
lar conceptualization of time, playgiarism is anything but a neutral or value-free
endeavor. Rather, it is an ideologically charged undertaking in Herrndorf’s blog
that is questionable, considering that the result of decontextualizing scientific
data is a loss of credibility in empirical research. As the following excerpt illus-
trates, the narrator’s goal is to deconstruct scientific research so as to shed light
on the pseudo objectivity of the hard sciences and create his own individual time-

The twelfth of August in my calendar is marked, grave stone shaped, my day of death, cal-
culated in the week after the surgery based on the first of Passig’s downloaded statistics,
seventeen point something months. The afternoon passes with a long beach walk in the
rain down to Sellin and swimming twice in 15 degree cold water. Gorgeous waves, gorgeous
everything. (AuS, Chap. 18)
142   Nora Berning

[Der zwölfte August in meinem Kalender ist eingekastet, grabsteinförmig, mein Todestag,
errechnet in der Woche nach der OP aufgrund der ersten von Passig runtergeladenen
Statistiken, siebzehn Komma irgendwas Monate. Der Nachmittag vergeht mit einem langen
Strandspaziergang im Regen nach Sellin runter und zweimaligem Baden im 15 Grad kalten
Wasser. Herrliche Wellen, herrlich alles.]

The alleged day of death is a day like any other. Hence, the narrator’s timeline of
his death diverges from the statistics. The narrator is disillusioned with scientific
research and starts to lean toward the other extreme, namely a romantic concep-
tion of the infinite and the relativity of all human values in the face of this infin-
ity. The irony in this context is that Wolfgang uses the mathematical concept of
infinity to deconstruct scientific studies that are primarily based on mathematical
principles or formulas. While this is not a problem per se, the fact that the narra-
tor instrumentalizes the notion of infinity in order to connote death and suicide
is definitely problematic because it may cause a copycat effect or what sociologist
David Phillips (1974) has dubbed the Werther effect.
Thus, the double-edged nature of online communities is augmented in the
suicide blog in the sense that the ethically sensitive issue of suicide is not only dif-
ficult to read and mourn but may also provoke impetuous and dangerous actions.
At the same time, the deregulated writing space of the Internet, and particu-
larly the blog, is a viable place for locating emotion related to death and suicide
because it is an open and less confrontational space than the mainstream media
or social networking sites such as Twitter, where people can comment instantly
on someone else’s state of mind. In Arbeit und Struktur, spaces are emotional-
ized and inextricably linked to the degenerative processes of advanced cancer
(Morris and Thomas 2007). The blog sheds light on space as a psycho-geograph-
ical category (Davidson et al. 2007), which points simultaneously to the experi-
ential dimension of space and the spatial dimension of experience (Berning et al.
2014b). The geography of his death that Wolfgang creates is deeply personal, yet
the fact that it is published online shows that (Wolfgang) Herrndorf’s actions are
motivated by a sense of sharing.

For some time already my empathy runs on a strange track. Previously, at some point I had
imagined that the nearness of death could possibly trigger hatred, hatred against the world,
envy of the survivors, perhaps even the desire to once again run amok and take along as
many as possible. In fact, I had even started a text in this sense. But the reverse is the case.
(AuS, Chap. 12)

[Seit geraumer Zeit schon läuft meine Empathie auf seltsamer Spur. Früher irgendwann
hatte ich mir mal vorgestellt, der nahe Tod würde möglicherweise Haß auslösen, Haß auf
die Welt, Neid auf die Überlebenden, vielleicht sogar den Wunsch, noch einmal Amok zu
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   143

laufen und möglichst viele mitzunehmen. Tatsächlich hatte ich mal einen Text in diesem
Sinne angefangen. Aber das Gegenteil ist der Fall.]

Wolfgang’s feeling of empathy is conspicuous because it is integrated into a

predominantly self-referential discourse. The writing space of the Internet runs
counter to an understanding of emotions as relational fluxes between people. On
the pragmatic level, it creates a pseudo-public sphere whereas on the semantic
level, the virtuality of socio-spatial life is “a matter of finding the right fit between
the medium and the form and the substance of the narrative content” (M.-L. Ryan
2004, 354). If one subscribes to the distinction between time-biased media and
space-biased media (Innis 1964), then the Internet can be conceptualized as a
space-biased medium because it obliterates space. If one adopts a deconstruc-
tionist perspective, one might even argue that Herrndorf subverts Innis’s distinc-
tion which goes back to Gotthold Ephraim Lessing by producing an emotional
geography that develops out of a personal, disjointed logic of mentally created
topologies that diminish the importance of both space and time.
This undoing of previously established truths takes place not only on the
level of form, but also on the level of the narrative content. In nonfictional illness
narratives, embodied emotions are typically connected to specific sites and con-
texts which serve as metaphors for caring/careless environments (e. g., home,
hospital, psychiatric ward). In Herrndorf’s blog, common stereotypes related to
space and, more concretely, to the contrast between home comfort and places
away from the patient’s home so typical of illness narratives is subverted through
the narrator’s use of irony.

The visit comes, the ward physician Dr. One brings concern to my enduring happy feeling.
Hypomania is the word. She would like to keep me here for a bit longer, and that is exactly
what I would like, too. I name my reasons, spacious rooms here vs. one-room hole at home,
fantastic food, rest, concentrated work and a garden practically just for me, adding that it is
like a holiday for me, that I consider it for the same reason as a waste of taxpayers’ money,
and with this overall assessment I could apparently confirm her diagnosis of hypomania.
(AuS, Chap. 2)

[Die Visite kommt, der Stationsärztin Dr. Eins macht mein haltbar fröhlicher Affekt Sorgen.
Hypomanie ist das Wort. Sie würde mich gern länger hierbehalten, und das ist genau das,
was ich mir auch wünsche. Ich nenne meine Gründe, Räumlichkeiten hier vs. Ein-Zimmer-
Loch zu Hause, fantastisches Essen, Ruhe, konzentriertes Arbeiten und ein Garten praktisch
für mich allein; füge hinzu, daß es wie Urlaub für mich sei, ich es aus demselben Grund für
Verschwendung von Steuergeldern hielte, und habe mit dieser Gesamteinschätzung ihre
Diagnose der Hypomanie offenbar befestigen können.]
144   Nora Berning

The narrator pretends to feel very much at home in the psychiatric ward where
he can work on his manuscripts without being disturbed. The excerpt can there-
fore be seen as indicative of the impact that embodied, emotional and situational
factors have on a person’s preference for place of death (Morris and Thomas
2007). It shows in what ways the experience of space affects the narrator’s sub-
jectivity and his interpersonal relationships, which are framed in such a way as
to foreground Wolfgang’s perception of people and things. In light of this, Herrn-
dorf’s blog can be seen not only as a medium of emotional geographies, but also
as an attempt to jump on the bandwagon of the neurological turn (Lovink 2010),
which refers to the deep-seated interest of the humanities and the media in the
workings of the human brain, (fictional and factual) minds and consciousness.
Arbeit und Struktur, a stream-of-consciousness blog, is a particularly fruitful
medium for studying minds, because it is able to shed light on suicidal subjec-
tivity in a way that no other medium can. The medium shares with the subject of
suicide the ability to function as a telling silence (Higonnet 2000). Like the topic
of suicide, the blog is abundantly open and transgressive, a medium of liminality
in the sense that “[i]t generates a multiplicity of interpretations and silences” (K.
Ryan 2004, 96). Herrndorf uses the writing space of the Internet as a means to
exhaust the therapeutic power of writing.

This time, simple volition is not sufficient, and I have to install a very vividly envisaged
Walther PPK in my head to shoot any unpleasant evolving thought: Bang, bang. Two
bullets, and I think of something else. […] After a few hours, maybe it is also a day, I notice
that it is cracking in my head, though I have not pressed the trigger. The Walther becomes
independent. That is welcome to me, it does only its duty. It cracks and bangs in my head
without my help, and the thought of death barely dips below the surface, while I am sitting
unfazed by anything at my computer and work. (AuS, Flashback, Pt. 5)

[Diesmal reicht eine einfache Willensentscheidung nicht aus, und ich muß eine sehr plas-
tisch vorgestellte Walther PPK in meinem Kopf installieren, um jeden unangenehmen
aufkommenden Gedanken zu erschießen: Peng, peng. Zwei Kugeln, und ich denke an
etwas anderes. […] Nach einigen Stunden, vielleicht ist es auch ein Tag, bemerke ich, daß
es in meinem Kopf knallt, ohne daß ich den Abzug gedrückt habe. Die Walther verselbstän-
digt sich. Das ist mir willkommen, sie tut nur ihre Pflicht. Es klickt und knallt in meinem
Kopf ohne mein Zutun, und der Todesgedanke taucht kaum noch bis unter die Oberfläche,
während ich von allem unbeeindruckt am Computer sitze und arbeite.]

While the narrator is writing his tale of death, the Walther PPK is fighting
against his tumor. In a way, the semi-automatic pistol serves the same function
as weapons in a computer game: it is a tool against unsolicited visitors. On a
photograph that shows the author’s Moleskine, which Herrndorf holds out to a
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   145

webcam, the audience literally sees the forces at play in the author’s head: the
Walther PPK versus Wilhelm Störer (the tumor) and Wolfgang (figure 2).

Figure 2: Walther, Wilhelm and Wolfgang (© 2010–2013 Wolfgang Herrndorf)

Underneath the drawing of the pistol can be read “the inseparable three” and
Barack Obama’s famous slogan “yes, we can!” The photograph is particularly
telling insofar as it speaks to the different spaces that the author (Herrndorf) and
the narrator (Wolfgang) inhabit. The narrator is heavily implicated in the emo-
tional geography that the author constructs, fighting his negative thoughts and
battling death, while the author is sitting at his desk and creates increasingly
grotesque scenarios and storyworlds.
In Arbeit und Struktur, ways of worldmaking (Goodman 1992 [1978]; Herman
2009) that go into literary world-construction are shaped by the increasingly
fragmented self of the author qua narrator and the feeling of being disconnected
from one’s own body. The narrator gradually loses control over his body—the
body referring in this context both to the body as a human organism and to the
body as text. The former is a site of intense emotions and struggle which exudes
pain and suffering. The reader becomes a witness of how the human “body turns
against itself, becoming murderer and victim simultaneously” (K. Ryan 2004, 95).
146   Nora Berning

The body as text is closely linked to the representational dynamics of the human
body. The textual as well as the physical body serve as a means of inscribing the
loss of identity.

One is suddenly no longer part of something that one was hitherto accustomed to perceiv-
ing as the self, as I, so questionable one has always found the synthetic construction of the
self on an intellectual level (but purely on an everyday basis this I surely existed), and then
it dissolves into the impersonal act of the average crazed representatives of this kind, very
useful and at the same time moronic, adapted from the process of evolution to the hard-
ships of the world. (AuS, Flashback, Pt. 10)

[Man steckt auf einmal nicht mehr drin in etwas, was man bis dahin als Selbst wahrzune-
hmen gewohnt war, als Ich, so fragwürdig man die synthetische Konstruktion des Ichs auf
einer intellektuellen Ebene schon immer empfunden hat (aber rein alltagstechnisch war
dieses Ich doch sicher vorhanden), und dann löst es sich auf in das unpersönliche Agieren
eines vom Evolutionsprozeß sehr sinnvoll und zugleich schwachsinnig an die Härten der
Welt angepaßten durchschnittlich durchgedrehten Vertreters der Art.]

What Wolfgang describes is the feeling as though one is not completely occu-
pying the body. His allusion to the Cartesian concept of the division of mind
and body can be conceived as a meta-reference, that is, a narrative strategy that
the author uses in order to convey the idea that the narrator’s loss of a sense of
direction and time which, as a recurrent theme, stands for the author’s anxiety
of getting lost in his own textual universe. The fear of losing a feel for the text is
mediated through metaphors derived from bodily experience. The text, to put it in
the words of George Lakoff and Mark Johnson (1999), is “philosophy in the flesh.”
It serves as a metaphor for thinking about the complex relation between life and
narrative, corporeality and culture, and moral concepts like freedom and deter-
minism. In short, the body as text invites reflection “on the nature of humanity
itself” (White 1980, 5).
In Herrndorf’s blog, the body serves as a map, for it renders intelligible or
even systematizes knowledge and experience. Its function is twofold: on the one
hand, the body serves “as part of a strategy of textual representation” (Punday
2003, ix) and as a relevant point of reference in terms of new ways of mapping
suicide. On the other hand, the body is a key element in challenging received
ideas that obfuscate critical knowledge about suicide. Furthermore, the body is
part of a personal ordering of a universe.

My body has exactly the same temperature and consistency as its surroundings, just like
the bedlinen, I am a piece of linen between other pieces of linen, by a strange coincidence I
became conscious, and I hope that it will always remain so. That is my first memory of this
world. (AuS, Introduction)
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   147

[Mein Körper hat genau die gleiche Temperatur und Konsistenz wie seine Umgebung, wie
die Bettwäsche, ich bin ein Stück Bettwäsche zwischen anderen Stücken Bettwäsche, durch
einen sonderbaren Zufall zu Bewußtsein gekommen, und ich wünsche mir, daß es immer so
bleibt. Das ist meine erste Erinnerung an diese Welt.]

The childhood memory of a moment when Wolfgang’s body and its surroundings
had the same temperature reflects a nostalgic feeling or yearning for being inte-
grated and connected to the material world. Since Herrndorf can neither restore
his health nor rehabilitate those feelings of connectivity and human relationship,
the blog represents a kind of ersatz for being connected to the world. Literary
techniques such as stream-of-consciousness are part of the author’s toolbox,
which is certainly shaped by famous literary examples such as James Joyce’s
novel Ulysses (1922) but that is shaped no less by the formal characteristics of the
digital medium. It is in the writing space of the Internet that the meta-level and
the object-level collapse into each other, as a consequence of which the author
and the narrator, text and body, become almost indiscernible. Arbeit und Struk-
tur poignantly illustrates the “fictionality of the real” (Zipfel 2009, 305), that is,
the impossibility of a clear-cut distinction between fact and fiction as well as the
need to acknowledge hybridization as a core catalyst for generic evolution, inno-
vation and change (Nünning and Schwanecke 2013).
Moreover, based on my analysis of the ethics and aesthetics of Herrndorf’s
suicide blog, it has hopefully become clear that Arbeit und Struktur fulfills two
additional cultural functions. First, Herrndorf’s use of the blog as a site of medial
self-fashioning highlights the cultural trend toward intermedial and multimodal
constructions of the self. Second, it underscores the need to delink medialization
from the rise of electronic mass media in the 1960s: the more recent developments
in the realm of digital media are an invitation to take into consideration medial-
ization as another core catalyst for genre evolution and change (Nünning and
Rupp 2013). To sum up, Herrndorf questions in his blog the therapeutic power of
writing and renders intelligible the generic conventions, narrative schemata and
building blocks of fictional end-of-life-stories (suicide novels) and their inability
to produce genuine knowledge about suicide.

3 Conclusion
In the present article I have argued that the analytical framework of CEN is best
understood as an emerging vector of narrative theory that reflects simultaneously
narratology’s continued trend toward diversification in the sense of increasing
interdisciplinarity as well as a phase of consolidation insofar as analysis of the
148   Nora Berning

forms and functions of narrative continues to be at the forefront of narratological

research. Drawing on the example of the autobiographical end-of-life story and,
more specifically, on its manifestation on the Internet in the form of a blog, I
have demonstrated that, by merging classical and postclassical concepts of nar-
rative theory, CEN is a fruitful framework for studying the thanatopoetics and the
negotiation of values surrounding the phenomenon of suicide. Suicide blogs are
an integral element of postmodern literature and culture. They warrant attention
because they foster new and innovative ways of knowledge production regarding
the sensitive issue of suicide, and also because they serve as indicators of new
forms of life (Wittgenstein 1968 [1953]) and nonfictional narratives of illness.
In A Poetics of Postmodernism (2000), Linda Hutcheon argues that postmod-
ern society is interested in margins rather than in centers. One need be neither a
technological enthusiast nor a media determinist to think that suicide blogs will
contribute to contemporary society’s growing fascination with margins and shape
our horizon of expectations (sensu Jauss) with regard to confessional genres and/
or modes of storytelling (Gilmore 2001). Against this background, narratological
analyses of sister genres of the autobiographical end-of-life story like drug auto-
biographies (Zieger 2007), for example, are but one of many fruitful avenues for
further research.
Furthermore, there are still many untrodden paths in the interdisciplinary
field of medical humanities that this article could not explore in depth. Although
I have briefly touched on the neurological turn, which is noticeable not only in
the humanities and in Internet criticism but also in the natural sciences, it is
important to underline that productive interfaces between the humanities, the
social sciences and the arts extend far beyond the ‘neuro turn’. With my contri-
bution, I hope to have shown that factual illness narratives provide a good start-
ing point for delving deeper into the important role of narratives in terms of con-
structing forms of medical knowledge as well as its application in research and
public health, on the one hand, and the role of narratives in generating knowl-
edge in the social sciences regarding mental health, social relations and human
development, on the other hand. As new paradigms of narratology emerge for
showing that the gap between living and telling is a lot thinner than it is some-
times supposed (see Raphaël Baroni’s contribution to this volume), it is likely
that the interdisciplinary scope of narrative theory, in general, and rhetorical or
ethical narratologies that conceive of narrative as a “purposive communicative
act” (Phelan 2007, 203), more concretely, will become much more relevant in the
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   149

Works Cited
Basseler, Michael, Ansgar Nünning and Christine Schwanecke. 2013. “The Cultural Dynamics of
Generic Change: Surveying Kinds and Problems of Literary History and Accounting for the
Development of Genres.” In Basseler, Nünning and Schwaneke, eds., 1–40.
Basseler, Michael, Ansgar Nünning and Christine Schwanecke, eds. 2013. The Cultural
Dynamics of Generic Change in Contemporary Fiction: Theoretical Frameworks, Genres and
Model Interpretations. Trier: WVT.
Baumbach, Sibylle, Herbert Grabes and Ansgar Nünning, eds. 2009. Literature and Values:
Literature as a Medium for Representing, Disseminating and Constructing Norms and
Values. Trier: WVT.
Berning, Nora. 2011. Narrative Means to Journalistic Ends: A Narratological Analysis of Selected
Journalistic Reportages. Wiesbaden: VS Verlag für Sozialwissenschaften.
Berning, Nora. 2013. Towards a Critical Ethical Narratology: Analyzing Value Construction in
Literary Non-Fiction across Media. Trier: WVT.
Berning, Nora, Ansgar Nünning and Christine Schwanecke, eds. 2014a. Reframing Concepts in
Literary and Cultural Studies: Theorizing and Analyzing Conceptual Transfers. Trier: WVT.
Berning, Nora, Philipp Schulte and Christine Schwanecke, eds. 2014b. Experiencing Space—
Spacing Experience: Concepts, Practices, and Materialities. Trier: WVT.
Bronfen, Elisabeth. 1992. Over Her Dead Body. Death, Femininity and the Aesthetic.
Manchester: Manchester University Press.
Brooks, Peter. 1985. Reading for the Plot: Design and Intention in Narrative. New York: Knopf.
Charon, Rita. 2008. Narrative Medicine: Honoring the Stories of Illness. Oxford: Oxford
University Press.
Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in
Fiction. Princeton, NJ: Princeton University Press.
Davidson, Joyce, Liz Bondi and Mick Smith, eds. 2007. Emotional Geographies. Aldershot, GB:
Derrida, Jacques. 1987 [1980]. The Postcard: From Socrates to Freud and Beyond. Translated by
Alan Bass. Chicago: The University of Chicago Press.
De Vigan, Delphine. 2013. Nothing Holds Back the Night. London: Bloomsbury.
Eagleton, Terry. 2012. The Event of Literature. New Haven: Yale University Press.
Frey, James. 2004. A Million Little Pieces. New York: Random House.
Genette, Gérard. 1980 [1972]. Narrative Discourse: An Essay in Method. Translated by Jane E.
Lewin. Foreword by Jonathan Culler. Ithaca, NY: Cornell University Press.
Gibson, Andrew. 1996. Towards a Postmodern Theory of Narrative. Edinburgh: Edinburgh
University Press.
Gilmore, Leigh. 2001. The Limits of Autobiography: Trauma and Testimony. Ithaca, NY: Cornell
University Press.
Giltrow, Janet, and Dieter Stein, eds. 2009. Genres in the Internet: Issues in the Theory of Genre.
Amsterdam: John Benjamins.
Goodman, Nelson. 1992 [1978]. Ways of Worldmaking. Indianapolis: Hackett.
Hayles, N. Katherine. 2008. Electronic Literature: New Horizons for the Literary. Notre Dame, IN:
University of Notre Dame Press.
Helms, Gabriele. 2003. Challenging Canada: Dialogism and Narrative Techniques in Canadian
Novels. Montreal: Queen’s University Press.
150   Nora Berning

Herman, David. 1997. “Scripts, Sequences, and Stories: Elements of a Postclassical

Narratology.” PMLA 112 (5): 1046–1059.
Herman, David. 1999. “Introduction: Narratologies.” In Narratologies: New Perspectives of
Narrative Analysis, edited by D. H., 1–30. Columbus: The Ohio State University Press.
Herman, David. 2009. Basic Elements of Narrative. Oxford: Wiley-Blackwell.
Herring, Susan C., Lois Ann Scheidt, Elijah Wright and Sabrina Bonus. 2004. Bridging
the Gap: A Genre Analysis of Weblogs. In Proceedings of the 37th Annual Hawaii
International Conference on System Sciences.
genreanalysisofweblogs.pdf (Accessed 22 November 2015)
Herring, Susan C., Lois Ann Scheidt, Elijah Wright and Sabrina Bonus. 2005. “Weblogs as a
Bridging Genre.” Information Technology & People 18 (2): 142–171.
Herrndorf, Wolfgang. 2010–2013. Arbeit und Struktur.
(Accessed 22 November 2015)
Higonnet, Margaret. 1985. “Suicide: Representations of the Feminine in the Nineteenth
Century.” Poetics Today 6 (1/2): 103–118.
Higonnet, Margaret. 2000. “Frames of Female Suicide.” Studies in the Novel 32 (2): 229–242.
Hook, Sidney. 1927. “The Ethics of Suicide.” International Journal of Ethics 37 (2): 173–188.
Hutcheon, Linda. 2000. A Poetics of Postmodernism: History, Theory, Fiction. New York:
Hutchins, Michael. 2012. “Intent to Mislead: W. G. Sebald’s Notion of Authenticity.” In
Authentisches Erzählen: Produktion, Narration, Rezeption, edited by Antonius Weixler,
79–94. Berlin and Boston: De Gruyter.
Hydén, Lars-Christer. 2005. “Medicine and Narratives.” In Routledge Encyclopedia of Narrative
Theory, edited by David Herman, Manfred Jahn and Marie-Laure Ryan, 293–297. London
and New York: Routledge.
Innis, Harold A. 1964. The Bias of Communication. Toronto: University of Toronto Press.
Jameson, Fredric. 1983 [1981]. The Political Unconscious: Narrative as a Socially Symbolic Act.
London: Methuen.
Joyce, James. 2002 [1922]. Ulysses. New York: Random House.
Joyce, Michael. 1987–90. afternoon, a story. Cambridge: Eastgate Systems.
Kapp, Volker. 1987. “Von der Autobiographie zum Tagebuch (Rousseau—Constant).” In
Selbstthematisierung und Selbstzeugnis: Bekenntnis und Geständnis, edited by Alois
Hahn and Volker Kapp, 297–310. Frankfurt a. M.: Suhrkamp.
Knapp, Lore. 2012. “Christoph Schlingensiefs Blog: Multimediale Autofiktion im Künstlerblog.”
In Nünning, Rupp, Hagelmoser and Meyer, eds., 117–132.
Korthals Altes, Liesbeth. 2014. Ethos and Narrative Interpretation. The Negotiation of Values in
Fiction. Lincoln and London: University of Nebraska Press.
Lakoff, George, and Mark Johnson. 1999. Philosophy in the Flesh: The Embodied Mind and its
Challenge to Western Thought. Chicago: The University of Chicago Press.
Lejeune, Philippe. 1989 [1973]. “The Autobiographical Pact.” In On Autobiography, edited
by Paul John Eakin, translated by Katherine Leary, 3–30. Minneapolis: University of
Minnesota Press.
Lenhart, Amanda, and Susannah Fox. 2006. Bloggers: A Portrait of the Internet’s
New Storytellers.
PIP%20Bloggers%20Report%20July%2019%202006.pdf (Accessed 22 November 2015)
 Critical Ethical Narratology as an Emerging Vector of Narrative Theory   151

Lovink, Geert. 2010. Mybrain.Net: The Colonization of Real-Time and Other Trends in Web 2.0. (Accessed 22 November
Mandle, Carol Lynn. 1984. “Suicide—A Human Problem.” Educational Horizons 62 (4): 119–127.
McNeill, Laurie. 2009. “Brave New Genre, or Generic Colonialism? Debates over Ancestry in
Internet Diaries.” In Giltrow and Stein, eds., 143–161.
Miller, Carolyn R., and Dawn Shepherd. 2004. Blogging as Social Action: A Genre Analysis
of the Weblog.
(Accessed 22 November 2015)
Miller, Carolyn R., and Dawn Shepherd. 2009. “Questions for Genre Theory from the
Blogosphere.” In Giltrow and Stein, eds., 263–290.
Monforte-Royo, Cristina, Christian Villavicencio-Chávez, Joaquín Tomás-Sábado and Albert
Balaguer. 2011. “The Wish to Hasten Death: A Review of Clinical Studies.” Psycho-
Oncology 20 (8): 795–804.
Morris, Sara M., and Carol Thomas. 2007. “Placing the Dying Body: Emotional, Situation and
Embodied Factors in Preferences for Place of Final Care and Death in Cancer.” In Emotional
Geographies, edited by Joyce Davidson, Liz Bondi and Mick Smith, 19–32. Aldershot, GB:
Nesbet, Anne. 1991. “Suicide as Literary Fact in the 1920s.” Slavic Review 50 (4): 827–835.
Newton, Adam Zachary. 1995. Narrative Ethics. Cambridge, MA: Harvard University Press.
Nünning, Ansgar. 2003. “Narratology or Narratologies? What is Narratology?” In Questions and
Answers Regarding the Status of a Theory, edited by Tom Kindt and Hans-Harald Müller,
239–275. Berlin: Walter de Gruyter.
Nünning, Ansgar. 2013. “Semantisierung literarischer Formen.” In Metzler Lexikon Literatur-
und Kulturtheorie, edited by A. N., 684–685. Stuttgart: Metzler.
Nünning, Ansgar, and Jan Rupp. 2012. “‘The Internet’s New Storytellers’: Merkmale, Typologien
und Funktionen narrativer Genres im Internet aus gattungstheoretischer, narratologischer
und medienkulturwissenschaftlicher Sicht.” In Nünning, Rupp, Hagelmoser and Meyer,
eds., 3–50.
Nünning, Ansgar, and Jan Rupp. 2013. “Media and Medialization as Catalysts for Genre
Development: Theoretical Frameworks, Analytical Concepts and a Selective Overview
of Varieties of Intermedial Narration in British Fiction.” In Basseler, Nünning and
Schwanecke, eds., 201–234.
Nünning, Ansgar, Jan Rupp, Rebecca Hagelmoser and Jonas Ivo Meyer, eds. 2012. Narrative
Genres im Internet: Theoretische Bezugsrahmen, Mediengattungstypologie und
Funktionen. Trier: WVT.
Nünning, Ansgar, and Christine Schwanecke. 2013. “Crossing Generic Borders—Blurring
Generic Boundaries: Hybridization as a Catalyst for Generic Change and for the
Transformation of Systems of Genres.” In Basseler, Nünning and Schwanecke, eds.,
Nünning, Vera, and Ansgar Nünning. 2002. “Produktive Grenzüberschreitungen:
Transgenerische, intermediale und interdisziplinäre Ansätze in der Erzähltheorie.” In
Erzähltheorie transgenerisch, intermedial, interdisziplinär, edited by V. N. and A. N., 1–22.
Trier: WVT.
Phelan, James. 2007. Experiencing Fiction: Judgments, Progressions, and the Rhetorical Theory
of Narrative. Columbus: The Ohio State University Press.
152   Nora Berning

Phillips, David. 1974. “The Influence of Suggestion on Suicide: Substantive and Theoretical
Implications of the Werther Effect.” American Sociological Review 39 (3): 340–354.
Punday, Daniel. 2003. Narrative Bodies: Toward a Corporeal Narratology. New York: Macmillan.
Puschmann, Cornelius. 2012. “Technisierte Erzählungen? Blogs und die Rolle der Zeitlichkeit im
Web 2.0.” In Nünning, Rupp, Hagelmoser and Meyer, eds., 93–114.
Ryan, Katy. 2004. “Falling in Public: Larsen’s Passing, McCarthy’s The Group, and Baldwin’s
Another Country.” Studies in the Novel 36 (1): 95–119.
Ryan, Marie-Laure. 2004. “Will New Media Produce New Narratives?” In Narrative across Media:
The Languages of Storytelling, edited by M.-L. R., 337–359. Lincoln and London: University
of Nebraska Press.
Scheidt, Lois Ann. 2006. “Adolescent Diary Weblogs and the Unseen Audience.” In Digital
Generations: Children, Young People, and the New Media, edited by David Buckingham
and Rebekah Willett, 193–210. Aylesbeare, GB: Mallory International.
Schlingensief, Christoph. 2008–2010. Schlingenblog.
(Accessed 22 November 2015)
Shankar, Ravi. 2010. “Dispersal and Renown: An Investigation of Blogs, Listservs and Online
Journals.” In Beyond the Screen: Transformations of Literary Structures, Interfaces and
Genres, edited by Jörgen Schäfer and Peter Gendolla, 521–542. Bielefeld: transcript.
Skelton, John. 2003. “Death and Dying in Literature.” Advances in Psychiatric Treatment 9:
Tophinke, Doris. 2009. “Wirklichkeitserzählungen im Internet.” In Wirklichkeitserzählungen.
Felder, Formen und Funktionen nicht-literarischen Erzählens, edited by Christian Klein and
Matías Martínez, 245–274. Stuttgart: Metzler.
Walker, Jill. 2003. Weblog Definition.
of_weblog_definition.html (Accessed 22 November 2015)
White, Hayden. 1980. “The Value of Narrativity in the Representation of Reality.” Critical Inquiry
7 (1): 5–27.
Wittgenstein, Ludwig. 1968 [1953]. Philosophical Investigations. Oxford: Blackwell.
Zenner, Roman. 2005. Hypertextual Fiction on the Internet: A Structural and Narratological
Analysis. Doctoral Dissertation submitted to the Rheinisch-Westfälische Technische
Hochschule Aachen.
Zerweck, Bruno. 2001. “Historicizing Unreliable Narration: Unreliability and Cultural Discourse
in Narrative Fiction.” Style 35 (1): 151–178.
Zieger, Susan. 2007. “Pioneers of Inner Space: Drug Autobiography and Manifest Destiny.”
PMLA 122 (5): 1531–1547.
Zipfel, Frank. 2009. “Autofiktion: Zwischen den Grenzen von Faktualität, Fiktionalität und
Literarität?” In Grenzen der Literatur, edited by Simone Winkel, 285–314. Berlin: De
Manja Kürschner (Kiel)
The Fictionalization of History in
Metahistoriographic Fiction after the
Constructivist Challenge

1 Historical Fiction—Quo Vadis

In 2004, Margaret Drabble’s The Red Queen caused some confusion and surprise
among readers and critics. The novel imitates historiography, since it contains
many of the ingredients that, according to renowned historian Perez Zagorin, are
required to create a good historical narrative:
A history, unlike a novel in the realist genre, consists not only of specimens
of narrative but of arguments, footnotes, of documentation and justification,
acknowledgments of what isn’t known, discussions and evaluations of sources
and evidence, and critiques of the view of other scholars (Zagorin 1999, 12).1
Drabble’s fiction combines historical facts about Lady Hyegyǒng, an eight-
eenth-century Korean Crown Princess, with the illusion of the Princess haunting
twenty-first century Britain as a ghost. Although it is clear that what we read is
fiction, the historical bits and pieces, as well as the procedure of writing history,
are taken seriously. The fictionalized version of the historical Princess controls
two modern protagonists who come into possession of Lady Hyegyǒng’s memoir,
which, in turn, relates to a historical document of immense importance in Korean
history. Despite its autobiographical form, the memoir and the information con-
veyed in it are part of the commonly accepted historical knowledge about the
Korean court of the eighteenth century, and the official Korean court chronicles
confirm it, which is why the memoir is categorized as solid historical evidence for
the events at the Korean court. By explicitly commenting on the origins and the
genesis of this memoir, the novel reflects upon the methods of historical research
and writing. It points out that historical knowledge about the Crown Princess is
treated as a fact on an intrafictional as well as on an extrafictional level, which is
quite uncommon for a postmodernist historical novel.

1 Note that in the 1990s, Zagorin entered into a heated debate with Keith Jenkins and Frank R.
Ankersmit, who represent a postmodernist textualist approach to the writing of history and who
challenge the supremacy of historical narrative over fiction.

DOI 10.1515/9783110555158-008
154   Manja Kürschner

Even if The Red Queen explicitly emphasizes the process of fictionalizing the
memoir, it does not seem to mix fact and imagination; instead, it marks factual
information as such. The inscribed author in Drabble’s fiction, also named Mar-
garet Drabble, quotes, for example, Korean Kisaeng poetry and Lady Sŏnhŭi’s
epitaph in one of the embedded narratives; she even gives references for these
quotations in her acknowledgements. In her ‘Note on Sources’, Drabble assesses
the different editions and translations of the Memoirs of Lady Hyegyŏng. The
novel concludes with bibliographic information about these editions and adds
further book recommendations for research regarding the historical background
of the period. David Jays, book reviewer for the Observer, calls Drabble’s attempt
to fictionalize history a failure because,

[l]ike a self-appointed anthropologist, she diligently explains culture and customs, and her
wintry reflections suck the blood out of court life. In chilly posterity, the princess fashions
her melodramatic experiences into a stilted seminar. […] The past ceases to be strange or
beautiful and subsides under a dust of explication.2 (Jays 2004)

Jays blames the author for having written a novel that is too erudite and that lacks
vividness, being “long on detail, but short on imagination.” Having anticipated
an aesthetic illusion, which might allow for a more emotional approach to past
experience, The Red Queen does not meet Jays’s expectations. In a review for the
San Diego Union-Tribune, Bart Thurber, expresses his astonishment about the
fact that Drabble places the focus on meticulously researched historical facts and
sources instead of playfully trying to blur the lines between fact and fiction:

Now, given that Drabble is one accomplished lady, the author of 14 or 15 prior novels,
several of which wax seriously postmodernistic, my first thought was that she was going
to Nabokov us to death with this, teasing us with nonexistent or endlessly self-referential
references, shading us into the ways we construct and are constructed by our reading, but,
well—no. I checked. Her sources are real. This novel is not Despair. Her academics, her intel-
lectuals, are all real people and her translations are genuine, all three of them. (Thurber

In comparison to her postmodernist case studies of female protagonists in the

1960s and 1970s, Drabble’s prodigious output continues with her sixteenth novel,

2 Notwithstanding this criticism, the effects of this extensive heteroreferentiality are also widely
praised in reviews of The Red Queen, which point out that the diary of the Princess “is especially
interesting for the light it casts on a way of life almost unknown to contemporary westerners,
and for this the novel is both important and fascinating.” See Mary Whipple’s review “Margaret
Drabble ‘The Red Queen’” (2004).
 The Fictionalization of History in Metahistoriographic Fiction   155

which Thurber aptly summarizes as being “seriously, even overwhelmingly,

scholarly.” These reactions of surprise and disapproval indicate that Drabble’s
way of writing about history in the realm of fiction deviates not only from the tra-
ditional historical novel, but also from the well-established postmodernist model
of historiographic metafiction, as defined by Linda Hutcheon (1988).3 Monika
Fludernik summarizes the main characteristics of the first prototypical examples
of historiographic metafiction:

In the post-modern world, it is argued, one’s worst fantasies are becoming true; in fact,
actual events exceed fictional scenarios in their grotesqueness, paradoxicality and incom-
prehensibility. As Raven Quickskill says in Ishmael Reed’s Flight to Canada, “Who is to say
what is fact and what is fiction?” It is this situation which critics have postulated to be at the
root of historiographic metafiction, and which is argued to reflect the climate of the Ameri-
can 60s and 70s. It is therefore no coincidence that the great historiographic metafictionists
are writing precisely during that period, producing representations of a chaotic world such
as those found in Heller’s Catch-22, Vonnegut’s Slaughterhouse-Five, Pynchon’ s Gravity’s
Rainbow, or in the work of Hawkes and DeLillo. (Fludernik 1994, 84)

In order to account for the otherness of Drabble’s fiction, I will introduce and
examine a subgenre of historiographic metafiction that I call post-constructivist
metahistoriographic fiction.4 Fludernik has already predicted the advent of a new
form of self-referential historical novel as an upcoming trend in the mid-1990s:

3 With the advent of the narrative and cultural turn in historiography and literary studies in the
1960s, metahistoriographic and metafictional commentaries about the nature and function of
historiography occur in large number in both disciplines. While representatives of narrative his-
toriography such as Arthur C. Danto, Hayden White or Frank R. Ankersmit demonstrate the ways
in which historians apply narrative techniques similar to those employed by novelists in order to
create a coherent story of the past, historiographic metafiction is characterized as a genre that,
according to Linda Hutcheon (1988, 48), merges metahistoriographic and metafictional com-
ments with storytelling. Hutcheon postulates the rise of historiographic metafiction in the 1960s
and 1970s, listing novels that “distrust history’s ability and will to convey ‘truth’” and that are no
longer interested in “recounting the facts.”
4 In analogy to Ansgar Nünning (1995, 284), who criticizes the concept ‘historiographic metafic-
tion’ for its misleading focus on metafiction instead of metahistoriography, I suggest a different
term for the subgenre I investigate in this paper. Aware of the fact that the umbrella term ‘histo-
riographic metafiction’ has already become canonical, I consider ‘post-constructivist metahis-
toriographic fiction’ a more appropriate term, but do not suggest replacing the former concept.
Rather than proposing to use Nünning’s term ‘fictional metahistoriography’ as a substitute for
the entire genre of historiographic metafiction, I think it is necessary to focus on a subgenre of
historiographic metafiction whose approach of fictionalizing history has changed. Instead of
displaying either a positivist or a radical constructivist attitude towards the past, post-construc-
tivist metahistoriographic fiction does not represent history as true or false, but as more or less
156   Manja Kürschner

A new, more serious, mode of historiographic metafiction seems to be hatching, one that is
less playful, more specifically concerned with ‘history’ (in different ways) and less simplis-
tically and dichotomously mythological than most of the historiographic metafictions of the
1960s and 1970s. Should one therefore, as I have suggested, desist from calling such texts
historiographic metafictions and talk instead of ‘the new historical novel’? That, I trust,
history will tell. (Fludernik 1994, 101)

Similar to Fludernik’s approach, I do not aim to provide a fixed typology of the

many different varieties of historiographic metafiction. I rather suggest having a
closer look at one specific group that can be differentiated from what we consider
the canonical examples of the genre.5 Whereas historiographic metafiction high-
lights the impossibility of writing reliable history, post-constructivist metahisto-
riographic fiction written after 2000 refutes radical constructivist theories that
postulate an all-encompassing unreliability of historical writing and storytelling.
Taking up the constructivist challenge, this new form of metahistoriographic
fiction presents history as a product of an agreement: a specific historical account
is treated as valid as soon as a scientific community agrees on it and as long as
it is not proven false by a competing explanation. Novels such as The Red Queen
hence reflect the cultural discourses of the last two decades, which foreground
the limits of contemporary historiography, but also start re-emphasizing its

plausible: positivist claims that history reconstructs the ‘real’ past are replaced by the belief that
history offers plausible hypotheses about what might have happened in the past.
5 In literary studies, novels such as John Fowles’s The French Lieutenant’s Woman (1969), Gra-
ham Swift’s Waterland (1983) and Julian Barnes’s Flaubert’s Parrot (1984) are mentioned as pro-
totypical British texts that exemplify Hutcheon’s genre classifications. Prototypical examples of
American historiographic metafiction are, for instance, E. L. Doctorow’s Ragtime (1974) or Thom-
as Pynchon’s Gravity’s Rainbow (1973) and Mason & Dixon (1997). In these novels, history is pre-
sented as “yarn” or “reality-obscuring drama” (Swift 1983, 53 and 34).
6 It echoes a pragmatic view on writing history. According to historian Dominick LaCapra (1983,
38), historical inquiry is not meant to end all arguments and explanations about what happened
in the past, but to make these explanations “as informed, vital, and undogmatically open to
counterargument as possible.” Additionally, historian Richard Evans (1997, 79) points out that
we do not have to question every single detail about the past, since there are at least some histor-
ical facts, such as the height of a tower, which are beyond doubt. “Only if new evidence is found
to amend or cast doubt on the historian’s account of a fact […] does revision at this level takes
place.” These pragmatic assumptions that writing history is still a viable procedure for deduc-
ing a plausible past from textual representations of past events have to be distinguished from
Hans Kellner’s (2000, 280) or Hayden White’s poststructuralist claim that “when it comes to the
historical record, there are no grounds to be found in the record itself for preferring one way of
 The Fictionalization of History in Metahistoriographic Fiction   157

Consequently, Drabble’s fiction defamiliarizes the process of writing history

without completely undermining it. It asks serious questions about the writing
and nature of history beyond postmodernist joking, and it intersubjectively
inserts acknowledged narratives of what might have happened in the past. Based
on the idea that provisional historical knowledge can still be produced and
transmitted—on the premise that this knowledge is acknowledged by a scientific
community and considered historically and culturally variable—Drabble’s novel
refutes the idea that fiction and fact, as well as past and present, have indistin-
guishably merged:

Time past and time present, London and Seoul, seem to be flowing through one another.
They have not merged, they remain distinct, but they coexist, in some dreamlike time of cor-
respondences. They do not fuse or melt. (Drabble 2004, 333–334; emphasis added)

2 After the Constructivist Challenge

The subgenre discussed in this article uses specific narrative strategies to under-
score this thematic focus. I will identify three of these strategies before I locate
them in Margaret Drabble’s quintessential post-constructivist metahistorio-
graphic fiction: the use of heteroreferentiality; the employment of explicit meta-
historiographic, metafictional and metanarrative commentaries; and the applica-
tion of unreliable narration. These narrative strategies are the ones most affected
by the changed premises of metahistoriographic fiction after the constructivist
challenge. Since all three strategies are employed in both historiographic meta-
fiction and post-constructivist metahistoriographic fiction, the two types of
novels cannot be differentiated by their inherent narrative strategies. Rather, the
functions and effects of the same narrative devices differ in both cases.
First, historical novels (and in particular historiographic metafiction) cannot
do without instances of heteroreferentiality, i. e., intertextual references to fic-
tional and non-fictional texts, as well as extratextual references to historical
events, figures, objects or places. It is necessary to analyse in which ways the

construing its meaning rather than another” (1983, 136–137) as well as from Sir Geoffrey Elton’s
resentful retort to the poststructuralists:
In battling against people who would subject historical studies to the dictate of literary critics
we historians are, in a way, fighting for our lives. Certainly for the lives of innocent young people
beset by devilish tempters who claim to offer higher forms of thought and deeper truths and
insights—the intellectual equivalent of crack. (Elton 1991, 41)
158   Manja Kürschner

texts refer to these entities outside of themselves. In contrast to historiographic

metafiction, novels such as Drabble’s The Red Queen affirm the view that histo-
riography proceeds under agreed rules and corresponds to specific criteria such
as continuity, coherence or causality—criteria that specify what evidence is the
most plausible, the most scientific. Examining the footnotes, bibliographies, quo-
tations or non-fictional intertexts in these novels supports the claim that meta-
historiographic fiction takes up established historical knowledge on which a sci-
entific community agrees.
Novels such as Margaret Drabble’s The Red Queen, James Robertson’s The
Fanatic (2000) or Adam Thorpe’s Hodd (2009) discuss the interplay of what Wolf-
gang Iser (1983, 122) calls the imaginary and the real in narratives about history
and historiography. Neither the real nor the imaginary automatically turn into
fiction once they are embedded into a fictional text; instead, they are fiction-
alized and thus turned into linguistic signs. The fictionalized real can interact
with the fictionalized imaginary. While fictionalization is an essential process to
enable the contact of the real with the imaginary on the same ontological level,
some genres of fiction do not cease to mark the different origins of the narrative
facts in the story.
Similarly, The Red Queen does not modify the Iserian real in order to satirize
or tease its readers, who still wonder whether the historical information given
might be fictionalized beyond recognition. It no longer shares the assumptions of
historiographic metafiction that, according to Fludernik, allow “for the ironic (ab)
use of, and the self-reflexive play with, factual parameters” (1994, 91) due to the
fact that “historical ‘reality’ itself is […] conceived of as fantastic and chaotic” in
the “predicament of general disorientation” (92). The novel leaves no doubt that
historiography and fiction can be distinguished due to conventions of reading
and to intratextual and paratextual markers. In that, it echoes and exemplifies
Fludernik’s apt remark that “one can exaggerate the fictionality of historiogra-
phy” (82).
Second, post-constructivist metahistoriographic fiction stays true to the genre
conventions of historiographic metafiction which prescribe explicitly-stated
markers that give away the constructedness of the text. Metafictional autorefer-
entiality in metahistoriographic fiction makes the status of fiction explicit, and
leaves no doubt about it. Metahistorical and metahistoriographic comments chal-
lenge the claim of positivist historiography to offer ‘true insights’ into the past.
However, the narrators of the more recent subgenre, who comment on the condi-
tions of historical knowledge, do not draw an entirely pessimistic conclusion as
to the possibility of generating historical knowledge altogether.
Post-constructivist metahistoriographic fiction is a reaction to the radical
claims of postmodernist thought in historiographic metafiction and—without
 The Fictionalization of History in Metahistoriographic Fiction   159

abstaining from irony and meta-commentary—affirms the view that consensus

on historical facts can be reached. This renewed interest in history corresponds
to the rise of neo-realism in fiction. Josh Toth acknowledges the return to realism
in postmodernist fiction as “ultimately symptomatic of a broad epistemological
renewal of faith—that is, faith in the promise (of mimesis, of communication,
etc.) and the impossible possibility that it will be fulfilled” (Toth 2010, 119).
Without rejecting postmodern claims, “neo-realism seems willing to do what
postmodernism was not (yet) willing, or able, to do: embrace both realism and
metafiction as equally contingent ‘language games.’” (119). Post-constructivist
metahistoriographic fiction consequently uses references to reality and other
markers of factuality to counterbalance the numerous markers of fictionality
and to characterize the postmodernist obsession with meta-commentaries as yet
another ideology.
Third, I should hasten to add that the use of unreliable narration in novels
such as Drabble’s The Red Queen serve as a means to distinguish between histo-
riographic metafiction and its subgenre. Following Tamar Yacobi, I take narra-
tive unreliability to be a “mechanism of integration” applied by the readers; an
“inference that explains and eliminates tensions, incongruities, contradictions
and other infelicities the work may show by attributing them to a source of trans-
mission” (1981, 119). Since the 1990s, Kathleen Wall (1994) and Bruno Zerweck
(2001) have argued that unreliable narration is no longer a heuristic category for
the interpretation of postmodernist novels because of the lack of a reliable coun-
Despite the recent trend to dismiss the concept of narrative unreliability
altogether, reading the narrator as unreliable remains a useful interpretative
strategy if the narration concerned emphasizes the question of reliability, both
thematically and structurally. Narrative instances in historiographic metafiction
frequently underline the limits of the story told, the act of storytelling and the
storytellers in the realm of historical narratives. To pose and answer the question
as to whether parts of the story are reliably told is hence constitutive to a genre
that does not present its narrators as infallible historians, but as human beings
with subjective reasons for their way of (re)telling history.
Unlike historiographic metafiction, the more recent subgenre stresses the
fact that it remains possible to present the course of events in a reliable fashion.
Whenever narrative unreliability is evoked in the texts, it is limited to small pas-
sages and is seldom used to question historically established facts. When used
selectively, it displays the belief that a subjective perception of past events will be
considered as intersubjectively valid and reliable if it were to meet a
160   Manja Kürschner

consensus which obtains about cognitive standards of acceptability and plausibility (such
as scope, explanatory power, consistency, adequacy) in evaluating the provisional truth-
claims of historical inference to the best explanation. (Robinson 2011, 12)

We should not forget that there is a structural analogy between historiography

and narrative unreliability in this matter. Assuming that it is worthwhile sticking
to the heuristic value of both, historiography and narrative unreliability can still
be distinguished from their counterparts, i. e., fiction and narrative reliability. If
one assumes that the historical text can be regarded as a plausible representation
of the past (which, however, cannot be proven beyond all doubt), one can also
assume that in the realm of fictional stories there is still a possibility of a reliable
report in the sense of trustworthiness and validity.
Therefore, recent metahistoriographic fiction rejects radical calls for merging
history and fiction into fiction as well as calls for repealing the concepts of reli-
ability and unreliability. It is this difference that determines the major function
of the subgenre of historiographic metafiction: to favour a contemporary histori-
ography which acknowledges its limits of reliability but which at the same time
responds to the need for reliable information.

3 “They do not fuse or melt”—Margaret Drabble’s

The Red Queen
Margaret Drabble’s The Red Queen illuminates the ways in which heteroreferenti-
ality, explicit metahistoriographic, metafictional and metanarrative commentar-
ies, as well as unreliable narration, are employed as narrative strategies in proto-
typical post-constructivist metahistoriographic fiction.
Regarding the strategy of marking heteroreferentiality with the help of nar-
rative signposts of factuality, a first observation reveals that Drabble’s fiction
spatially separates those parts of the text that allude to the historiographic back-
ground and historical research on the topic. The novel contains a prologue, an
afterword, acknowledgements, a note on sources and a bibliography. Not only
does this narrative frame create the illusion of an editor-narrator who tries to
make sense of a found manuscript before daring to fictionalize it, but it also
encompasses meticulous descriptions of the editing process and of the research
carried out by the inscribed author, Margaret Drabble. It also foregrounds the
mechanisms and pitfalls of historiography by revealing the deficiencies of the
historical manuscript on which the novel is based. All of this sounds familiar
when recalling historical novels of the nineteenth century—and yet it is different.
 The Fictionalization of History in Metahistoriographic Fiction   161

Although some of the time-honoured literary traditions (such as the topos of the
lost and found manuscript) reoccur, the historical facts evoked in The Red Queen
are not invented, but rather backed up by bibliographic notes.
The narrative is inspired by the four Memoirs of Lady Hyegyŏng. Lady
Hyegyŏng, Crown Princess of eighteenth-century Korea, narrates her life at court.
In her fourth memoir, Lady Hyegyǒng talks about the death of her husband,
Prince Sado, and the reign of King Yŏngjo, the twenty-first king of the Korean
Joseon Dynasty. This period of reign has become famous because of a royal fili-
cide: King Yŏngjo killed his son, Prince Sado, because the latter was accused of
randomly killing people in the palace.
The Crown Princess’s ensuing attempts to explain the motives of the differ-
ent political agents in the conflict turn into the non-fictional historical basis of
Drabble’s fictionalization. The Red Queen subsequently fills a gap in the official
history of Korea, but it does not imply that history itself is a form of fiction. It
enriches the historical knowledge handed down in the court chronicles by taking
into account the memoir of an eyewitness. The latter is less objective, but never-
theless provides plausible reasons.
The novel opens with a prologue in which a novelist-narrator and inscribed
author states that she created a fictional account based on Lady Hyegyǒng’s
memoirs. In order to mark her borrowings, the novelist-narrator explicitly

I have supplied some invention, and added some interpretations, most of which are overtly
displayed as interpretations, rather than facts. There are (and have been) many possible
interpretations of the story, and mine is only one of them. You will find details of sources and
a bibliography at the end of this volume. (Drabble 2004, viii; emphasis added)

Clearly indicating the difference between the Iserian real and the imaginary by
spatially separating them from one another in different parts of the novel, The
Red Queen suggests that the real can be neatly distinguished from the imaginary.
However, in her bibliography, the inscribed author includes one fictitious source
which she has mixed in with the other sources of verified historical knowledge:

The memoirs are known in English under various titles, and there are at least three transla-
tions, taken from variant texts. (The translation by Thea Ŏ. Landry mentioned in my text is,
of course, a fiction.) (Drabble 2004, 355–356)

By adding two narratives that unfold the transmission of the fictionalized memoir
from the Korea of the eighteenth century into contemporary Britain, Drabble
chooses to show the fictionalized historical document as part of a complex
network of communication. Simulating the transmission of the document
162   Manja Kürschner

through time and space, the novel emphasizes that establishing history is a col-
lective, culturally variable activity based on consensus. This activity is dependent
on the different agents who assess and combine historical facts to form a coher-
ent historical narrative as well as on the readers of the resulting product.
Despite the heteroreferential markers of factuality, the novel is brimming
with meta-commentaries: reviewer Mary Whipple (2004) criticizes the mixture
of metafiction, metanarration and metahistoriography for “lacking in sub-
tlety.” While turning Lady Hyegyǒng’s historical autobiography into a novel, the
inscribed author in the primary narration proceeds with caution and incessantly
reminds the readers of the fact that they are merely reading a fictional adaptation:

I have turned her story into a novel, of a kind. This is because I am a novelist, and, for better
and for worse, writing novels is what I do. I do not know if this is what she would have
wanted. She wanted something, but this may not have been it. It may well be that she would
have utterly deplored the liberties I have taken with her story. Being dead, she has not had
much say in the matter. She has had no control over how her readers interpret or adapt or
translate her story. (Drabble 2004, vii–viii)

This passage contains ironic metafictional, metahistoriographic and metanarra-

tive statements about the process of fictionalizing a historical document, deci-
sions about how to tell the story and the ways in which the past can be distorted
by historians or political agents. In outlining the characteristics and limits of her
profession, the inscribed author cunningly concludes: “Novelists […] are not to
be trusted. They steal; they borrow; they appropriate. You should never tell them
anything, if you want to keep it a secret” (Drabble 2004, 351). Her ironic warning
foreshadows what is going to happen: the inscribed author in the novel acquires
a manuscript that she turns into fiction.
Subsequently, the novel’s entire structure serves as a metafictional comment
on the ontological status of the novel by reduplicating the communicative situ-
ation in the prologue of the inscribed author. In analogy to the real-life author,
the inscribed author illustrates the relation between the real and the imaginary
in the process of fictionalization while underscoring that the product, a “work of
fiction and fancy” (Drabble 2004, viii), needs to be clearly distinguished from its
historical origin.
In comparison to the historical memoir, the novel adds markers that render
certain sections unreliable. It is vital to explore the nature and the effects of these
unreliably narrated passages, because they support what has been said so far
about the novel’s approach to the ongoing discourse on historiography. Narrative
 The Fictionalization of History in Metahistoriographic Fiction   163

unreliability in The Red Queen is not limited to the intranarrative, internarrative

or intertextual realm, but extended to extratextual unreliability.7
When the inscribed author admits that the fictionalized Korean historian
JaHyun Kim Haboush “does not endorse [her] interpretation, and has had no
influence over the point of view or overall tone” (Drabble 2004, viii), she mirrors
Haboush’s real-life reaction. The historian Haboush outside the novel expresses
her dismay with the novel in an interview:

There is a kind of moral authenticity that we are looking for, and in that sense, The Red
Queen is Margaret Drabble’s imagination—it’s very much embedded in the British post-war
literary generation, where enlightenment comes from the West. (Park 2004)

In the novel as well as in real life, Haboush thus accuses the novelist of having
manipulated Lady Hyegyǒng’s historical voice and of relocating the narration
to the present Western culture. Thereby, she casts suspicion on the authenticity
and reliability of the historical account in the novel. Both Margaret Drabble as
the empirical author of The Red Queen and the fictionalized version of herself,
which she inscribes into the novel, meet this criticism by referring to their right
to fictionalize and invent as part of their profession. Although Drabble indicates
Haboush’s dismay due to the distortion of the ‘original’ voice, she stages the
embedded narrative about Lady Hyegyǒng as a reliable narration: she invites a
comparison of Haboush’s translation and her own account while only admitting
slight, but clearly marked deviations from this translation.
Additionally, however, Drabble creates the impression of Lady Hyegyǒng
being a cunning and calculating woman—an impression that is not evoked in
Haboush’s translation. The fact that Drabble’s narration sullies Lady Hyegyǒng’s
reputation as an innocent victim of Korean history to a small extent might have
increased Haboush’s anger. At an extratextual level, the limits of the metahisto-
riographic novel are evident: a reader from a different cultural and ideological
background might not share the novel’s claim that history can be transmitted as
a commonly accepted explanation of the past.
At an intranarrative level, the representation of Lady Hyegyǒng as a calculat-
ing storyteller results in a stigmatization of the princess-narrator as morally unre-
liable: the Princess helps her husband to hide his deviant deeds from his father’s
view in order to avoid the king’s rage. Instead of condemning her husband for
raping her ladies-in-waiting, torturing her servants and finally killing them, the
Princess starts defending these cruel deeds by reducing Sado to the status of a

7 See Per Krogh Hansen’s distinction (2007, 241) between the four different forms of unreliability.
164   Manja Kürschner

child: “But in truth I do see us, as from afar, like two dolls in a distant pageant.
Two small, overdressed, unhappy, innocent dolls” (Drabble 2004, 87).
As this statement indicates, the Crown Princess is well aware of her audi-
ence and addresses it directly by allaying the doubts referring to the narrator’s
reliability. She enhances the impression of innocence while listing the attributes
‘small’, ‘overdressed’, ‘unhappy’ and ‘innocent’, which are combined with the
noun ‘doll’. Directed and even dressed by others, the prince and princess are por-
trayed as passive and non-autonomous objects that cannot be held responsible.
The discrepancies between the narrator’s attempt to protest her innocence, and
the information that the Crown Princess actually assists her husband, can be nat-
uralized by calling the princess unreliable. This trend persists when deceptive
unreliability, as defined by Jan Stühring, can be associated with the princess-nar-
rator in the first of the three embedded narratives. Stühring defines his notion of
deceptive unreliability as follows:

A narrative n of a work of fiction f is deceptively unreliable if and only if n justifies (at least
temporarily) a belief b, such that b is ›p in f‹, and not all optimal possible interpretations
agree on p in f. (Stühring 2011, 107)

In other words, a narration deceives the readers if it suggests and justifies a spe-
cific reading that is later refuted by other justified readings. In comparison to
stereotypical examples of deceptive unreliability, such as Ambrose Bierce’s “An
Occurrence at Owl Creek Bridge,” the degree of deception in The Red Queen is
nevertheless small: it only consists of the princess-narrator’s attempt to create
the impression that she is innocent. Notwithstanding this ruse, later passages
reveal that the princess has actively participated in the events that lead to the
murder of her husband. Although Lady Hyegyǒng tries to win over her audience,
textual discrepancies prevent the readers from seeing her as an innocent victim of
court politics. When describing the first of her husband’s numerous killings, the
narrator temporarily deceives the readers:

I remember the illustrated text I was reading; I remember the pattern of blossom and but-
terfly; I remember the gilt thread; I remember even the vermilion and turquoise of the cloth
spool on which the thread was bound. It was a tranquil domestic scene. […] So picture me,
innocently employed, sitting on my low, silk-cushioned rosewood couch. Then suddenly
my husband Prince Sado burst in, through the outer chamber, past the ladies-in-waiting,
carrying before him a strange round object stuck on the end of a short spike—it looked to be
about the size of a large cabbage. […] Trembling, I removed my glasses, and then I saw what
I saw. Sado was bearing before him a severed head. It was not a papier-maché mask from
a peasant puppet show, but a real head. […] I do not remember if I screamed or not. Later,
my ladies assured me I conducted myself with dignity, but I cannot remember what I did or
said. (Drabble 2004, 84)
 The Fictionalization of History in Metahistoriographic Fiction   165

This part of the narrative contains numerous strategies to enhance the cruel visual
impression on the readers. It is an adequate example for the narrator’s skill in
selecting, arranging and presenting events which does not miss its effects. Intro-
ducing the scene with a stereotypical domestic and peaceful feminine setting, the
male intrusion of violence seems aggressive and hostile.
The fictionalized Lady Hyegyǒng foregrounds her vivid memories of the tran-
quil scene before the assault by repeating the phrase ‘I remember’ four times in
one sentence. Thus, she stages herself as an innocent victim who tries to protect
her family against male capriciousness: the gendered space of the princess is vio-
lently invaded by Sado, who forces his way into the last private resort of the prin-
cess, past the outer chamber and the ladies-in-waiting. The narrator then defers
the identification of the ‘strange round object’ and again alludes to the foremost
peaceful atmosphere by comparing the object to a harmless cabbage. When she
finally reveals that her husband is not holding a cabbage but a severed head on a
spike, she turns once more to a less violent comparison, a ‘papier-maché mask’,
which elucidates the discrepancies between reality and her seemingly innocent
attempt to comprehend and categorize the object.
Finally facing reality, her repeated claim not to remember what happened in
and after this recognition is contradicted to the repetition of ‘I remember’ at the
beginning of her account: while the narrator remembers every tiny detail on the
embroidery she made two hundred years ago, she denies having a memory of her
reaction to the severed head. These discrepancies between existing and missing
memories make readers aware of the fact that the narration is a carefully con-
structed tale. After all, the narrator has a vital interest in convincing her audience
that the events took place in the exact same way as described in her account. Her
claim, however, is undermined by her confession to have forgotten many crucial

Strangely, I cannot remember what happened to my little brothers at this time. Were they
with me and my mother in Pansong-bang, or with my father and the wet nurse? It is imma-
terial, but it is strange that I cannot remember. My memories are full of gaps. (Drabble 2004,

Complaining about memory gaps is a topos that frequently occurs in historio-

graphic metafiction aimed at underlining the factual unreliability of eyewit-
nesses in historiographic research. To fight this stigma of factual unreliability, the
Crown Princess constantly compares her mental superiority and sanity with her
husband’s insanity: “The madder Sado became, the more I believed in reason. I
took refuge in reason and in the life of the mind. Posterity is witness to my ration-
ality” (Drabble 2004, 72). However, Lady Hyegyǒng’s rationality is undermined
166   Manja Kürschner

by two important narrative facts: first, the Crown Princess is shown to be very
superstitious and less rational than she pretends to be and; second, contrary to
the non-fictional historical text, she is presented as a ghost in Drabble’s fiction.

4 Spectral Transformations
The radical transformation of a historical eyewitness into a ghost-narrator in a
fictional realm serves as a marker of fictionality. We have to ask ourselves what
benefits this transformation brings, if any. One possible benefit is that a ghost
can overcome the spatial and temporal restrictions and thus turns into a tran-
scendental historian who functions both as a chronicler and as an eyewitness
for the historical events. Using a ghost in the embedded story serves to make a
point that cultural narratology has highlighted since its emergence: we need con-
text-sensitive interpretations of texts, since the transmission and the reception of
(historical) narratives are culturally dependent.
Similar to many other conventional homodiegetic narrators in fictional texts,
the narrating I of the first part of the novel writes history, or in this case herstory,
by maintaining a temporal, spatial and emotional distance from the experienc-
ing I of the eighteenth century. Although the Crown Princess cannot look into
the minds of her husband or other members of the court, she can explain their
potential motives with more hindsight. In this, she imitates a historian’s attempt
to render an account of what actually happened.
A second advantage of this narrative trick is that it enables anachronistic
remarks such as the narrator’s musings on psychoanalysis or psychological ter-
minology, although the story this terminology refers to is 200 years older than
Freud. When analysing her mother’s behaviour after having given birth to her
brother, the narrator, Lady Hyegyǒng, asks: “Was it a form of postnatal depres-
sion? Such a condition was not officially recognized or named in those days,
though it was common enough” (Drabble 2004, 12). The use of a ghost narrator
is a device to show how contemporary knowledge about the Korea of the eight-
eenth century replaces the knowledge visible in the historical document of Lady
Hyegyǒng, and in what ways historical knowledge is affirmed or refuted by a later
generation of historians. It enables a meta-commentary on the nature and devel-
opment of historical knowledge.
Third, the ghost represents the link between the first and the following two
narratives: a character-narrator who has lived in one culture and period can
 The Fictionalization of History in Metahistoriographic Fiction   167

compare it to the social and political evils of another.8 In an interview, the empir-
ical author Drabble explains her decision to let her character-narrator haunt the
entire setting of the novel on all diegetic levels:

Well, I made her a ghost so that she could leap out of her own period and comment on the
present day, because one of the things I was trying to do in the book was to point out that
progress isn’t all it’s meant to be, that people still die very unpleasant deaths. (Lee 2007,

The ghost as a marker of fictionality deviates from a realistic realm and enables
a comparison of different factualities in different times and cultures. Instead of
concluding that the Korea of the eighteenth century cannot be understood or
compared to Great Britain in the twenty-first century, the novel uses the ghost
of the Korean Crown princess as a mouthpiece to ask: “How much of the past, I
wonder, lingers in the air?” (Drabble 2004, 160). The remaining parts of the text
contain the answer to this question: history influences the present and can be
narrated and transmitted in the acknowledged form of historical facts.
Thus, despite the drawbacks and problems of narrating the past, The Red
Queen still claims to offer a factual portrait of eighteenth-century Korea. The
resulting picture corresponds to what Western readers believe Korea to have
looked like two hundred years ago. Along cultural lines, however, the question
remains whether a British author is entitled to modify, and is capable of critically
assessing, Korean history in her fiction.
In analogy to other examples of post-constructivist metahistoriographic
fiction, The Red Queen exemplifies a strategy for representing historical knowl-
edge as well as doubts concerning this knowledge. Alan Robinson summarizes
this strategy, which applies not only to Drabble’s fiction but also to non-fictional
postmodernist historiographies:

Although the avowedly provisional truth-claims of constructionist historians cannot be

proven, acting pragmatically on the assumption that consensual views bear some plausible
relation to truth is a strategy which works. (Robinson 2011, 27)

Thus, despite the signposts of fictionality, such as explicit metafiction and unre-
liable narration and despite the limits of transcultural understanding, The Red

8 Hence, the Crown Princess argues, for example, that the political leaders of Korea are frequent-
ly overrated: while King Sejong is said to have invented the Korean alphabet although he has
only promulgated it, Kim Jong-il has not invented the massed dance notation, but only written a
theory about it (Drabble 2004, 35).
168   Manja Kürschner

Queen claims to transmit intersubjectively valid historical knowledge: the nar-

rators may be biased and may try to evoke sympathy in the reader, but they are
still shown to render a reliable report which corresponds to the historical memoir.
Moreover, narrative unreliability is limited to only a few specific passages.
In their article on unreliability in autobiography, Dan Shen and Dejin Xu affirm
the view that a verdict on a text or a narrator’s unreliability depends on the genre
conventions and on the comparison with other texts of the same or a different

Hence we can see the functioning of two representative mechanisms: the generic mecha-
nism: how one views the ‘factualness’ of autobiography as a genre affects one’s response to
the autobiographer’s unfaithful presentation of facts, hence bearing on the interpretation
of ‘factual unreliability’; and the comparative mechanism: the factual (un)reliability of the
narration is judged in relation to other works of the same genre, and its ideological (un)
reliability is judged in relation to the ideological behavior of other comparable persons […].
(Shen and Xu 2007, 78–79)

In comparison to historiographic metafiction, post-constructivist metahistorio-

graphic fiction considerably reduces the scope of unreliability. Notwithstanding
all uncertainty and unreliability, novels of this type produce the effect that fic-
tional and historiographic discourse can be distinguished and that the partial
unreliability of the narrators does not exclude the possibility that their reports
contain reliable information. Heteroreferential comments do not turn out to be
invented and, if they are, they are marked as such. Historical information given
in the fiction actually refers to the non-fictional historiographic sources indicated
in the footnotes or the bibliography.
Hence, there is little postmodernist play on historical facts. Nevertheless, the
subgenre employs both: strategies that enhance the impression of factuality and
metafictional comments that clearly point out the status of the text as a whole.
As a result, the reader is made aware of the factuality of some passages, and of
the fictionality of others, with the help of intratextual markers of fictionality and

5 Conclusion: Constructivism Reloaded

Despite its similarities to historiographic non-fiction, recent metahistoriographic
fiction does not pretend to serve the same functions. Its added value, in compari-
son to non-fictional texts about the past, is that the three narrative strategies (the
 The Fictionalization of History in Metahistoriographic Fiction   169

use of reliable heteroreferentiality the application of explicit meta-commentary

and the curtailing of unreliable narration) are used to satirize historiography.
Having said this, I want to grant full force to the claim that the nature of
satire has always been twofold. On the one hand, a specific convention is ridi-
culed and disclosed as a more or less arbitrary tool to convey a specific message.
This is what happened to positivist historiography in historiographic metafiction
in the 1960s and 1970s. On the other hand, satire helps to redeploy a strategy
that has been subjected to close scrutiny in case it turns out to be a necessary
way of dealing with what surrounds us. Satirical genres, such as historiographic
metafiction or its subgenre, post-constructivist metahistoriographic fiction, take
up contemporary discussions about how to write and assess history. They either
ridicule the attempt to write history or explore ways of coping with the difficul-
ties of such an endeavour. The latter type of fiction tends to answer the call for
viability instead of relativity when it comes to generating historical knowledge.
Nevertheless, I have also highlighted the limits of this subgenre: its suggestion
that the real and the imaginary can be neatly separated might be valid within the
story world, but this is refuted by evaluations of extrafictional reviewers. These
critics either disapprove the high amount of reference in fiction or the still-too-
liberal adaptation of historical narratives in the realm of the novel.
It clearly remains that the interaction of the real and the imaginary in fiction-
alized historiographic accounts needs to be examined further. Only if we start
considering post-constructivist metahistoriographic fiction as a form of histori-
cal novel which uses constructivist historiography for new and pragmatic ends
will we be in a position to assess new ways of dealing with history in fiction and
outside. Much more work needs to be done if we want to gain a more complex
picture of narrative unreliability and reliability of historiography in recent meta-
historiographic fiction. Eventually, the ghost narrator in The Red Queen con-
cludes and—with it—summarizes the state of the art in this field of research: “No
story is ever finished. Mine continues” (Drabble 2004, 165).

Works Cited
Elton, Geoffrey R. 1991. Return to Essentials. Some Reflections on the Present State of Historical
Study. Cambridge: Cambridge University Press.
Evans, Richard. 1997. In Defence of History. London: Granta Books.
Fludernik, Monika. 1994. “History and metafiction: Experientiality, causality, and myth.” In
Historiographic Metafiction in Modern American and Canadian Literature, edited by Bernd
Engler, 81–101. Paderborn: Schöningh.
170   Manja Kürschner

Hansen, Per Krogh. 2007. “Reconsidering the unreliable narrator.” Semiotica 165 (1/4):
Hutcheon, Linda. 1988. A Poetics of Postmodernism: History, Theory, Fiction. New York:
Iser, Wolfgang. 1983. “Akte des Fingierens oder Was ist das Fiktive im fiktionalen Text?” In
Funktionen des Fiktiven, edited by Dieter Henrich and Wolfgang Iser, 121–151. München:
Wilhelm Fink Verlag.
Jays, David. 2004. “Seoul destroying: Margaret Drabble’s historical Korean fable, The Red
Queen, is long on detail, but short on imagination.” The Guardian. 22 August. http://www. (Accessed 13 December 2013)
Kellner, Hans. 2000. “Narrativity in History: Post-structuralism and Since.” In The Narrative
Reader, edited by Martin McQillan, 275–283. London and New York: Routledge.
LaCapra, Dominick. 1983. Rethinking Intellectual History: Texts, Contexts, Language. Ithaca, NY:
Cornell University Press.
Lee, Young-Oak. 2007. “An Interview with Margaret Drabble.” Contemporary Literature 48 (4):
Nünning, Ansgar. 1995. Von historischer Fiktion zu historiographischer Metafiktion. Band I.
Theorie, Typologie und Poetik des historischen Romans. Trier: WVT.
Park, Ed. 2004. “A Talk With the Translator of the Memoir You Have to Read Next.” The Village
Voice. 26 October.
translator-of-the-memoir-you-have-to-read-next/full/ (Accessed 26 December 2013)
Robinson, Alan. 2011. Narrating the Past: Historiography, Memory and the Contemporary Novel.
Basingstoke: Palgrave Macmillan.
Shen, Dan, and Dejin Xu. 2007. “Intratextuality, Extratextuality, Intertextuality: Unreliability in
Autobiography versus Fiction.” Poetics Today 28 (1): 43–87.
Stühring, Jan. 2011. “Unreliability, Deception, and Fictional Facts.” JLT 5 (1): 95–108.
Swift, Graham. 1983. Waterland. New York: Poseidon Books.
Thurber, Bart. 2013. “Go ask Margaret: ‘The Red Queen’s lost her head, and her way, in
Drabble’s ambitious, confused tale.” U-T San Diego. 24 October. http://www.utsandiego.
com/uniontrib/20041024/news_mz1v24queen.html (Accessed 26 December 2013)
Toth, Josh. 2010. The Passing of Postmodernism: A Spectroanalysis of the Contemporary. New
York: State University of New York Press.
Wall, Kathleen. 1994. “The Remains of the Day and Its Challenges to Theories of Unreliable
Narration.” The Journal of Narrative Technique 24 (1): 18–42.
Whipple, Mary. 2004. “Margaret Drabble ‘The Red Queen’.” Mostly Fiction Book Reviews. 19
December. (Accessed 13 December 2013)
White, Hayden. 1983. “The Politics of Historical Interpretation: Discipline and De-Sublimation.”
In The Politics of Interpretation, edited by W. J. T. Mitchell, 136–137. Chicago: The University
of Chicago Press,
Yacobi, Tamar. 1981. “Fictional Reliability as a Communicative Problem.” Poetics Today 2 (2):
Zagorin, Perez. 1999. “History, the Referent, and Narrative: Reflections on Postmodernism
Now.” History and Theory 38 (1): 1–24.
Zerweck, Bruno. 2001. “Historicizing Unreliable Narration: Unreliability and Cultural Discourse
in Narrative Fiction.” Style 35 (1): 151–178.
Silke Lahn (Hamburg) and Stephanie Neu (Mannheim)
Towards a Crossing of the Divide between
Fiction and Non-Fiction in European
Television Series and Movies: The Examples
of the Italian Romanzo Criminale and the
Danish Klovn
Non-fiction and fiction have generally been understood to be two distinct entities.
Narratives were supposed to belong either to a factual representation of reality
or to an invented representation. The only problem was to settle on an agreed
definition for the categorial distinction of fictional and factual narratives—an
endeavour which, to date, has yielded mainly an ongoing debate about determin-
ing acceptable criteria for fiction.
Currently we find ourselves in the nascent phase of various forms of New
Media which are struggling to define themselves and their relationship with
existing genres in Old Media. This process is not only establishing new ways for
how a (real or an invented) story can be told, but it is also influencing the old
ones. Seen from this perspective, playing with the postulated divide of fiction
and non-fiction seems a logical consequence: techniques and strategies that were
long regarded as typical for one or the other now travel back and forth. The result
is a growing number of new hybrid forms in which the receiver is frequently faced
with the difficulty of deciding whether a narrative, in whole or part, is fictional or
factual. Does this mean that the divide between fiction and non-fiction has finally
become obsolete? On the one hand, Gérard Genette once claimed that “pure” fic-
tional or factual narratives are mere thought experiments, only found in the “poe-
tician’s test tube” (1990, 772). On the other hand, we still have a strong intuition
that differentiating narratives according to “real” and “invented” content does
After an intense debate during the second half of the twentieth century that
failed to achieve generally agreed positions, Richard Walsh (2007) proposed, in
the tradition of the Chicago School, to redefine fictionality not as an ontological
category but as a rhetorical mode.
In our paper, we will test drive this suggestion by looking at two recent and
iconic examples in European storytelling which mix supposedly irreconcilable
elements of fictional and factual origin: Romanzo Criminale, an Italian movie
(2005) and a television series (2008–10) based on a novel of the same name; and
Klovn, a Danish television series (2005–09) and the subsequent film Klovn—The

DOI 10.1515/9783110555158-009
172   Silke Lahn and Stephanie Neu

Movie (2010). All representations refer to real persons while at the same time
employing storytelling patterns typically associated with fictional narratives.
To begin, we give a short overview of the most relevant approaches to fiction

1 The Problem of Fiction in Literary Theory

1.1 Classical approaches to defining fiction

Following Jean-Marie Schaeffer (2014 [2009]), we subdivide the debate on how to

differentiate fiction from non-fiction into three groups.
1) The semantic definition is inspired by philosophical theories of reference
and rests on the idea of an ontological divide between fiction and non-fiction:
either a narrative has a reference to the “real” world (in the case of factual nar-
ratives) or not (in the case of fictional narratives). A challenge for this approach
arises from so-called immigrant objects (Zipfel 2001, 92), elements imported from
the real world into a fictional universe (e. g., the notorious example of Napole-
on’s appearance in a novel). Lubomír Doležel offers a solution by stating that
real-world elements, when appearing in a fictive world, undergo “a substantial
transformation at the world boundary into non-actual possibles, with all the
ontological, logical and semantical consequences” (1988, 485). However, Doležel
does not explain what this mystic conversion (e. g., of real-world persons) means
in concrete terms; we agree with Frank Zipfel (2001, 94) that it is extremely coun-
ter-intuitive to label real persons or places in fiction as fictive objects out of the
desire to achieve theoretical rigour. Wolf Schmid (2010 [2005], 32), on the other
hand, highlights the consequent danger of gradating the fictive.
2) A syntactic definition seeks to distinguish fictional and factual narratives
by means of distinctive formal criteria that are inherent to the text. Probably the
best-known representative is Käte Hamburger, who describes assumed fiction-sig-
nals in third-person narratives. This occurs with the use of the epic preterit that
ceases to indicate the past in fictional contexts, the use of verbs of inner action
to portray a character’s inner thoughts and the use of free indirect speech (1973
[1957/1968], 59–133). Because of her logico-linguistic approach, Hamburger was
led to exclude first-person narration from the realm of fiction which, as Schaeffer
(2013, 189) rightly points out, is a highly counter-intuitive conclusion. Moreover,
as highlighted by Walsh (2007, 44–45), the supposed essential formal character-
istics of fictional narratives undergo changes and are thus not universally valid.
Henrik Skov Nielsen (2011, 114) also objects to a syntactic definition, pointing
 Towards a Crossing of the Divide between Fiction and Non-Fiction   173

out that alleged fictional techniques can be borrowed by non-fictional texts. It

follows then that inherent signals of fiction are neither necessary nor sufficient to
identify a work of fiction as such.1
3) A pragmatic definition is proposed by John Searle. By way of contrast
to Hamburger, he dismisses the existence of a “textual property, syntactical or
semantic” (Searle 1975, 325) that allows a clear distinction between fictional and
factual narratives. Instead, he identifies as the decisive criterion the “illocution-
ary intentions of the author” (325) and defines fiction as a pretended speech act.
Searle’s approach has been contested, among others, by Kendall Walton (1990),
who rejects the idea that intentionality can suffice as a valid criterion for the defi-
nition of fiction. Nevertheless, Schaeffer points out the importance of the inter-
play between a “fictional intention” and the perception of a given work as fiction:
“[I]f it is true that fictional intention cannot define fiction as a pragmatic stance,
it is nevertheless the existence of a shared intention which explains the fact that
the emergence of fictional devices has the cultural and technical history it has”
(2014 [2009], 190). In short: fictionality, as defined in approaches which empha-
size the pragmatic dimension, is not based on intrinsic features of a given work
but rather on conventions and on a kind of contract between author and receiver.2
Nielsen (2011, 113–114) terms all three approaches that seek to define cate-
gorial differences between fiction and non-fiction as separatist or exceptionalist.
Furthermore, he identifies two positions that do not consider fiction and non-fic-
tion to be opposites:
4) A panfictionalist approach claims that all narratives are artificial and
therefore fictionalizing because they are the result of a transformation of facts
into language. Hayden White, one of the best known representatives of this posi-
tion, states that historians, for instance, charge real events with the symbolic
significance of a plot structure, a feature that is also typical of fiction (1978, 92).
Nielsen is a resolute opponent of panfictionalist approaches: even if we agreed
that all fiction is artificial and that all narrative is artificial, the logical conse-
quence would not be that all narrative is fictional (2011, 114).
5) A similarist or non-fictionalist approach, inspired by cognitive theory and
furthered by Monika Fludernik’s Towards a ‘Natural’ Narratology (1996), is sup-

1 Two years later, Nielsen and his co-authors in Fiktionalitet were to outline ten theses about
fictionality, where this position is summarized as thesis 5 (Jacobsen et al. 2013, 44). Again two
years later, Nielsen et al. reworded this conclusion in the article “Ten theses about fictionality”
as thesis 6: “No formal technique or other textual feature is in itself a necessary and sufficient
ground for identifying fictive discourse” (Nielsen, Phelan, Walsh 2015, 66–67).
2 For a more detailed critique of Searle’s approach, see Zipfel (2001, 185–195).
174   Silke Lahn and Stephanie Neu

ported by David Herman, for example. According to Herman, conversational nar-

ratives are prototypical, and all other narrative forms derive from these. Moreo-
ver, he argues that we always activate the same cognitive repertoire and thus the
same translation apparatus to understand all narratives (2009, 6). Diverging from
Herman’s very popular approach, Nielsen and his co-authors argue that we acti-
vate a particular translation machine when dealing with fictionality (Jacobsen et
al. 2013, 19–20).
To summarize the debate, most recent contributions agree that none of the
elements thus far examined deliver clear-cut criteria for identifying fictional nar-
ratives. On the other hand, approaches that eliminate the divide between fiction
and non-fiction are also widely rejected. It seems that everybody recognizes
fiction when they are confronted with it, but why this happens is still a mystery.

1.2 Richard Walsh’s rhetorical revision

A new approach to tackle the mystery of fiction was presented by Richard Walsh
in his 2007 book The Rhetoric of Fictionality, a deliberate echo of Wayne C. Booth’s
seminal work The Rhetoric of Fiction (1961). Walsh strives to come up with a con-
vincing pragmatic definition. For his starting point, he assumes that the reader
does not expect literal truthfulness; it is rather the presumption of relevance that
drives the reader’s search for an appropriate interpretative context. Thus, Walsh
(2007, 23–29) re-reads Dan Sperber’s and Deirdre Wilson’s Relevance Theory,
itself based on Paul Grice’s Maxims, and applies their insights to fictionality. “The
relevance theory model allows for a view of fiction in which fictionality is not a
frame separating fictive discourse from ordinary or ‘serious’ communication, but
a contextual assumption” (30; emphasis added). Contextual assumptions result
from paratexts as well as from cultural context in a broader sense:

A rhetorical definition of fictionality is pragmatic, in that its criteria are not ultimately
inherent in the narrative itself, but are contextual. The rhetoric of fictionality is brought
into play whenever a narrative is offered or taken as fiction, regardless of issues of form,
style, or reference. (44)

Walsh further stresses that this understanding of fictionality is not of any ahis-
torical essence due to the fact that the conventional forms of fiction are subject to
change. From this, it follows that the “rhetorical scope and import of fictionality
itself are both historically and culturally variable” (45).
 Towards a Crossing of the Divide between Fiction and Non-Fiction   175

With regard to their relevance for our later discussion about the hybrid forms
Romanzo Criminale and Klovn, we would like to outline three of Walsh’s main
1) Instead of fiction, which is more a kind of “category or genre of narrative”
(7), we should focus on fictionality, understood as a rhetorical mode. According to
Walsh, fictionality “is best understood as a communicative resource rather than
as an ontological category” (36), even though he does maintain the idea of its
“categorical distinctiveness” (45). Fictionality as a communicative strategy “is
apparent on some scale within many nonfictional narratives, in forms ranging
from something like an ironic aside […] to full-blown counterfactual narrative
examples” (7). In other words, fictionality can occur in every sort of communica-
tion, even on as limited a scale as a single word.3
2) Fictional communication is not a second-order phenomenon, but rather
it is of fundamental relevance for our real lives and therefore a serious means
“within a real-world communicative framework” (16): “The knowledge offered by
fiction, however, is not primarily specific knowledge of what is (or was), but of
how human affairs work, or, more strictly, of how to make sense of them—logi-
cally, evaluatively, emotionally” (36). In addition, we can understand it “as the
exercise of our narrative understanding” (50–51). This is a relevant observation
in regard to hybrid forms and their impact on “real life,” as we will see later in
our examples.
3) Walsh rules out the narrator as a relevant entity in literary communica-
tion. He takes his departure from Genette’s classification and states “that all
homodiegetic and intradiegetic narrators are equally represented and therefore
characters” (9). For the remaining category of extradiegetic-heterodiegetic narra-
tors, Walsh sees no reason to distinguish these from the real author. He assumes
that a distinction between authorial and narratorial personality was established
to be “available to interpretation as a meaningful aspect of the text’s own rep-
resentational rhetoric” (80).4 This is a tempting proposal for the field of filmic
representation, as the question of whether a narrator is relevant or not is often
discussed in audio-visual narratives. We will come back to this point later.

3 An example for the fictionality mode within everyday communication established in just one
word could be: “Have you noticed how hard Sue is working on her new project? I wonder when
she is sleeping. Extraterrestrial?”
4 The argument that unreliable narration justifies this duality is also rejected by Walsh (9, 78–80
and 174). In his view, considerations like these are unnecessary if we understand that the real
author is fictionalizing.
176   Silke Lahn and Stephanie Neu

1.3 The extension of Walsh’s rhetorical approach to hybrid

forms by Danish scholars

Walsh’s approach has served as the basis for further theoretical considerations,
especially for Danish narratologists at Aarhus University.5 This may be due to the
fact that there is a prominent tendency to mix fictional and documentary formats
in Scandinavia’s contemporary literature and media culture. This situation has
called for a theoretical framework, as emphasized notably by Rolf Reitan (2013,
54). According to him, Walsh’s book represents the beginning of a new paradigm
(Reitan 2011, 52), as it “opens up traditional narratology for pragmatic dimen-
sions that have been and still are unknown to it” (Reitan 2013, 55).6
Reitan’s colleague, Henrik Skov Nielsen, also picks up the idea of fictionality
as a rhetorical resource. He maintains “that fictionalization does not indicate a
technique through which one removes oneself from the world, but a wide range of
options to interact with the world” (Nielsen 2013, 81; cf. Nielsen 2011, 116).7 While
Walsh stresses the importance of the paratext for the “production” of fictionality,
Nielsen also outlines the importance of “techniques of fictionality”:

Using any of a range of techniques of fictionality (including omniscience, free indirect dis-
course, simultaneous narration, imaginative supplementation, and counterfactual narra-
tive) will locally produce fictionality that similarly invites certain interpretative operations
at least towards parts of the narrative—without necessarily turning the whole narrative into
a fictional text. (Nielsen 2010, 282; original emphasis)

In his inaugural lecture, Nielsen acknowledges that it is helpful to differentiate

between fictionality and fiction but observes that he is unable to find in Walsh’s
book “what this differentiation specifically implies for an understanding of fic-
tionality outside of fiction and outside of works” (2011, 116). Nielsen therefore
calls for an extension of Walsh’s approach, not only to fictional genres other than
literature but to non-fiction narratives in general.
Nielsen later published together with other members of the Aarhus-based
research group “Centre for Fictionality Studies” a handbook entitled Fiktionalitet

5 See, for example, “Fiktionens forandringer” [The changes of fiction], a special issue of the
journal Spring (2011) edited by Rolf Reitan et al., and “Fiktion og fortælling” [Fiction and narra-
tive], a special issue of the journal K&K. Kultur og klasse (2013) edited by Per Krogh Hansen et al.
Most of the articles in these publications take Walsh (2007) as their departure point.
6 All translations from the Danish are by Silke Lahn; all translations from the Italian are by
Stephanie Neu.
7 In Fiktionalitet, Jacobsen et al. summarize this position as thesis 9: “Fiction is a fact. Fictionali-
ty is a reality” (2013, 45). A corresponding independent thesis is not listed in Nielsen et al. (2015).
 Towards a Crossing of the Divide between Fiction and Non-Fiction   177

(Jacobsen et al. 2013) in which, like Walsh, they strictly differentiate fictionality
and fiction. While Walsh was wary of presenting his understanding of fictionality
bluntly,8 Jacobsen et al. offer a redefinition of the term to fictionalize, which they
understand both as a communication strategy from the sender’s perspective and
an interpretative hypothesis from the receiver’s perspective:

Fictionality is a property of (mainly narrative) communication (visual, written, verbal,

etc.) which a sender can signal that the communication possesses and which a receiver
can assume that the communication possesses. These two actions can be called ‘fiction-
alization’. Fictionalization serves as a signal that the narrated does not in any immediate
sense describe being and reality as it exists, but on the contrary describes the possible, the
not-being, the potential, the future, the exaggerated, etc. (2013, 9)

The authors insist that this use of to fictionalize does not mean “that something
non-fictive is transformed into something fictive but, on the contrary, that a rela-
tion to reality is ‘foregrounded’ which is non-limited and often non-referential”
(43–44). Fictionalization thus differs from other communicative acts in that it
does not deal with reality—that is, it is not referential—and it differs from lies
because it does not seek to deceive; rather, it signals that reality is not depicted
as it is (9).
Another Aarhus scholar, Johanne Helbo Bøndergaard, explores what she
calls a literary forensic mode. According to her, in order to deal with past events,
“contemporary literary representations […] tend to use strategies conventionally
associated with fictional representation as well as conventions associated with
non-fictional genres and historical traces that seem to ‘prove’ an authentic rela-
tionship to the past” (2012, 3).9 Bøndergaard ties the inclusion of such “forensic”

8 Cf. Simona Zetterberg Gjerlevsen: “Richard Walsh expounds many of the benefits which derive
from using the notion of fictionality, but he does not come up with an exact definition of the
concept” (2013, 57). Reitan (2013, 3) states that it can sometimes be difficult to decide whether
Walsh presents premises or conclusions; moreover, some expressions are explained as late as at
the end of the book.
After completion of our article, Gjerlevsen offered a full definition of the term: “fictionality, as a
fundamental rhetorical mode, is understood as a means to communicate what is invented and
as such transgresses the boundaries of both fiction and narrative. In this perspective, fictionality
is not bound to any genre or limited to narrative representation. Building on this conceptual
framework of fictionality as an autonomous concept, a definition of fictionality as intentionally
signaled invention in communication has been put forth.” (2016, § 1)
9 Such traces of the past could be photographs, for example, but also “documents and other
scraps of evidence” (59). It is important to highlight that these “scraps of evidence” do not have
to be authentic; what counts is “the implicit suggestion of authenticity or truth that clings to the
evidence” (38).
178   Silke Lahn and Stephanie Neu

(or “factual”) elements to one of the benefits of fictionality, as implied by Walsh

and elaborated further by Jacobsen et al.: “Communicating through fictionality,
then, is part of our cognitive ability to imagine the world as being different from
the way it really is” (2013, 62). The insertion of forensic elements can be therefore
an invitation to question ‘our’ reality: the reality these elements presumably refer
to is “opened up for narrative embroidery […], imagination and emotional and
ethical evaluation” (68). As we will see, this notion is of crucial importance for
texts such as Romanzo Criminale.
It is striking that almost all narratological studies cited so far have been pri-
marily interested in fictionality and the problem how to define it. But what about
factuality? Is it a kind of default case and so self-explanatory that we do not even
have to theorize it? In our view, the answer is no. How, then, is factuality as a
rhetorical strategy in fiction to be determined? How many factual elements are
necessary in order for a work to change sides—across the postulated categorial
ontological divide? And how do fictionalization and factualization interact? In our
discussion of Romanzo Criminale and Klovn, we shall seek to answer these ques-
tions while keeping in mind Nielsen’s observation that “fictionalization is a way to
do something with and about reality intentionally and with a purpose” (2013, 79).
We shall also attempt to outline possible ways to deal with factuality by looking at
stylistic devices in the sense of textual, acoustic and visual arrangements.

2 Romanzo Criminale: Between Truth, Facts and

The story told in Giancarlo De Cataldo’s novel Romanzo Criminale [RC] is set in the
1970s and 1980s in Rome. It gives an account of the “Banda della Magliana,” a real
gang of criminals. In 2005, the novel was made into a movie, directed by Michele
Placido with the screenplay co-authored by De Cataldo. A television series soon
followed, directed by Stefano Sollima (also co-authored by De Cataldo): from
November 2008 to December 2010, two seasons were broadcast on the Italian
pay-TV channel Sky Cinema 1.

2.1 Fact-based imaginative supplementation

RC is a typical example of a story based on real events perpetrated by real persons.

The story is told in genres which are conventionally marked and perceived as fic-
tional. The paratext of all three representations of RC sends unequivocal mes-
 Towards a Crossing of the Divide between Fiction and Non-Fiction   179

sages: the book is labelled “novel”; disclaimers at the beginning of the movie
and in the end credits of the series clearly state that the narrated events are “free
adaptions” of real events and that the characters are the product of the authors’
imagination.10 Nevertheless, references to real events are clearly evoked, as for
example by a rhetorical question on the book cover: “In the very recent past, a
band of street gangsters tried to take hold of the city of Rome. Did this really
happen?” (De Cataldo 2006 [2002]). Additional material on the DVD-edition of
the movie conveys a similar message: one of the features is dedicated to the inter-
play between “fiction and reality.” To put it in a nutshell: on the one hand, the
movie deals with reality, while on the other it contains “imaginative supplemen-
tation” (Nielsen 2010, 282).
This imaginative supplementation affects the plot and the main characters.
The characters are shaped according to the genre conventions of the crime novel:
a genre evoked by the title itself, Romanzo Criminale (crime novel). In the novel
as well as in the movie and television series, the three gang leaders—Dandi, Lib-
anese and Freddo—are fixed in predetermined roles from the beginning: Dandi is
the good-looking playboy who enjoys good clothes, an elegant lifestyle and beau-
tiful women; Libanese is an angry young man while Freddo, as his name (“freddo”
means “cold” or “cool”) suggests, is the gang’s coldblooded mastermind. As is
common in crime novels and movies, the gangsters have an antagonist: Com-
missario Scialoja, the police officer who tries to put an end to the gang’s criminal
activities. Also, as is often the case in the tradition of the hard-boiled novel or film
noir, a love plot counterbalances the crime plot: Patrizia, a prostitute, is pursued
romantically by both Dandi and Scialoja.
Due to their conformity with genre patterns, it is impossible to tell which
characters are purely fictive and which are based loosely on real characters, for
there are no intrinsic features which allow such hypotheses. It is only by consult-
ing text-external material like articles, books and documentaries about the real
Magliana gang that we can identify characters with real counterparts—Dandi,
Libanese, Freddo—and those without real counterparts, such as Scialoja and
Patrizia. As borne out by the many blog postings on the show, such interest in the
historical “veracity” of the story told in RC is indeed en vogue among RC fans, who
try to determine which of the characters’ features and actions correspond to those
of the real criminals.11 This interest is certainly triggered by the paratext (e. g., the

10 Cf. the movie RC, 00:27, and the first episode of the first season of the series RC, 59:50.
11 See, for example, the Italian Wikipedia website dedicated to the television series http://, which includes a chapter entitled “Corre-
spondence with real facts.”
180   Silke Lahn and Stephanie Neu

book cover). Nevertheless, there are also text-internal formal features which may,
in addition to the suggestions made in the paratext, direct the interpretation of RC
in a more “factual” direction.

2.2 Formal techniques of fictionality and factuality:

a comparative outlook
An amalgamation of real and invented elements becomes more obvious in the
movie adaptation of RC. Among other similar examples, we find television news
images about the kidnapping of the politician Aldo Moro by a terrorist group in
1978. With regard to Bøndergaard’s approach outlined above, the scenes about
the discovery of Moro’s dead body in the movie could be described in terms of a
“forensic” mode. Thanks to the audio-visual montage, the historical footage is
integrated smoothly into the sequence of events. First, we see in colour the image
of a helicopter, filmed from below. Immediately following, we see images in black
and white which show how Moro’s body is found in the trunk of a car. The point
of view from high above suggests that we are watching the scene from the per-
spective of someone aboard the helicopter. Off-camera, we hear the blurred voice
of one of the kidnappers talking on the phone and revealing the whereabouts
of the dead body. At the same time, we hear, also off-camera, the distinct sound
of a violin playing a sad tune. Obviously, this music is not part of the historical
footage but, like the background music employed within the whole movie, serves
as an implicit comment on the tragedy unfolding before the viewer’s eyes.
The following images show some of the movie’s characters playing billiards
and watching the news on television, where a speaker comments on the images
that we, the audience, have just seen (54:19–54:40). Hence, the close connection
is further underlined between the authentic images and the presumed fictive ele-
ments within the movie.
In the series, news about Moro’s kidnapping is also shown. As in the movie,
one of the characters is watching the television news, where he (and the audi-
ence) is informed of the kidnapping (season 1, episode 5). But the impact, we
would suggest, is different: in the series, it is more akin to a “citation” so as
to anchor the story in a certain historical context. In the movie, the authentic
footage and the scenes featuring the film’s actors are put on the same level, since
they are shown in rapid succession, as if they belonged to the same sequence
of events. This minor interest in “forensic evidence” within the series could be
understood as a signal for an evolution within the cosmos of RC: thanks to the
novel, the movie and the public discussion about them, it can be assumed that
the viewer is well aware of the interaction with the “real world” in RC. Against this
 Towards a Crossing of the Divide between Fiction and Non-Fiction   181

backdrop, the producers of the series seem to be less concerned with signalling
real-world references.12
At this point, one question arises: does the insertion of such “forensic” mate-
rial alter the movie as a whole, or at least parts of it? The answer is a simultaneous
“no and yes.”
Although the insertion of the documentary footage does not turn the movie
RC into a documentary, we observe a kind of factuality mode which triggers
certain hypotheses about the link between invention and “real events.” If the
viewer also possesses information about the historical subtext of RC, these scenes
may unfold their full interpretative potential, inviting viewers to ask such ques-
tions as: Which parts of the movie are based on real events? Does the sequence of
events correspond to the real criminal activities of the “Banda della Magliana”?
Were they really involved in major political crimes such as the kidnapping of

2.3 Truth issues

These questions are indeed asked by readers and viewers. RC—novel, movie and
series—is perceived in terms of referentiality, i. e. as a useful source for under-
standing Italy’s recent past: “Romanzo Criminale has become—in common
knowledge—one of these works one relies on when it comes to discovering
hidden truths and refreshing the memory of certain historical arguments” (Nan-
nucci 2013).13 Here, truth is understood as something which is “really” there and
can be discovered. But this is only one of the many shades of meaning associated
with “truth” when it comes to RC. In an interview, De Cataldo also describes the
novel in terms of truth, as opposed to truthfulness:

It is a fatal error to consider Romanzo Criminale a story of the Magliana gang. […] There is
the need to draw a metaphorical and mythological line out of the bare facts and point to the
heart of a false story: a story which, therefore, is more true, and in any case more convincing
than the ‘official’ story. (D’Attis 2002)

12 Apparently, their interest is focused on the narrative potential of innovative formal tech-
niques such as an unsteady camera, blurred margins of the images, frequent low angle shots—
often in accordance with point-of-view shots—and jump cuts. Indeed, the series is frequently
cited as an example of innovative genre fiction in Italy (cf. Marino 2009).
13 For similar comments, see (Ac-
cessed 14 December 2014).
182   Silke Lahn and Stephanie Neu

Here, truth is understood in terms of plausibility—a point of view also shared by

critics who highlight the function of the novelist as a seeker of truth: “[T]he novel-
ist-poet chooses a different approach to effective reality than historiography; but
despite this, they are not less committed to the search for ‘truth’” (Sørensen 2010,
88). Truth is thus akin to filling in the gaps in the official history. This matches
one of the functions of fictionality as outlined above, namely being a means of
toying “with what is not the case and could never be the case, with what is not
the case but could be the case, with what should have been the case, and so on”
(Nielsen et al., 2015, 64).
What can be observed here is what Bøndergaard, referring to Walsh (2007,
51–52), calls an “exercise of our narrative—counterfactual—imagination” (2012,
63). As mentioned above, the “knowledge offered by fiction […] is not primarily
specific knowledge of what is (or was), but rather of how human affairs work, or,
more strictly, of how to make sense of them—logically, evaluatively, emotionally”
(Walsh 2007, 36). This, we would agree, also applies to RC. Its specificity, however,
is that it is not only a general “exercise” about how human affairs work but rather
refers to a specific real-world context: it offers reinterpretations of recent events
in Italian history which are probably counterfactual but nevertheless plausible.
In order to perceive RC as such, it is necessary to know about its reference to
real-life events: “[De Cataldo’s] aim has been to reactivate this knowledge, still
conserved within the collective memory; without this knowledge, Romanzo Crim-
inale would have remained incomprehensible, or it would have been something
completely different” (Sørensen 2010, 96). The “forensic” or “factual” elements
outlined above help to trigger this connection, suggesting a direct correspond-
ence between the story and real events. Thus, fictional and factual elements are
not contradictory but rather interact with and enhance each other.

3 Klovn: Feedback Loop Between Film Art and

Real World
Klovn is a Danish comedy series that aired on the pay-TV channel TV Zulu from
2005 to 2009. Each episode was about 25 minutes long, with sixty episodes in total
over six seasons. The show, widely regarded as the best Danish sitcom ever, was
written by the famous stand-up comedians Casper Christensen and Frank Hvam.
In 2010, a movie sequel was released which went on to become the most suc-
cessful film in Denmark of the previous ten years. Both the series and the movie
present the two scriptwriters as the main characters, using their reals names and
(ostensibly) playing themselves. Frank and Casper run a small film company and
 Towards a Crossing of the Divide between Fiction and Non-Fiction   183

share a great deal of their spare time together. The focus is generally on Frank,
who is generally depicted in everyday activities (in the office with Casper and
other colleagues, at home with his partner Mia, during leisure activities with Mia,
Casper, his girlfriend Iben and various acquaintances).
The comedy in both the series and the movie is built around awkward and
embarrassing situations, mainly arising from Frank’s selfishness and clumsiness.
Klovn is particularly discomfiting for the viewer because of its frequent “politi-
cal incorrectness” and inappropriate responses to sensitive topics: for example,
Frank frequently clashes with Muslim immigrants, fails to treat disabled persons
with consideration or allows things go awry because of his “gay panic” nature.
Canadian film critic Simon Howell justly describes the show as a “collusion
between three recent strains of popular comedy”: the “comedy of embarrass-
ment,” as found in Larry David’s Curb Your Enthusiasm; the Judd Apatow brand of
“manchild comedies”; and the “debauched bro humor” of The Hangover (Howell

3.1 Conflicting markers of factuality and fictionality

In Klovn, various signposts indicate a factual narration. Foremost amongst these

is the fact that the scriptwriters are identical to the performers who are (osten-
sibly) identical to the characters. This self-appearance can be understood as a
medial modified transfer of the canonical formula author = narrator = character
for literary autobiographical narration and thus as a sign for a referential pact
between author and reader (e. g., Genette 1990, 764), thus giving the impres-
sion that the series is a documentary. This impression is strengthened by the
fact that the dialogues are delivered very naturally14 and by the large number
of Danish celebrities who appear in person and with their real names as friends
and business partners of Frank and Casper (e. g., legendary soccer player Michael
Laudrup, award-winning director Bille August or famous television presenter Jarl
Friis-Mikkelsen). Moreover, Klovn gives the impression of being filmed exclusively
on location (in the real streets of Copenhagen, in real restaurants and offices).
The technical design of the series also resembles that of a documentary because
of the hand-held camera, imperfect light management or haziness when chang-
ing focus. This impression of reality is further underlined by the use of tracking
shots that follow a character into a building from outside without a cut whilst

14 From the DVD’s bonus material, the viewer learns that the dialogues are not fully scripted but
delivered spontaneously.
184   Silke Lahn and Stephanie Neu

disregarding the difficulty of varying light conditions; sometimes we can even

register that the camera light is switched on when the character enters a room
from the outside. In short, Klovn’s presentation corresponds to that of reality
television programmes featuring the domestic life of celebrities, the most iconic
example being The Osbournes (MTV, 2002–05) about the heavy metal singer Ozzy
Osbourne and his family.
Other elements in Klovn, however, do not follow the protocol of factual nar-
ration. For instance, the persons depicted are obviously not aware of the camera.
And when the angle changes in the next shot, no camera can be seen in the posi-
tion from which we saw the previous shot. Moreover, we are informed by the
opening credits that the series is “Based on real events,” meaning that the series
is based on real events but does not render real events. And in the end credits, we
learn that all characters are credited with one single actor—sometimes with the
same name on the left and on the right, sometimes not.
Due to the conflicting signals, the viewer may be uncertain as to whether s/he
is being presented with a fictional or a referential pact for the first few episodes.
But soon, s/he will decide on that Klovn’s story line is a thoroughly scripted nar-
rative, as each episode is a perfectly arranged and fully coherent whole which is
too coherent to be merely observed. In fact, Klovn displays a “setup-payoff joke
structure, in that every single character detail and comic possibility becomes a
Chekov’s Gun, waiting to be deployed for maximum effectiveness” (Howell 2011).
Now, how can we account for the fact that well-known people appear in a fic-
tional work ‘as themselves’ using their real names and, in so doing, are being
referential to the real world?

3.2 Self-appearance and autofiction

In a literary narrative, the author can use any real human that s/he so wishes.
In an audiovisual representation, however, the real live humans have to appear
‘as themselves’ in person before the camera. Thus, self-appearance is a genuine
filmic means which strengthens the impression of factual narration. In the case
of Klovn, we are (ostensibly) supposed to accept a referential-autobiographical
pact, since Frank and Casper are depicted with an abundance of real biographical
detail about Hvam and Christensen. Klovn-Casper’s girlfriend Iben, for instance,
is a popular Danish actress; she is played by Iben Hjejle, who was not only Chris-
tensen’s partner in real life during the filming but actually is an acclaimed actress.
And just as Hvam, Klovn-Frank was raised in Northern Jutland, far away from
Copenhagen—a fact that often serves as an explanation for his failings. Moreover,
 Towards a Crossing of the Divide between Fiction and Non-Fiction   185

Christensen’s and Hvam’s collaborations on earlier television series and Hjejle’s

film roles are frequently referred to in Klovn.
In literary theory, the combination of autobiography and fiction is known
as autofiction, an expression coined by French author Serge Doubrovsky in the
preface to his ‘novel’ Fils (1977) and introduced into academic research by Gérard
Genette in his essay “Fictional Narrative, Factual Narrative” (1990). Discussions
of the phenomenon can be divided into two main positions. First, autofiction is
defined as a special form of fictional narration (e. g., by Genette). But this still
leaves us with the problem, for example, of how to interpret the fact that the
author and one of the characters share the same name. Second, autofiction is
defined as the simultaneous combination of an autobiographical and a fictional
pact without offering the reader the chance to classify the text, in whole or in
part, according to one of the two pacts. Zipfel stresses that the last definition pre-
serves the pragmatic paradox of the autofictional claim “It is me and it is not me,”
thus maintaining the distinction between the two pacts. Zipfel assumes that the
receivers experience the specific aspects of both pacts, not by following both of
them at the same time, but by changing constantly from one pact to the other and
back again in the course of reading (2009, 302–306). Other receivers may feel that
both the referential and the fictional pact is broken and thus react with disap-
proval. This may be the reason why audience responses to an autofictional work
are usually very polarized.

3.3 Jacobsen’s term “fictiobiographism” in the field of audio-

visual works

The concept autofiction, which is of literary origin and based on the author entity,
can arguably cover the performance of Christensen and Hvam. However, it does
not cover the self-appearance of real persons in audiovisual representations who
are not the scriptwriters.
Because of this shortcoming, Louise Brix Jacobsen (also a member of the
Centre for Fictionality Studies at Aarhus University) introduced in 2008 the
Danish term fiktiobiografisme, which includes any kind of self-acting in audiovis-
ual narratives. Based on Walsh’s definition of fictionality as a rhetorical strategy,
Jacobsen’s concept of fictiobiographism highlights the idea that fictionality can be
used in all communication situations and in all media. In the light of series such
as Klovn and its source of inspiration Curb Your Enthusiasm, she understands
fictionality as a “performative self-fashioning strategy” (2012, 13 and 305) and
stresses its special potential in audiovisual works (14). According to Jacobsen,
self-acting results in “biographical undecidability” (2012, 12), as the ‘self-actors’
186   Silke Lahn and Stephanie Neu

both are and are not the same person at the same time. Such works are “biogra-
phy and not biography at all. Both fiction and nevertheless not fiction” (2008, 1).
Still, she insists (like Walsh and Nielsen) on the existence of a borderline between
reality and fiction, even though it cannot be definitively localized. Faced with
this situation, it is the declared aim of Jacobsen’s concept of fictiobiographism to
problematize and challenge this borderline.15
Jacobsen points out that rumours form a central ingredient of many so-called
fictiobiographical works. Gossip in tabloids or social media about real celebrities
(e. g., rumours about unfaithfulness or homosexuality) is taken up, (ostensibly)
confirmed and amplified by them ‘as themselves’ in the series.16 Following the
broadcast, the rumours as enriched in the narrative may, in turn, be picked up
again and further propagated by other media (Jacobsen 2008, 4). By this, a series
like Klovn can initiate a “feedback loop” (Walsh 2007, 128) so that, in the end, it
may be hard to say whether a rumour originated from a real incident or a fictive
presentation. Hence, Jacobsen defines fictiobiographism as intermedially based
and even as transmedial (Jacobsen 2012, 14 and 119).
And what is in it for a self-acting VIP affected by slander? Celebrities appear-
ing ‘as themselves’ in a fictiobiographical work are provided with an opportunity
to playfully deal with these rumours in a fictional context, and thus a charm-
ing effect of self-mockery, since rumours are deliberately seized upon and even
amplified, or because character traits are portrayed in an exaggerated way.
For the case of Klovn, we can state that the viewers are not changing constantly
between the referential and the fictional pact (as Zipfel assumes with regard to lit-
erary narratives), but that both contracts are fully present at the same time. And it
seems that this applies not only for the narrative but for Hvam’s and Christensen’s

15 For a detailed delineation of the term fictiobiographism, see Jacobsen’s Ph. D. thesis (2012, 33–
37). Reflecting the fact that biographism is an expression with a negative connotation, Jacobsen
proposes to expand its meaning: “Biographism becomes instead an expression for a trend where
it is common and partly quite accepted to speculate about the private lives of self-appearing
persons” (35). In her view, the spread of the phenomenon indicates a general artistic tendency
across media and art, which is why she adopts the suffix -ism (4). We question whether the in-
tended revaluation of the pejorative expression biographism can be successful, or even desirable,
although we very much approve of the concept itself. Our second objection to Jacobsen’s term is
the fact that a series like Klovn does not aim to portray its characters’ whole life story, or even an
essential part of it, as the word part ‘biographi-’ would imply.
After completion of our article, Jacobsen published a contribution in English, in which she re-
places fictiobiographism with vitafication after “sparring” with international colleagues “in the
process of developing a term more suitable in English” (2015, 266).
16 In Bøndergaard’s words, this makes the rumour and the performer a “forensic” element in a
fictional surrounding (2012).
 Towards a Crossing of the Divide between Fiction and Non-Fiction   187

non-fictional media appearances as well. In real interviews, for instance, it can

be hard to decide whether the journalist is talking to the Klovn-characters or to
Hvam and Christensen as they frequently and with relish use opportunities like
these to challenge the divide between fiction and non-fiction from reality’s side
as well. The result of the series’ specific autofictional or fictiobiographical nature
and the actors’ behaviour outside the show fully blur the borderline between
Christensen’s and Hvam’s real lives, on the one hand, and Klovn-Casper’s and
Klovn-Frank’s fictional lives, on the other.
To give an example of how fictionality can even invade reality itself and cause
undecidable situations: at the Skanderborg music festival in 2010, an apparently
confused Frank Hvam dressed only in his dirty underwear suddenly entered the
stage and pushed the musician Thomas Helmig aside, who was in the middle of
his famous song “Stupid Man.” Frank Hvam managed to yell desperately into
the microphone for some Bo, crying out hysterically that this Bo had misunder-
stood everything. Two roadies then forced him off the stage. Helmig commented
“Stupid man indeed” and started singing the song again.17 To understand this
situation adequately requires the contextual knowledge that this real festival
setting provided the background for a scene in the film sequel of the series,
Klovn—The Movie, which was released later in 2010. But at the time, the Skan-
derborg audience only saw a seedy man acting extravagantly. And what about
Thomas Helmig? Was he himself in that situation, or was he giving a fictionalized
version of himself? This remains an unsolvable question.

3.4 Functions of fictionality in Klovn

In Fiktionalitet, Jacobsen et al. (2013, 71–72) identify three functions of “fiction-

alization” on the side of the sender: 1) self-fashioning strategy; 2) a critique of
contemporary society; 3) genre creation and genre critique. Here, we have mainly
focused on the first-mentioned strategy but are convinced that Klovn deploys all
Especially in Klovn, the viewer can identify another function of fictionality on
the side of the sender. The show may be understood as a sacrifice on the altar of
the Law of Jante, a well-known list of rules in Scandinavia with its central com-
mandment: Do not think you are in any way special. As there is no denying that
Christensen and Hvam have become very special in Denmark, their self-appear-

17 The scene can be looked up online at (Accessed

14 December 2014).
188   Silke Lahn and Stephanie Neu

ance in the series and the foibles of their ‘homonym’ Klovn-characters are not
only to be received as self-deprecating humor and self-mockery but also as a form
of atonement in order to make themselves and their success a little less ‘special’
A function of fictionality on the side of the receiver in Klovn might possibly be
termed an “exercise,” extending a concept originally introduced by Walsh (2007,
50–51): the viewer can safely experience all manner of what appear to be real
embarrassing situations without being involved in them personally and learn
that selfishness and narrow-mindedness do not pay, as illustrated by Frank, who
is repeatedly punished for his failings. In this light, Klovn conveys a highly ethical

4 Conclusion
Hybrid forms such as Romanzo Criminale and Klovn can be defined as transmodal
in that they employ patterns of both fictionality and factuality, creating new
forms for which the terms fiction and non-fiction seem inadequate. Against this
background, it is important not only to outline the rhetorical quality of fictionality
(on which literary theory has hitherto focused) but also to define the rhetorical
quality of factuality. In our view, both fictionality and factuality can be seen as
“invitations” to entertain “certain interpretative operations” (Nielsen 2010, 282).
The two invitations can be accepted together but also separately: in this case, the
perception of the work is altered but not compromised.
The receiver of satirical and ironic representations who accepts both invita-
tions may have difficulties identifying the forensic elements and understanding
the line of attack; s/he may have to do considerable research in order to ferret
out all the work’s ideas and allusions to reality, especially when the distance
between the events portrayed in the work and the receiver is more than a gener-
ation. Klovn manages to avoid this dilemma. A self-mocking self-appearance of
a VIP which plays on real gossip or biographical data may provide extra amuse-
ment for those in the know. However, most of Klovn’s story lines are universally
understood because they deal with general human foibles. While RC—both the
television series and the movie—can also be enjoyed as an entertaining piece of
crime fiction, the integration of forensic material invites the receiver to consider
RC in light of the historical facts alluded to in the story. This is certainly an extra
effect which helps to explain why (especially in Italy) RC has been perceived as
a truthful statement about that country’s recent history. What comes into play
in both RC and Klovn is the understanding of fictionality as “the inaugurating
 Towards a Crossing of the Divide between Fiction and Non-Fiction   189

move of a specific rhetoric, which enables a process of imaginative exploration of

values” (Walsh 2007, 168) and—especially RC—the fulfilment of a function asso-
ciated by Walsh with factual narratives: the “factual enrichment of the reader’s
cognitive environment” (31).
In summary: with their proposal of fictionalization as a strategy for both the
sender and the receiver, Jacobsen et al. (2013) seek to reconcile apparent con-
trasts and provide an alternative to the either-or-question “fiction or non-fic-
tion.” Their approach is not only a theoretical elaboration within a narratological
framework; it also offers a set of heuristic tools applicable to a variety of texts,
as demonstrated in our two examples. The receiver is thereby offered a strategy
that undermines the categorization of television series and movies like Klovn as
fiction or non-fiction (Jacobsen et al. 2013, 72). Moreover, their approach allows
one to take into account the intuitive distinction between fiction and non-fiction
we all experience everyday,18 even though we are hard put to substantiate this
distinction with a syntactic or a semantic approach, or any other approach.
With this in mind, we wish to conclude with three further remarks regarding
possible future developments in the field of fictionality and factuality studies.
First, we argue in favour of expanding the “techniques of fictionality” by
including elements of the narrative discourse. The representation of characters,
as we have shown here with our two audiovisual examples, can be shaped, to a
certain extent, by genre conventions which are perceived conventionally as fic-
tional. The employment of certain techniques in the representation of characters
can thus function as a “technique of fictionality,” too.
Our second point refers to the declared execution of the narrator. Here,
Nielsen (2011, 119) agrees with Walsh, stating that the narrator is a theoretical
fallacy. In his view, understanding the author’s activity as fictionalizing and
inventionality makes the narrator dispensable: s/he loses all raison d’être because
this concept replicates the author unnecessarily. Accordingly, Nielsen concludes:
“The rebirth of the author and of fiction must therefore be paid for with the nar-
rator’s life” (2011, 119). For a more or less classical narratologist, this is quite a
bullet to bite. Nielsen may have a point in challenging the idea of a narrator when
it comes to borderline cases such as Scott Fitzgerald’s The Great Gatsby, “where
something is narrated in great detail from an episode where the narrator himself
was absent” (Nielsen 2010, 278). But to dismiss the narrator completely because
of this occasional problem—which could probably be solved within narratologi-
cal models (e. g., via the concept of unreliable narration)—would mean to suspend

18 Cf. Nielsen (2011, 114): “It is as difficult to keep an idea of difference as it is impossible to
eliminate ideas like these.”
190   Silke Lahn and Stephanie Neu

a valuable heuristic tool. Instead, we propose opting for a more balanced solu-
tion, subscribing not to the elimination of the narrator in favour of the author but
maintaining that both entities must be taken into account according to the text
at hand. Although we distance ourselves from biographism, we are well aware
that an author—understood as “production entity”—may express his views (for
example, in interviews), which can then be perceived as “ancillary texts” (Walsh
2007, 45) of the work in question, thus influencing the perception of a narrative
as fictional or not.19
Our third remark concerns possible counterparts of fictionality and fiction-
alization. With hybrid forms such as RC and Klovn, we are confronted with a sit-
uation that has yet to be squarely met: if, as mentioned above, fictionalization is
understood as a signal that the narrated events do not represent reality but rather
describes the possible, the non-existing, then what about the opposite case in
which it is implied that the narrated does represent reality and the existing? We
have tried to take this into account by using the counter-expressions to factualize
and factualization. In our view, the idea of a “rhetoric of factuality” as a counter-
part to the “rhetoric of fictionality” deserves to be taken further.

Works Cited
Primary sources
De Cataldo, Giancarlo. 2002. Romanzo Criminale. Torino: Einaudi. [novel]
Klovn. Broadcast 2005–09 by TV2 Zulu. Denmark. Directed by Mikkel Nørgaard and Niels
Gråbøl. Written by Casper Christensen and Frank Hvam. [television series]
Klovn—The Movie. 2010. Directed by Mikkel Nørgaard. Written by Casper Christensen and Frank
Hvam. Denmark: Zentropa Productions, TV2 Zulu. [movie]
Romanzo Criminale. 2005. Directed by Michele Placido. Written by Giancarlo De Cataldo,
Sandro Petraglia and Stefano Rulli. Italy: Cattleya, Warner Bros et al. [movie]
Romanzo Criminale. Broadcast 2008–10 by Sky Cinema. Italy. Directed by Stefano Sollima.
Written by Leonardo Valenti, Barbara Petronio and Daniele Cesarano. [television series]

19 Speaking about the author’s ethos, in her seminal work Ethos and Narrative Interpretation:
The Negotiation of Values in Fiction (2014), Liesbeth Korthals Altes outlines a promising way of
how to take into account authorial statements for the interpretation of fictional texts.
 Towards a Crossing of the Divide between Fiction and Non-Fiction   191

Secondary sources
Bøndergaard, Johanne Helbo. 2012. Forensic Literature. Evidence and Fictionality in
Contemporary Memory Narratives. MA thesis. Aarhus: Aarhus universitet.
Booth, Wayne C. 1983 [1961]. The Rhetoric of Fiction, 2nd edition. Chicago: The University of
Chicago Press.
D’Attis, Nino G. 2014. “Intervista a Giancarlo De Cataldo.”
Intervista_a_Giancarlo_De_Cataldo.htm (Accessed 14 December 2014)
Doležel, Lubomír. 1988. “Mimesis and Possible Worlds.” Poetics Today 9: 475–496.
Fludernik, Monika. 1996. Towards a ‘Natural’ Narratology. London and New York: Routledge.
Genette, Gérard. 1990. “Fictional Narrative, Factual Narrative.” Poetics Today 11 (4): 755–774.
Translated by Nitsa Ben-Ari and Brian McHale.
Gjerlevsen, Simona Zetterberg. 2013. “En seriøs sprogteoretisk fiktionalitetsdefinition.” In
Hansen et al., eds., 57–68.
Gjerlevsen, Simona Zetterberg. 2016. “Fictionality.” In living handbook of narratology, edited
by Peter Hühn, Jan Christoph Meister, John Pier and Wolf Schmid: http://www.lhn. (Accessed 22 June 2016)
Hamburger, Käte. 1973 [1957/1968]. The Logic of Literature. Translated from the 2nd revised
edition by Marilynn J. Rose. Bloomington and Indianapolis: Indiana University Press.
Hansen, Per Krogh, Henrik Skov Nielsen and Stefan Kjerkegaard, eds. 2013. “Fiktion og
fortælling” [Fiction and narrative]. Special issue of K&K. Kultur og klasse 115.
Herman, David. 2009. Basic Elements of Narrative. Malden, MA, etc.: Wiley-Blackwell.
Howell, Simon. 2011. “Fantasia 2011. Klovn offers debauched manchild comedy,
Scandinavian-style.” 11 July.
debauched-manchild-comedy-scandinavian-style/(Accessed 14 December 2014)
Jacobsen, Louise Brix. 2008. “Klovnen og rygtet.” 16:9, no. 28. September. http://www.16-9.
dk/2008-09/side04_feature1.htm (Accessed 14 December 2014)
Jacobsen, Louise Brix. 2012. Fiktiobiografisme. Fiktionalisering som performativ strategi i dansk
film og tv fra 2005 og frem. Ph. D. thesis. Aarhus: Aarhus universitet.
Jacobsen, Louise Brix. 2015. “Vitafiction as a Mode of Self-Fashioning: The Case of Michael J.
Fox in Curb Your Enthusiasm.” Narrative 23 (3): 252–270.
Jacobsen, Louise Brix, Stefan Kjerkegaard, Rikke Andersen Kraglund, Henrik Skov Nielsen,
Camilla Møhring Reestorff and Carsten Stage. 2013. Fiktionalitet. Frederiksberg:
Korthals Altes, Liesbeth. 2014. Ethos and Narrative Interpretation: The Negotiation of Values in
Fiction. Lincoln and London: University of Nebraska Press.
Marino, Marco. 2009. “Romanzo Criminale La Serie—Stagione 1.” 19 February. http://serial.
(Accessed 14 December 2014)
Nannucci, Teresa. 2013. “Doppio schermo. Romanzo criminale, dal film alla serie.” 5 March.
filmromanzo-criminale-la-serie/ (Accessed 14 December 2014)
Nielsen, Henrik Skov. 2010. “Natural Authors, Unnatural Narration.” In Post-Classical
Narratology: Approaches and Analyses, edited by Monika Fludernik and Jan Alber,
275–301. Columbus: The Ohio State University Press.
Nielsen, Henrik Skov. 2011. “Fiktion og fortælling. Tiltrædelsesforelæsning.” Kritik 201:
192   Silke Lahn and Stephanie Neu

Nielsen, Henrik Skov. 2013. “Fiktionalitetens Retorik.” Reception 71: 78–90.

Nielsen, Henrik Skov, James Phelan and Richard Walsh. 2015. “Ten Theses about Fictionality.”
Narrative 23 (1): 61–73.
Reitan, Rolf. 2011. “The Rhetoric of Fictionality: Intoduktion til Richard Walsh.” In R. R., Poul
Behrendt and Marianne Barlyng, eds. “Fiktionens forandringer” [The changes of fiction].
Special issue of Spring 31/32: 22–56.
Reitan, Rolf. 2013. “Walsh, kort fortalt.” In Hansen et al., eds., 33–56.
Schaeffer, Jean-Marie. 2014 [2009]. “Fictional vs. Factual Narration.” In Handbook of
Narratology, edited by Peter Hühn, Jan Christoph Meister, John Pier and Wolf Schmid, vol.
2: 179–196. Berlin and Boston: De Gruyter.
Also available in: living handbook of narratology:
fictional-vs-factual-narration (Revised 20 January 2013)
Schmid, Wolf. 2010 [2005]. Narratology. An Introduction. Translated by Alexander Starritt.
Berlin and New York: De Gruyter.
Searle, John R. 1975. “The Logical Status of Fictional Discourse.” New Literary History 6:
Sørensen, Gert. 2010. “Letteratura noir e storiografia. Le voci del doppio Stato in Romanzo
criminale di De Cataldo.” In Memoria in “Noir.” Un’indagine pluridisciplinare, edited by
Monica Jansen and Yasmina Khamal, 87–106. Frankfurt am Main et al.: Peter Lang.
Walsh, Richard. 2007. The Rhetoric of Fictionality. Narrative Theory and the Idea of Fiction.
Columbus: The Ohio State University Press.
Walton, Kendall. 1990. Mimesis as Make-Believe. Cambridge, MA: Harvard University Press.
White, Hayden. 1978. “The Historical Text as Literary Artifact.” In H. W. Tropics of Discourse.
Essays in Cultural Criticism, 81–100. Baltimore: Johns Hopkins University Press.
Zipfel, Frank. 2001. Fiktion, Fiktivität, Fiktionalität. Analysen zur Fiktion in der Literatur und zum
Fiktionsbegriff in der Literaturwissenschaft. Berlin: Erich Schmidt.
Zipfel, Frank. 2009. “Autofiktion. Zwischen den Grenzen von Faktualität, Fiktionalität und
Literarität?” In Grenzen der Literatur. Zu Begriff und Phänomen des Literarischen, edited
by Simone Winko, Fotis Jannidis and Gerhard Lauer, 285–314. Berlin and New York: De
Brian Richardson (College Park, Maryland)
Unnatural Narrative Theory:
A Paradoxical Paradigm
I will begin with some necessary definitions: what I call mimetic narratives are
those works of fiction that model themselves on or substantially resemble non-
fictional works. Mimetic works systematically attempt to depict the world of our
experience in a recognizable manner; this is the traditional goal of works that
strive for realism or verisimilitude. Nineteenth-century realist fiction is a major
subspecies of the mimetic tradition. I define an unnatural narrative as one that
contains significant antimimetic events, characters, settings, or frames. By
antimimetic, I mean representations that contravene the presuppositions of non-
fictional narratives, violate mimetic expectations and the practices of realism,
and defy the conventions of existing, established genres.
There are many models of narrative theory, classical and postclassical, being
employed today. A number of them come readily to mind: rhetorical, structur-
alist, Stanzelian, Tel Aviv school, cognitivist, Hamburg school and so forth.
It is quickly apparent, however, that nearly all current models are mimetically
centered, that they in fact are part of a larger paradigm that privileges mimetic
constructs and/or conversational natural narrative models and views fictional
constructs as generally or typically having a fundamental similarity to analogous
constructs in nonfictional narratives. That is, most narrative theories begin by
treating fictional narrators as if they were like human storytellers; fictional plots
like events we might experience in our lives; fictional temporality like time in the
real world. More generally, narrative theory, classical and postclassical, presup-
poses a stable, consistent identity in all its forms; most narrative theories don’t
discuss spaces that are impossible, causality that moves backward, narrators
who are not single, consistent personages, or characters whose selves split off
or merge with other characters. And yet such fictional constructs permeate post-
modern fiction and drama.
I will argue instead for the importance of the unnatural narrative model as an
essential corrective to the limitations, omissions, and blind spots of mimetic the-
ories.1 In doing so, I will not claim that ours is a totally new paradigm that should

1 To reiterate: in this essay, I will be using the terms “mimetic” and “antimimetic.” Mimetic
works systematically attempt to depict the world of our experience in a recognizable manner;
this is the traditional goal of works that strive for realism or verisimilitude. Nineteenth-century
realist fiction is a major subspecies of the mimetic tradition. An antimimetic (or antirealist) work

DOI 10.1515/9783110555158-010
194   Brian Richardson

replace older ones. I will argue that instead of supplanting them, our model
should instead supplement them. Mimetic theories have done excellent work but
have not exhausted all the important narratives of the world and the narrato-
logical concepts necessary to encompass them. Thus, in the case of narration or
temporality, one may retain Genette’s ultimately mimetic accounts, but they must
be augmented by new categories that can include the many important texts that
elude or defy Genette’s model. Unnatural narratology thus offers a much more
expansive vision of the field of narrative studies, but it may only require a par-
tially new paradigm.
Let us examine an example of an unnatural text. I offer the following passage
from Mark Leyner’s My Cousin, My Gastroenterologist:

He’s got a car bomb. He puts the keys in the ignition and turns it—the car blows up. He gets
out. He opens the hood and makes a cursory inspection. He closes the hood and gets back
in. He turns the key in the ignition. The car blows up. He gets out and slams the door shut
disgustedly. He kicks the tire; he takes off his jacket and shimmies under the chassis. He
pokes around. He slides back out and wipes the grease off his shirt. He puts his jacket back
on. He gets in. He turns the key in the ignition. The car blows up, sending debris into the air
and shattering windows for blocks. He gets out and says, Damn it! (1995, 59)

It quickly becomes evident that it is useless to apply a mimetic framework to this

text. One cannot be blown up by a car bomb and then offhandedly inspect the
damage to a car that has just been destroyed. There is no script, code, or model
from our experience that corresponds to being blown up several times. This is
not a mimetic representation of a possible event that we could experience in the
world we inhabit; it is in fact a parody of such an occurrence.2 It is obvious that a
very different kind of play with representation is occurring, and it is antimimetic.
This requires us to jettison mimetic expectations—such events can only happen
in fiction, never in nonfiction—and then to be receptive to the new, unexpected
pattern that might be emerging, whether it is a parodic, abstract, or follows some
other design. The fabula for such an event would have to state that the car was

like Beckett’s Molloy defies the conventions of mimetic (or realist) representation that are ad-
hered to in a work like Anna Karenina.
2 It is true that in this case we need to be aware of the mimetic model to realize that it is being
violated, as well as to derive humor from the divergence. However, the model is not the same as
its violation, even if we need the model to perceive the violation. In other cases, however, like the
novels of Robbe-Grillet, humorous or parodic intentions are by no means always evident; the an-
timimetic elements may seem instead like mistakes or deliberately contrived nonsense. I discuss
the status of impossible fictional worlds in the second chapter of my book, Unnatural Narrative:
Extreme Narration in Modern and Contemporary Fiction (2006).
 Unnatural Narrative Theory: A Paradoxical Paradigm   195

blown up and at the same time it was not blown up, that there were in fact several
contradictory fabulas—something only unnatural narratology is prepared to do.
As readers, we will abandon our sense of probability and may focus instead on
the numerous impossible repetitions of the explosion. Even if we shift around
looking for a non-literal way to read the material (is this an allegory of the blindly
self-destructive nature of Western technological society? is it a parable of the total
self-absorption of an egoistic individual?), we appreciate the work as one that
cannot be recuperated by conventional strategies of representation.
Our experience of starting cars or visiting service stations will only help us up
to a point here, and no story we have ever heard about either is likely to allow us
to understand these events (no one is likely to say, “Oh, yeah—I heard about cars
like that in Nevada!”). Nor will any story about car bombs help much, either: this
narrative is following a different trajectory, creating an impossible blend. One
suspects that its author delights in this new path and is presumably indifferent
to the feelings of confusion or resentment it might cause in more traditional, lit-
eral-minded, or insistently mimetic readers. More adventurous readers, however,
follow along as they carefully observe how many repetitions there will be, how
they will be varied, and how far they will extend: there would seem to be a rule
to the effect that unnatural events generally progress to ever greater levels of
extravagance and impossibility. We can even try to apply blending theory to these
events and suggest that Leyner is taking a common situation, turning the key in
the ignition when the battery is dead and nothing happens, and merging it with
the video thriller scenario of the Hollywood car bomb scene. Though this account
may explain some features of the text, it misses many of the most important ones;
it does not allow us to predict what will happen next, what the consequences will
be, or what other strange new blends might emerge.
What does a single-model narratology do with a text like Leyner’s that resists
the ordinary conventions of mimetic representation? In almost all cases, it mar-
ginalizes, ignores, or denies this kind of text. An antimimetic narrative theory
is essential to supplement existing, monomorphic paradigms. We can see this
clearly in a number of areas, many of which center on the representation of a
sequence of related events. To demonstrate the importance of the unnatural para-
digm, I will offer an analysis of the theory and practice of the fabula, emphasizing
unusual, unnatural and impossible narrative sequences.
Fabula, or the story that we infer from a text or performance, is one of the
foundational ideas of narrative theory. This concept, established by the Russian
formalists, has been around for nearly a century and is referred to in a variety
of ways, including the French structuralist term histoire. Meir Sternberg has
indicated its importance for narrative theory, asserting that “actional discourse,
whether literary or historical or cinematic, presupposes temporal extension
196   Brian Richardson

[which] provides a natural principle of coherence, one that enables the narrator
to construct his presentational sequence, […] according to the logic of progres-
sion inherent in the line or chain of events themselves; from earlier to later and
from cause to effect” (1981, 60–61).
As his metaphors of line and chain indicate, Sternberg here reveals himself
to be trapped by mimetic presuppositions. As Monika Fludernik has pointed out,
“the story vs discourse opposition seems to repose on a realist understanding of
narrative” (1996, 334). A consistent fabula can indeed be derived from most every
nonfictional or conversational natural narrative, as well as the mimetic or realist
works of fiction that strive to resemble these discourse types. There remain,
however, a number of varieties of unnatural fabulas that elude the mimetic model
which narrative theory needs to account for. Looking at these can help us for-
mulate a revised concept of the fabula and thereby disclose what the unnatural
paradigm can provide.
Some texts, like Leyner’s piece, have several contradictory sequences of
events (e. g., Robbe-Grillet’s La Jalousie 1957; Harold Pinter’s The Basement 1966;
Anna Kavan’s Ice 1967; Robert Pinget’s Passacaille 1969; Robert Coover’s “The
Babysitter” 1969; Caryl Churchill’s Traps 1977). Some of the different, incompat-
ible endings all present in Coover’s novella include the following: the babysitter
accidently drowns the baby; the husband who hired her comes back early to have
sex with her; a shy neighborhood boy visits her; two neighborhood boys encoun-
ter the father with the babysitter; the father surprises the boys with the babysitter;
the babysitter chases off the boys; the babysitter is raped and accidently killed by
the boys; the family returns to find all is well; and the mother learns that the chil-
dren have been murdered, her husband is gone, there is a corpse in the bathtub,
and her house has been destroyed.
Churchill goes to some length to clarify the antimimetic nature of her drama
and simultaneously to ensure that her work will retain its unnaturalness and not
be simply reduced to a conventionalizing interpretive strategy. She affirms in her
preface that Traps:

is like an impossible object, or a painting by Escher, where the objects can exist like that
on paper, but would be impossible in life. In the play, the time, the place, the characters’
motives and relationships cannot all be reconciled—they can happen on stage, but there is
no other reality for them. […] the characters can be thought of as living many of their pos-
sibilities at once. There is no flashback, no fantasy, everything that happens is as real and
solid as everything else within the play. (1985, 71)

The description Churchill provides of her play could well be a guide for the inter-
pretation of the more extreme kind of unnatural narratives.
 Unnatural Narrative Theory: A Paradoxical Paradigm   197

Ursula Heise has observed that such narratives “project into the narrative
present and past an experience of time which normally is only available for the
future: time dividing and subdividing, bifurcating and branching off continu-
ously into multiple possibilities and alternatives” (1997, 55). Instead of one event
precluding several other possible options, many incompatible possibilities can
be seen to have been actualized. In none of the examples noted in this section
can one easily extract a single, consistent story from a fixed sjuzhet the way one
might in any natural or realistic narrative.3 Alain Robbe-Grillet, referring to the
contradictory fabula in La Jalousie, stated: “It was absurd to propose that in the
novel […] there existed a clear and unambiguous order of events, one which was
not that of the sentences of the book, as if I had diverted myself by mixing up a
pre-established calendar the way one shuffles a deck of cards” (1965 [1963], 154).
He went on to state that for him there existed no possible order outside of that
found within the pages themselves. This text does not mimic realistic narratives
whose sjuzhets will divulge a single fabula; here one has only an indeterminate,
contradictory fabula—one that narrative theory needs to recognize and investi-
A narrative can circle back on itself as the last sentence becomes the first
sentence and thus continues for eternity, its fabula infinite (e. g., Joyce’s Finneg-
ans Wake 1939; Nabokov’s “The Circle” 1936; Beckett’s Play 1963). In the case of
what I call denarration (Richardson 2005 and 2006, 87–94), the discourse erases
the fabula as it is being articulated. At the beginning of Robbe-Grillet’s In the
Labyrinth (1965 [1959]) we are told that “outside it is raining […] the wind blows
between the bare black branches” (141); in the next sentence this setting is denar-
rated as we are informed instead that “outside the sun is shining: there is no tree,
no bush to cast a shadow” (141). Inside the room, there is fine dust that coats
every surface; this dust in turn generates what will become the definitive weather
beyond the walls of the house: “Outside it is snowing” (142).
In other works, time passes at different speeds for different groups of people.
Thus, in Shakespeare’s A Midsummer Night’s Dream, four days pass for the nobles

3 For additional discussion of many of these forms, see my articles “Beyond Story and Dis-
course” (2002) and “Unnatural Stories and Sequences” (2013).
4 It should be obvious that for me, a narrative does not have to be self-consistent; such a re-
quirement is, from my perspective, just another mimetic prejudice. Some time ago Gerald Prince
suggested that La Jalousie was not a narrative: it “is a novel, of course, but a pseudo-narrative
one” (1982, 65). Such an argument sidesteps the debate over narrative (why can’t a narrative
include contradictory events?), even as it neglects to explore the definition of a novel. What is
a novel but an extended fictional narrative in prose? How can a work easily qualify as a novel
without also being a narrative?
198   Brian Richardson

in the orderly city while—at the same time—two days pass in the enchanted forest
(cf. Richardson 1987). In Virginia Woolf’s Orlando (1928), twenty years pass for
the protagonist while three-and-a-half centuries pass for those around him (her);
similarly, in Caryl Churchill’s play, Cloud Nine (1979), twenty-five years pass for
the characters while a full century passes for the rest of the world. These cases
result in a dual or multiple fabulas, a possibility unimagined by most narratolo-
Another experimental technique employs features of the discourse to create
the fabula. These can be termed textual generators. We see them at work in the
passage from In the Labyrinth just cited. Exactly as the dust in the generative
room determines that snow will be the weather outside, other surface images on
the inside of the house generate additional objects in the storyworld: the impres-
sion of a letter opener becomes a soldier’s bayonet; a rectangular impression pro-
duces the mysterious box that the soldier carries; a desk lamp gives rise to a street
lamp outside in the snow, which in turn yields up a soldier leaning against it,
clutching the box; and a realistic painting, “The Defeat at Reichenfels,” literally
brings to life the military events it depicts. The descriptions here bring into being
the events they suggest as the discourse creates the fabula.
In other works the fabula is variable. In “choose-your-own-story” texts such
as Raymond Queneau’s 1967 proto-hyperfiction, “Un conte à votre façon,” the
reader is offered a series of options to choose from; the fabula is multilinear and
variable, though once a particular event is selected, it becomes fixed; this is the
principle around which many hyperfictions are constructed. Ana Castillo’s The
Mixquiahuala Letters (1986) operates along similar lines. The book is composed
of a series of letters sent by one of the characters, but not all are intended to
be apprehended by the reader. Instead, the author offers three different reading
sequences depending on the reader’s sensibility. Thus, the conformist is told to
begin with letters 2 and 3 and then to go to number 6, while the cynic is to start
with letters 3 and 4 before going on to number 6. The quixotic reader is offered yet
another different sequence: 2, 3, 4, 5, 6. It is important to note that each sequence
produces a different fabula, and recognize in our definitions that this is also an
option that writers may turn to.
Related to these works is the multiple ending that offers several possible
conclusions. Malcolm Bradbury’s “Composition” (1993 [1976]) tells the story of a
new teaching assistant at a Midwestern university during the Vietnam War. After
completing his course on composition (but before turning in the final grades),
he is invited to party with two of his female students. The evening itself is fairly
innocent, though some extremely compromising photos are taken. The next
morning, the instructor receives a sample Polaroid and a request for a higher
grade for another student who has neglected to turn in a composition in order
 Unnatural Narrative Theory: A Paradoxical Paradigm   199

to more fully engage in political struggles; he has to decide what to do, knowing
that if the pictures get circulated he is sure to lose his position. The earlier sec-
tions of the work are numbered 1 through 4; the final section offers three different
resolutions, designated 5A, 5B, and 5C. In the first option, the instructor quietly
raises the grade and saves his job. In the second, he corrects the grammar of the
letter, sends it back to the blackmailers and defiantly turns in the correct grade.
In the third, he agrees with the student that grades are crap and all words are
inadequate; he destroys the grade sheet and abandons all academic drudgery in
order to move on and devote himself to life and love. The text offers no indication
of which of these possibilities will (or has been) be actualized; each option has
a certain plausibility. I don’t see this as a hermeneutic test in which the reader
needs to determine which is the most likely decision as much as the demonstra-
tion of a series of options that the reader is implicitly invited to choose from. As
the instructor is informed by one of the other characters, “You have to write your
own ending” (141). Here we have a fabula that forks into multiple incompatible
directions at its end.
In the texts I have just been considering, we have a partially variable sjuzhet
that, once selected, produces different fabulas. It might be objected that after
the choices have been made, each resulting narrative is entirely consistent, even
mimetic. Porter Abbott explains, however, that narrative “is something that
always seems” to come after the events it depicts; “to be a re-presentation” of
them (2008 [2002], 36); it is the violation of this sense of the pastness of the nar-
rative events that is foregrounded by such multilinear fabulas—if the events actu-
ally did occur in the storyworld’s past, they could not be altered.
These examples clearly show the need for an expanded framework to account
for other kinds of unnatural stories, including infinite fabulas; dual or multiple
storylines with inconsistent chronologies; inherently vague and unknowable
fabulas; internally contradictory fabulas; denarrated fabulas; and repeated, mul-
tiple versions of the same essential story. Elsewhere in The Cambridge Introduc-
tion to Narrative, Abbott writes: “narrative discourse is infinitely malleable […],
but as we take in information from the discourse we sort it in our minds, recon-
structing an order of events we call the story [fabula]. Insofar as it is a story, it
has its own length of time and an order of events that proceeds chronologically
from the earliest to the latest” (17). It should be readily apparent that this account
is far too limited; it and other standard definitions of fabula need to be revised,
updated, and reconceived. Unnatural narrative theory is designed to account for
such inventive cases, analyze their own unexpected trajectories, and provide a
model that allows them to be understood theoretically.
The mimetic/antimimetic opposition that I have been foregrounding in this
paper has existed for some time and has been explicitly discussed. We recall the
200   Brian Richardson

deep concern expressed by Henry James when he stated in an essay on Trollope

that a novelist should “regard himself as a historian and his narrative as a history.
[…] He must relate events that are assumed to be real” (1972 [1883], 175). This is an
excellent description of the goals of mimetic fiction, for James’s insistence on this
poetics exemplifies the mimetic prejudice that has dogged literary and narrative
theory since the Renaissance. James denounces all refusals to assume the pose of
verisimilitude; he complains:

Certain accomplished novelists have a habit of giving themselves away which must often
bring tears to the eyes of people who take their fiction seriously. I was lately struck, in
reading over many pages of Anthony Trollope, with his want of discretion in this particu-
lar. In a digression, a parenthesis or an aside, he concedes to the reader that he and his
trusting friend are only “making believe.” He admits that the events he narrates have not
really happened, and that he can give his narrative any turn the reader may like best. Such
a betrayal of a sacred office seems to me, I confess, a terrible crime, and it shocks me every
whit as much in Trollope as it would have shocked me in Gibbon or Macaulay. (James 1972
[1884], 30–31)

There are many ironies here, the most evident of which is that author and reader
actually are making believe and the narrative can be given any turn that is desired.
James claims to be shocked by a “terrible crime” that is both common and almost
as old as fiction itself. Furthermore, James was more expansive and meticulous
than Fielding, Thackeray, or Trollope in explaining his methods of constructing
his novels, but he limited this critical discourse to his prefaces. James’s outrage
was aroused not by this practice, but its placement within the fictional work and
its consequent disruption of mimetic pretenses. As Linda Westervelt explains,
“James does not criticize the intrusion of Trollope’s narrator, but Trollope’s
destruction of the illusion that the events in the story actually occurred” (1983,
The same opposition was noted by Boris Tomaševskij in 1925. He, however,
gives it an opposite valorization by reversing James’s privileging of the mimetic:

Two literary styles may be distinguished in terms of the perceptibility of [literary] devices.
The first, characteristic of writers of the nineteenth century, is distinguished by an attempt
to conceal the device; all of its motivation systems are designed to make the literary devices
seem imperceptible, to make them seem as natural as possible—that is to develop the lit-
erary material so that its development is unperceived. But this is only one style, and not a
general aesthetic rule. It is opposed to another style, an unrealistic style, which does not
bother about concealing the devices and which frequently tries to make them obvious, as
when a writer interrupts a speech he is reporting to say he did not hear how it ended, only
to go on and report what he has no realistic way of knowing. (Tomaševskij 1965 [1925], 94)
 Unnatural Narrative Theory: A Paradoxical Paradigm   201

To maintain both styles, both aesthetic rules, narratology needs a comprehen-

sive paradigm that provides a dual or divided poetics, one for mimetic works,
the other for antimimetic narratives. By definition, a mimetic model cannot com-
prehend antimimetic works that violate mimetic practices. A complete narrative
theory requires a binocular vision and a dialectical poetics. Unnatural narratol-
ogy is the more capacious paradigm that supplies the missing parts, the missing
theory, the missing vision.
Intriguingly, this discussion leads to another paradoxical observation. One
may think of narratives as being of three fundamentally different types: nonfic-
tional, which are subject to falsification; mimetic fictions, which substantially
mimic nonfictional practices but cannot themselves be falsified; and antimimetic
works, which flaunt their fictionality and cannot be reduced to the other types.
We cannot simply rely on our experience of personal interactions in the real
world if the fictional text defies those experiences; understanding people in real
life may be of little use when apprehending a fictional work that does not employ
human-like characters; there is little in our experience that corresponds to a char-
acter’s knowledge that he is a fictional character, and any experiences of our own
that resemble this situation may not be especially helpful in comprehending it.
It seems clear that if one insists on a unified, totalizing theory of narrative
that can embrace both fictional and nonfictional stories, one must ignore, dismiss
or deny the antimimetic. One cannot, in principle, have a single theory that is
based on mimetic entities and events as well as the violation of the principles
that mimetic representations require. On the other hand, if one desires a more
capacious theory that can include the antimimetic, a different divide erupts: that
between nonfiction and fiction. This in turn leads to a theory of fictional narratives
that are all quite different in technique, import and pragmatic status from nonfic-
tional narratives. Mimetic elements can be observed in antimimetic texts, just as
antimimetic moments will be found in substantially mimetic works. Noted, too,
will be the narrow divide between the conventions of fiction and the violations of
those conventions in unnatural narratives: it is a short but highly significant step
from first-person narration to second-person narration; from conventional omnis-
cience to omniscient narrators who plead ignorance; from complex characters
and events to mutually contradictory ones. In the end, one must choose between
an incomplete mimetic paradigm for fiction (mimeticism) or a divided, double or
even antithetical stance (the mimetic and the antimimetic); either choice entails
a very different conception of fiction, and each requires a different narratology.
In discussing mimetic models, it is important to note significant differences
among them, both in theory and in practice. Some theories unnecessarily limit
themselves to mimetic concerns. Structuralism attempts to transcend humanism,
yet Luc Herman and Bart Vervaeck (2005) have pointed out key instances of a
202   Brian Richardson

residual humanism that is pervasive in the practice of structuralist narratology.5

As I and others have argued elsewhere, most cognitivist narratologists employ an
extremely narrow mimetic framework, treating characters as people and narra-
tors like actual storytellers in the real world (see Iversen 2013 and Mäkelä 2013).
There is no inherent reason why this mimeticism should be so intransigent in
their work; in fact, some theorists, including Jan Alber (2009) and Marina Grish-
akova (2006), have shown how a cognitive approach can be effectively welded to
unnatural texts.
Other approaches seem to have the opposite problem, leaving little space for
unnatural concerns in their larger theoretical statements, yet trying nonetheless
to accommodate unnatural practices, if only in a very limited fashion. A rhetori-
cal model like that of James Phelan and Peter Rabinowitz is resolutely anchored
in a mimetic conception of fiction. For them, “narrative is somebody telling some-
body else, on some occasion, and for some purposes, that something happened
to someone or something” (Herman et al. 2012, 3). Such a definition leaves no
place for narratives like those of Samuel Beckett or many practitioners of the
nouveau roman or postmodernism. In Molloy (1951), for example, the status of
the narrator is inherently dubious: are there two or does Moran become Molloy?
How do we account for the radical, seemingly impossible change in Moran’s char-
acter? Is Molloy, who can cite Continental Rationalist philosophers but does not
know what the policeman means when he asks him for his papers, a plausible,
consistent entity? What events, if any, really occurred? The systematic denarra-
tion of the first half and the global erasure at the end of the second half suggest
that there may be no actual fabula to be recovered. Who could be the narratee of
these outrageous events, and what is the purpose of their narration? Needless
to say, attempting to answer such questions gets more difficult as we approach
more extreme unnatural texts like Beckett’s The Unnamable (1953), Robbe-Gril-
let’s La Maison de rendez-vous (1965) or Pinget’s Passacaille (1969). Despite the
general mimetic position indicated by his definition, Phelan has nevertheless
made important contributions to the theory and analysis of unnatural texts in his
studies of redundant telling. However, the danger of such a conception is that it
always treats the unnatural as a special case, possible addition or practice situ-
ated at the far end of the spectrum the theorist constructs, giving pride of place
and definitional centrality to mimetic poetics. I urge that the practice of confining

5 They note, for example, that “the fact that structuralist narratology holds on to concepts such
as hero and villain suggests that it still deals with characters in a very anthropomorphic way”
(2005, 70). They also point out that similar problems occur in Genette’s temporal concepts (63–
64) other aspects of this model (41–101).
 Unnatural Narrative Theory: A Paradoxical Paradigm   203

the unnatural to the end of a spectrum of narrative possibilities be resisted; it is

too important to be so easily marginalized.
A final irony is that the unnatural model seeks to restore the paradigm
employed by the Russian formalists but which has been largely abandoned by
the many theoretical stances that claim to build on their work. The newest para-
digm of narrative theory—unnatural narratology—thus calls for a resurrection of
one that is a century old. In addition to the work of Tomaševskij, I wish to draw
attention to Viktor Šklovskij’s theoretical pieces, one of which, “The Relationship
between Devices of Plot Construction and General Devices of Style” (1990 [1919]),
though long neglected, is especially helpful in analyzing nonmimetic principles
of story construction.
Unfortunately, however, a new mimeticism has recently begun to emerge
as narrative theories based on cognitive studies repeatedly insist on a homol-
ogy between human experience and literary interaction. The cognitive theories
developed by Richard Gerrig, David Herman and many others treat narrative in
a largely or almost exclusively mimetic manner. Thus, for example, they con-
sider literary characters essentially as if they were human beings or human-like
persons. This bias forces them to neglect and leave untheorized the thousands of
non- or anti-mimetic characters from those of Aristophanes to Beckett’s Unnam-
able to, for that matter, Bugs Bunny. What is especially disappointing in this
context is that this is a mistake that we have seen many times before in the history
of literary theory. We recall the naïve mimeticism that drove A. C. Bradley to spec-
ulate vainly on how many children Lady Macbeth had, or how many years Hamlet
studied in Wittenberg. This “mimetic fallacy” can be traced back at least as far as
Ben Jonson; now it threatens to enjoy a new resurgence.6
Those who appreciate either innovative literature or accurate theoretical
formulations can only hope that the new mimetic bias does not take hold. As
Lubomír Doležel has stated,

Mimetic doctrine is behind a very popular mode of reading that converts fictional persons
into live people, imaginary settings into actual places, invented stories into real-life hap-
penings. Mimetic reading, practiced by naïve readers and reinforced by journalistic critics,
is one of the most reductive operations of which the human mind is capable: the vast, open,
and inviting fictional universe is shrunk to the model of one single world, actual human
experience. (Doležel 1998, x)

6 For a sustained critique of cognitive mimeticism, see my response in Herman et al. (2012,
204   Brian Richardson

To reiterate: a narratological paradigm that restricts itself to standard, natural,

common, mimetic or conventional narrative forms suffers several debilitating lim-
itations. By contrast, a more open and expansive model offers numerous obvious
benefits, particularly when it comes to the narratives any narrative theory should
be able to cover. In conclusion, we affirm that fiction is different, often wildly
different, from nonfictional narrative; it is different by definition and in practice,
and this basic alterity needs to be taken account of in any narrative theory that
aims at comprehensiveness. André Malraux clarifies this basic opposition pithily:
“L’artiste n’est pas le transcripteur du monde, il en est le rival” (“The artist is not
the transcriber of the world, he is its rival”) (1976, v). This is a good motto for us

Works Cited
Primary sources
Bradbury, Malcolm. 1993 [1976]. “Composition.” In M. B. Who Do You Think You Are? Stories
and Parodies, 119–146. New York: Penguin.
Castillo, Ana. 1992. The Mixquiahuala Letters. New York: Doubleday.
Churchill, Caryl. 1985 [1978]. Traps. In C. C. Plays: One, 69–125. New York: Routledge.
Coover, Robert. 1969. “The Babysitter.” In R. C. Pricksongs and Descants, 206–239. New York:
Leyner, Mark. 1995. My Cousin, My Gastroenterologist. New York: Vintage.
Queneau, Raymond. 1967 [1961]. “Un conte à votre façon.” Les Lettres nouvelles julliet-
septembre, 11–14.
Robbe-Grillet, Alain. 1965. Two Novels by Robbe-Grillet: Jealousy [1957] and In the Labyrinth
[1959]. Translated by Richard Howard. New York: Grove.

Secondary sources
Abbott, H. Porter. 2008 [2002]. The Cambridge Introduction to Narrative. 2nd edition.
Cambridge: Cambridge University Press.
Alber, Jan. 2009. “Impossible Storyworlds—and What to Do with Them.” Storyworlds 1: 79–96.
Jan Alber, Henrik Skov Nielsen and Brian Richardson, eds. 2013. A Poetics of Unnatural
Narrative. Columbus: The Ohio State University Press.
Doležel, Lubomír. 1998. Heterocosmica: Fiction and Possible Worlds. Baltimore and London:
The Johns Hopkins University Press.
Fludernik, Monika. 1996. Towards a ‘Natural’ Narratology. London and New York: Routledge.
Grishakova, Marina. 2006. The Models of Space, Time and Vision in V. Nabokov’s Fiction:
Narrative Strategies and Cultural Frames. Tartu: Tartu University Press.
 Unnatural Narrative Theory: A Paradoxical Paradigm   205

Heise, Ursula. 1997. Chronoschisms: Time, Narrative, and Postmodernism. Cambridge:

Cambridge University Press.
Herman, David, James Phelan, Peter Rabinowitz, Brian Richardson and Robyn Warhol. 2012.
Narrative Theory: Core Concepts and Critical Debates. Columbus: The Ohio State University
Herman, Luc, and Bart Vervaeck. 2005. Handbook of Narrative Analysis. Lincoln and London:
University of Nebraska Press.
Iversen, Stefan. 2013. “Unnatural Minds.” In Alber et al., eds., 94–112.
James, Henry. 1972 [1883]. “Anthony Trollope.” In Miller, ed., 175–176.
James, Henry. 1972 [1884]. “The Art of Fiction.” In Miller, ed., 27–44.
Mäkelä, Maria. 2013. “Realism and the Unnatural.” In Alber et al., eds., 142–166.
Malraux, André. 1976. Les metamorphoses des dieux. Vol. 3: L’Intemporel. Paris: Gallimard.
Miller, James E., Jr. ed. 1972. Theory of Fiction: Henry James. Lincoln: University of Nebraska
Prince, Gerald. 1982. Narratology: The Form and Functioning of Narrative. Amsterdam: Mouton.
Richardson, Brian. 1987. “‘Time is Out of Joint’: Narrative Models and the Temporality of the
Drama.” Poetics Today 8 (2): 299–309.
Richardson, Brian. 2002. “Beyond Story and Discourse: Narrative Time in Postmodern and
Nonmimetic Fiction.” In Narrative Dynamics: Essays on Time, Plot, Closure, and Frames,
edited by B. R., 47–63. Columbus: The Ohio State University Press.
Richardson, Brian. 2005. “Beyond the Poetics of Plot: Alternative Forms of Narrative
Progression and the Multiple Trajectories of Ulysses.” In A Companion to Narrative Theory,
edited by James Phelan and Peter Rabinowitz, 167–180. Malden, MA: Blackwell.
Richardson, Brian. 2006. Unnatural Voices: Extreme Narration in Modern and Contemporary
Fiction. Columbus: The Ohio State University Press.
Richardson, Brian. 2013. “Unnatural Stories and Sequences.” In Alber et al., eds., 26–48.
Richardson, Brian. 2015. Unnatural Narrative: Theory, History, and Practice. Columbus: The
Ohio State University Press.
Robbe-Grillet, Alain. 1965 [1963]. For a New Novel: Essays on Fiction. Translated by Richard
Howard. New York: Grove.
Šklovskij (Shklovsky), Viktor. 1990 [1919]. “The Relationship between Devices of Plot
Construction and General Devices of Style.” In V. S. Theory of Prose, translated by
Benjamin Sher, 15–51. Elmwood Park, IL: Dalkey Archive Press.
Sternberg, Meir. 1981. “Ordering the Unordered: Space, Time, and Descriptive Coherence.” Yale
French Studies 61: 60–88.
Tomaševskij (Tomashevsky), Boris. 1965 [1925]. “Thematics.” In Russian Formalist Criticism:
Four Essays, translated by Lee T. Lemon and Marion J. Reis, 61–98. Lincoln: University of
Nebraska Press.
Westervelt, Linda A. 1983. “‘The Growing Complexity of Things’: Narrative Technique in The
Portrait of the Lady’.” The Journal of Narrative Technique 13 (2): 74–85.
Göran Rossholm (Stockholm)
Causal Expectation
This essay is an attempt to give the discussion about causation as a narrative
universal a new direction. The scholarly context consists of a discussion about
causality in narratives conducted by three philosophers, Noël Carroll (“On the
Narrative Connection” 2001), David Velleman (“Narrative Explanation” 2003) and
Gregory Currie (“The Content of Narrative” 2010, ch. 10). The term “narrative” is
used in a transmedial sense throughout the essay and thus includes narrative
films, theatre plays, etc.; most of the time my use of the term “reading” should be
understood in an accordingly wide and variable sense.1
Causality has been suggested as a characteristic of narrative ever since Aris-
totle, but it is rare that the concept as it is used in narratological contexts is crit-
ically analysed in terms of conditions. Carroll, however, does present such an
analysis. His thesis as to what kind of causality characterizes narratives is crit-
icized by Velleman, who refutes not only Carroll’s thesis of narrative causality,
but all similar theses, every version of what I will call a Weak Causality Model.
According to Velleman, causality is not what holds a narrative together. He argues
partly by counter examples, in particular one taken from Aristotle’s Poetics, and
he presents an alternative model, the Emotional Cadence Model. Currie, finally,
admits that there are non-causal stories, or at least stories with no causality at
their turning points, but, contrary to Velleman, he is not willing to throw the
idea of narrative causation out of the discussion. He meets the presumed counter
examples by attending to the notion of narrativity, instead of narrative, and by
formulating a combination of two narrativity models, the Compensation Model
and the Conflict Model.
My position will be rather close to Currie’s; still, my basic idea about narra-
tivity and causality will be quite different in one certain respect, coming closer to
Velleman’s proposal in this regard. Like Velleman, I will move the focus from the
story to the reader: I suggest that a narrative is something that gives rise to causal

1 The two terms in my title, causal and expectation, have been investigated in narratological
contexts by many scholars, just as related concepts such as plot and anticipation or chance and
surprise. Many of the results of these investigations are highly relevant in the present context, in
particular Sternberg’s (1978) observations on suspense, curiosity and surprise and, with respect
to the different kinds of reader expectation, Jauss’s (1982 [1977]) and Iser’s (1990, 1991) theories.
However, to make the line of argument clear and straight, I concentrate my references on the
publications of the three philosophers mentioned above.

DOI 10.1515/9783110555158-011
208   Göran Rossholm

1 Causal Troubles
Before entering the discussion, a few remarks are needed on causation in general
and on the role of causation in narrative in particular.
The concept of causality is philosophically and scientifically controversial,
and the terms “cause” and “effect” and their closest semantic relatives have
been used differently in different times and cultures. These circumstances call
for caution. Currie (2010, 28–41) has suggested that we can speak about depend-
ences in many cases instead of causation, and I agree. I shall come back to this
point below. For the moment, though, suffice it to say that the notion should
be interpreted as including sufficient causation as well as teleological forces of
all kinds—mechanical causes as well as intentional ones, full-blown sufficient
causes as well as tiny causal factors—and that the use of such a notion should
go along with determinism as well as non-determinism. I believe that the loose
concept of dependence is part of any more specific (and consequently more con-
troversial and more culturally relative) concept of causation.2
One of the most quoted examples of causation and narrative comes from
E. M. Forster: “‘The king died. Then the queen died’ is a story. ‘The king died.
Then the queen died of grief’ is a plot.” (Forster 1927 [1962], 93). Few, if any, have
followed Forster in drawing a line between story and plot by using the criterion
of causality; rather, the example is often used to shed light upon the question of
whether any coherent discourse reporting two or more successive events is a nar-
rative. A standard reference in this debate is Morton White’s (1965) classification
of annals (records of successive events), chronicles (records of successive events
with a single subject) and narratives (records of successive events with a single
subject which are causally organized).3 Of course, other genres may be put into
these three boxes: for instance, a witness report is about what a single subject has
observed but without claims on causal links between the events perceived and is
thus a “chronicle” but not necessarily a “narrative.” However, these distinctions
are stipulations, and I’m far from certain that we will ever find any stable linguis-
tic intuitions behind them. At most, we might say that many “annals” are less

2 The caution called for yields no requirement that every detail in the chosen concept—depend-
ence in this case—should be embraced as the cement and propeller of the universe by any inhab-
itant in the same universe. I will in the following stipulate that dependence should be factual,
not logical or conceptual—a distinction that is probably not universally acknowledged.
3 The first two categories mentioned in White’s annals-chronicles-narratives tripartition should
be taken in a broad sense; for instance, when a person makes an account about his wherea-
bouts in order to establish an alibi in a police investigation, the discourse may take the form of
a chronicle.
 Causal Expectation   209

narrative than most “chronicles” and that many “chronicles” are less narrative
than most “narratives.” This leaves us with several alternatives. For instance, we
might draw a definitional line between “narrative” and the other two categories,
or we might consider “chronicles” and “annals” (or only “chronicles”) as subspe-
cies of “narrative.” As already said, I will follow quite another line of reasoning,
and consequently I won’t go deeper into this.
The causality claim does not answer the question: “How much?” In the first
paragraph of the Grimm Brothers’ version of “Little Red Riding Hood,” the reader
is informed of the fact that the protagonist is given a hood which is red and made
of velvet. The colour plays a causal role in the text (the girl is called Little Red
Riding Hood because of the colour of the hood), but the textile material velvet
does not. Does this disqualify the whole text as a causal narrative? Of course not.
Every piece of information does not necessarily have a specific place in the causal
web of the story. (This is so even though, as will be shown at the end of this essay,
the textile material of the hood may be given a causal interpretation.)
An unqualified thesis of causality may include too much causality. A state-
ment such as “The sun shines” refers to a complex causal process, but it makes
a poor narrative, if any at all. Currie points out that a representation of a person
at different times is a causal representation because personal (gen)identity is a
causal affair (Currie 2010, 38); similarly, Mackie discusses whether persistence
in general presupposes causality (Mackie 1980, 156–157). However, none of these
cases illustrate narrative causality because, I suggest, they don’t necessarily yield
discourses that represent anything as a causal process. The sentence “The sun
shines” neither spells out nor tacitly presupposes any specific and separate tem-
poral items (events, circumstances, conditions) playing the roles of causal factors.
Thus, to represent something that de facto is a causal process is not enough: it
has to be represented as a causal process.4
Finally, we cannot require that a narrative should give us sufficient causes of
the events represented. Many events in most narratives are not presented as the
consequences of preceding determining causes; everything in every narrative is
not causally explained. Some genres, such as the gothic novel, are characterized
by the lack of convincing causal explanations, and most narratives, in particular
good ones, are such that the process of events cannot be easily foreseen. Surprise
is not only common in narratives but is often praised as a narrative virtue, and
in some genres (narrative jokes, for instance) it is constitutive. Thus, narratives
hang together by some looser link than sufficient cause and its effect.

4 For a more elaborate account of representing something as a causal process, see Rossholm
210   Göran Rossholm

2 Noël Carroll and the Weak Causal Model

In order to cope with the problem mentioned in the last paragraph above, Noël
Carroll puts forth a weak form of coherence. The cement he proposes has two ingre-
dients. First, a narrative discourse must be held together by “a unified subject,”
that is, it should live up to the requirements of a chronicle. Forster’s example
of story coheres in this way. We take it for granted that the queen mentioned in
second sentence is not just any queen but the one who is (or was) married to the
deceased king. Discourse coherence is prototypically expressed by pronouns.5
We can make Forster’s sentences more explicit by paraphrasing them: “The king
died. Then, his wife, the queen, died.”6 The second cementing element is causa-
tion, and this is the specifically narrative component of the connection. Carroll
argues that we can’t require that narrative events be connected only by sufficient
causal conditions (i. e., a condition which produces the effect); instead, he pro-
poses necessary parts of sufficient causal conditions, so called INUS-conditions,
as components binding a narrative together.7 As mentioned above, Currie sug-
gests (2010, 40–41), with reference to David Lewis (1986), that causation in a still
weaker sense could be paraphrased as dependence. One event or state of affairs
depends directly or indirectly (in the case of INUS-conditions) on another event
or state of affairs. He also takes two more steps to loosen up the causal knots still
more. First, he allows not only categorical causal conditions of different sorts,
but also probabilistic ones. Second, he accepts David Lewis’ suggestion that neg-
ative causal information may also be part of causal explanation. This idea means
that facts that are only seemingly causally connected to the event to be explained
are parts of the causal history of that event (Lewis 1986, 220) (see the example
below). Finally, as mentioned above, we cannot require of any causal model that
every piece of information should have a causal role to play, only that causation
dominates the picture, and that the crucial turns of the plot are causally related
to something else in the story.

5 On coherence and cohesion, see Michael Toolan (2014 [2009]).

6 I have advocated explicating this kind of coherence in terms of logical unitarity in Rossholm
7 The four letters are short for Insufficient but Non-redundant part of an Unnecessary but Suffi-
cient condition (see Mackie 1965, 1980). A concept very similar to Mackie’s INUS-condition was
introduced a few years earlier by Marc-Wogau (1962) under the name of “moment in a sufficient
 Causal Expectation   211

3 David Velleman and the Emotional Cadence

David Velleman presents counter examples to this model. He does not deny that
narratives often present sequences of causally connected events in accordance
with the model, but this is neither a sufficient nor a necessary condition for being
a narrative (the conditional terminology now used in its logical or conceptual
sense). There are non-narrative discourses that comply with the weak causal
model, and there are full-blooded narratives which do not comply: “A discovery
due to serendipity, a tragedy narrowly averted by dumb luck, a mundane act that
unforeseeably becomes the last in a life accidentally cut short […]” (Velleman
2003, 6). He provides examples drawing on narrative motifs such as improbable
reunions of twins separated since birth and narrative genres such as joke-telling,
and he finds support for his non-causal position in a passage from Aristotle’s

Tragedy, however, is an imitation not only of a complete action, but also of incidents arous-
ing pity and fear. Such incidents have the greatest effect on the mind when they occur unex-
pectedly and at the same time in consequence of one another; there is more of the marvel-
lous in them then than if they happened of themselves or by mere chance. Even matters of
chance seem more marvellous if there is an appearance of design as it were in them; as for
instance the statue of Mitys at Argos killed the author of Mitys’ death by falling down on him
when a looker-on at a public spectacle; for incidents like that we think to be not without a
meaning. (Aristotle 1952, 9:1452, p. 686)

This story, the story of Mitys, may of course be interpreted as causally connected.
In such an interpretation, the murderer is probably killed by Mitys’ spirit, or by
the statue coming to life. After all, people are intentionally killed by statues in
narratives  – as Don Juan is. But Aristotle is obviously presenting the case as a
story about happenstance and coincidence, not about causal effect, and Velle-
man does so, too.
How, then does Velleman analyse the concept of narrative? First, he moves
the focus from the content of the story to the emotional experiences of its audi-
ence. What makes certain representations narratives is not their internal struc-
ture but the feelings they evoke: “So long as we feel an anxiety relieved or a hope
dashed, we have the sense of hearing a story, even if we have no idea why events
took the relevant turn.” (Velleman 2003, 7) Then, he specifies emotive content as
instances of “emotional cadences.” The reader or spectator of a narrative expe-
riences a sequence of emotions, indefinite in number but ordered in a certain
way. The emotive sequence is initiated by an emotion that, in real life as well as
in reading experience, is not typically the consequence of another emotion but
212   Göran Rossholm

comes, in Velleman’s phrase, from “out of the blue,” and this feeling is followed
by one or more emotions which typically are emotion-determined. The sequence
ends in a feeling of resolution, that is a feeling with no standard emotional fol-
low-up. (You can hear Aristotle’s three-way split of “a complete action” into the
causally determined parts “beginning, middle and end” behind these words.)
One example: first, we have puzzlement (a feeling that is no standard conse-
quence of another feeling), both in the story and its audience, then curiosity, then
foreboding, then dismay and finally, as a consequence of the whole series, grief
(a feeling with no standard emotional consequence). This is the emotive cadence
of Oedipus Rex.

4 Gregory Currie, the Compensation Model and

the Conflict Model
Gregory Currie argues that Velleman’s narrative essence, the emotional cadence,
is no sufficient condition of the occurrence of a narrative. A scientist making a
discovery or formulating a proof may experience the same sequence of emotions.
I would like to add that neither does Velleman’s model provide a necessary condi-
tion. His concept of narrative is too narrow, confined to stories with closure. Many
narratives, written or not, literary or not, spontaneous or calculated, do not end
in resolution. Further, there are stories (with or without closure) with such a low
emotional temperature that the reader/spectator stays emotionally unaffected,
yet recognizing what s/he reads as a narrative.
Still, Currie admits that there are narratives with pivotal changes that just
happen, with no obvious causal background. He adds one of his own making:
“[A] story of man who successfully battles a life-threatening disease but who is
killed in an unrelated car crash the day after his discharge from the hospital.”
(Currie 2010, 32) There are, as we all know, a lot of such stories about sudden and
inexplicable misfortune or fortune.
Currie’s basic move in this troublesome situation (Velleman is correct in
his diagnosis, but his cure to ordinate “emotional cadences” doesn’t help) is to
attempt to find any narrative essence, any narrative connection, any defining
characteristic of all and only narratives. He thus turns to the gradual property of
narrativity. One discourse that is not a narrative may still be more narrative—i. e.,
have more narrativity—than another discourse: one definite narrative may be
more narrative than another; two discourses in the grey area between narrative
and non-narrative may differ in narrativity. Narrativity consists of a number of
entities and features such as causation, characters, chronology, closure, conflict,
 Causal Expectation   213

surprising turns, narrative point or tellability, complications, suspense, experen-

tiality, and so on. This move is far from new, the first to come to mind being that
of narrative theorist Gerald Prince. In addition to the candidates mentioned, he
suggests that “disnarration” is one more factor among the different features of
narrativity. By disnarration he means “explicitly referring to what did not take
place but could have.” (Prince 2008, 24) Each of these categories may be more
or less manifest in a particular discourse, and when we have enough of these
N-factors we have an exemplary narrative. This means that the loss of one feature
may be compensated by an overdose of another. In addition, Currie points to the
possibility that one salient feature may conflict with another. Take the story about
the man who survived a dangerous disease only to be killed in a car accident. The
point of the story might be to illustrate the unforeseeable changes and inexplica-
ble variability of human life and death—or something of that nature.

5 The Causal Expectation Model

Currie ends his discussion about narrative causation by saying that “causation
is important in narrative; by and large we expect narratives to tell us a lot about
causes, and to have their most salient events strongly embedded in causal con-
texts”; moreover, “narrativity varies to some extent with causal information.”
(Currie 2010, 38) These quotes indicate that causation holds an exceptional posi-
tion among the features of narrativity.
On the whole, I find Currie’s suggestion about counter examples right. The
step from narrative to narrativity, to narrative factors which may both compensate
for each other and conflict with each other, is convincing. But I think that there
is more to this complex question. Currie’s proposal does not have anything to say
about what makes all the flowers in the bouquet of narrativity narrative flowers.8
Neither does he make clear why causation seems to take a somewhat privileged
position among these different factors.
I suggest that the answer to these questions is hinted at in Currie’s own words,
just quoted: we, as readers, expect narratives to tell us a lot about causes. Nar-
rative, causally organized or not, is something that raises causal expectations.
This is, of course, no definition: other things, events and situations also evoke

8 Currie does not mention any more narrative factors than causality. The eclectic list presented
in the previous section is of my own making.
214   Göran Rossholm

causal expectations. But, I suggest, it is an important characteristic of narrative

in general.
I will devote the rest of this essay to clarifying what I mean by “causal expec-
tation” with respect to narrative. The proposal does not contradict Currie’s ideas:
in fact, it suggests answers to some questions raised by his model as well by the
weak causality model and the emotional cadence model.
First, a few stipulations about the term “expectation.” An expectation may be
conscious, but there are also unconscious expectations. An expectation may be
or involve a belief of what will come to pass, but not necessarily. When something
surprising happens, this is often testimony of the presence of a belief. If I turn
on the shower I expect water, but I don’t necessarily form any conscious belief
about it. I become surprised when nothing happens, and my surprise informs me
about my until-this-moment unconscious belief. I might expect several possible
outcomes in a certain situation. For instance, when going to the countryside for
a weekend, I have no definite belief that it will be very rainy or that it will be dry
and hot; but being prepared for any of these extreme forecasts, I bring both my
umbrella and my bathing suit. It is within my anticipatory perspective, and in this
sense I expect it. With respect to reading, Currie uses the apt term “salient pos-
sibility” (2010, 42–43).9 Both conscious and unconscious expectations and both
belief-expectations and non-belief-expectations may be positive (that a certain
event will/can happen) or negative (that something will not/may not happen).
I always carry about with me a number of unconscious negative expectations
such as, when expecting a student to come to my office I have no particular posi-
tive expectations about how he will be dressed, but still I feel surprised when he
arrives in his pyjamas.
Now, how does narrative causal expectation work? Currie’s formulation,
“expect narratives to tell us a lot about causes” (2010, 38), comprises reference to
expectations we already have before we start reading (or watching or whatever).
From, and even before, the very start, we expect to be informed about individuals
and events in a causal structure. During reading, this very sketchy anticipation
is filled in by expectations of more specified information about these individu-
als, specifications that in turn are further specified or revised or thwarted as the
reading goes on.
The initial expectation, which we can call a frame expectation since it frames
the whole reading and the whole story, is an external expectation: the stimu-
lus for what we anticipate is not part of the content of the narrative. Such an
external frame expectation may, however, be more specific, depending on our

9 Currie acknowledges Carroll (1990) for the idea about salient possibilities.
 Causal Expectation   215

acquaintance with the genre, the writer or some other external circumstance.10 If
we decide to read an Agatha Christie novel, we expect from the beginning—and
even before—that we will be informed about a murder which will necessitate an
investigation that exposes the murderer who, as a consequence, will probably be
caught and punished. If we decide to watch a programme of the TV-series Murder,
She Wrote, our external frame expectations will be even more detailed, particu-
larly if we have already watched previous adventures of this series.11 As fans of
Murder, She Wrote, we expect to be informed about a nice young couple who,
as a result of various causal turns, will unite in the end, and we will expect that
someone will be murdered. We will expect the victim to be an unsympathetic
character, both to us and to most of the characters in the story, and we will expect
the murderer to be someone other than one of the two young lovers. We will also
expect that the first person to be arrested will be proven innocent and believe that
protagonist Jessica Fletcher’s own friends have nothing to do with the crime. We
will expect the murderer to be identical to one of the characters with only a very
weak causal function in the story (aside from his/her presumed participation in
the murder, of course). We further expect that as a consequence of the murder,
Jessica Fletcher will conduct an investigation that has the effect of the murderer
being exposed and arrested. All this and much more belong to what the alert
viewer can anticipate as s/he switches on the TV. As the story unfolds, we form
expectations that specify these frame expectations: we predict that the wealthy
stockholder and tyrannical father will be killed, that his son will be arrested and
proven innocent by Jessica Fletcher and that the father’s cousin, a vague but
harmless character, will be exposed as the murderer. In the confession, always
given as a consequence of a set-up, he presents the causal material still to be
filled in in order to make the causal blanket complete.
Some of the expectations of causal links in narratives are external (mostly
genre-determined) and some are internal, that is, elicited by the story informa-
tion. Causal expectations may be divided along other lines as well, for instance,
into effect expectations and causal explanation expectations. When reading a nar-

10 Gérard Genette’s (1997 [1982]) theory of architextuality (generic membership plus horizon of
expectations) and paratextuality (the “threshold” beween the text and what lines outside) are
highly relevant in the context of what I call frame expectations.
11 “Murder, She Wrote is an American television mystery series starring Angela Lansbury as
mystery writer and amateur detective Jessica Fletcher. The series aired for 12 seasons from 1984
to 1996 on the CBS network, with 264 episodes transmitted. It is one of the most successful and
longest-running television shows in history, with close to 23 million viewers in its prime, and was
a staple of its Sunday night lineup for a decade.”
(Wikipedia.,_She_Wrote) (Accessed 10 August 2015)
216   Göran Rossholm

rative, I form an effect expectation when I receive story information that prompts
me anticipate some event. Something in the information refers to (some part of)
the cause of what I expect will happen. In the second case, causal explanation
expectation, something makes me expect that a certain event or state of affairs
will be given a causal explanation in the course of events to follow. The first case
is exemplified by the information that someone takes out a gun from a drawer in
preparation, the reader presumes, for a murder. The second kind of expectation
is evoked by, for instance, a scene in which we see someone spying on someone
else: at that moment, we have no idea why this is happening, but we expect to be
informed of the causal context of the event. It should be noted that this example
is also an example of an external expectation: we expect to be causally informed
because we are reading a narrative. Very often, this kind of expectation is com-
bined with an effect expectation. The obligatory confession scene in Murder, She
Wrote can serve as an example: here the confession is an anticipated effect of
Jessica Fletcher’s manipulations, and the information given by the murderer puts
prior events in their proper causal context.
Two points are to be made here. First, about internal and external stimuli: in
crime stories genre features also motivate many, maybe most, expectations trig-
gered by what happens in the story. Crime stories (at least certain kinds, to a lesser
extent stories à la Raymond Chandler) should give this kind of information sooner
or later. Second, about causal explanation expectations: the spying activities just
mentioned may be totally irrelevant to the causal structure of the story, and this
irrelevancy may be proven. Although no causality is found, the causal expecta-
tion has done its job by drawing extra attention to the (ir)relevancy of the spying
incidents: the episode has stimulated our causal curiosity, thereby strengthening
the causal expectation attitude. This hypothetical case corresponds to (a kind of)
disnarration in Prince’s sense and negative causality in Lewis’ sense. (Lewis’ own
example fits in nicely here: “Why was the CIA man there when His Excellency
dropped dead? – Just coincidence, believe it or not.” Lewis 1986, 220)
Which of these different kinds of expectation are what I have called causal
expectations? First, of course, all expectations of elements of causal explana-
tion, both causal frame expectations and more specific expectations of locat-
ing presented events in their proper causal context. This means that external
expectations (i. e., anticipations made on external grounds solely) may be causal
expectations insofar as they are causal explanation expectations. But can exter-
nal expectations be effect expectations? They may point to the same events as
internal causal effect expectations do, but they also have to link these events to
their (partial) causes. When we meet two bears or two goats or two brothers in a
fairy tale, we expect a third, and when the third one has arrived we don’t expect a
fourth one. This has nothing to do with causation. Let’s look at one of the exam-
 Causal Expectation   217

ples in Murder, She Wrote. We expect that the one who is initially put behind bars
is innocent. But of course he or she is not innocent because of the fact the he or
she is the first one arrested. However, this, and other, external expectations are
still very similar to internal effect expectations. They have the same form:

The reader is informed that X is the first to be arrested, and therefore the reader expects that
X will be released.
The reader is informed that X has a watertight alibi, and therefore the reader expects that
X will be released.12

Only in the second case are two causally related facts mentioned. In the story,
that is. For in both cases a causal link is established within the reader: in the
first case, his conviction about the arrest of X, and in the second, his convic-
tion about the alibi—both of them causal antecedents of his expectation. But
only in the second case do the mental causal poles, conviction and expectation,
mirror a similar relation outside the reader. However, the difference is invisible
in this formulation because the external basis of the information is left unsaid.
A more explicit formulation of the first case should mention that the reader is
also informed that in every Murder, She Wrote adventure the one who is initially
arrested for the murder is innocent:

The reader is informed that X is the first to be arrested and therefore expects that X will be
released, because the first arrested is almost always innocent.
The reader is informed that X has a watertight alibi and therefore expects that X will be
released, because persons with reliable alibis are innocent.

The third, law-like, italicized line makes all the difference. In the second triple it
exemplifies an internal law—something both Jessica Fletcher and the sheriff can
agree on. The third is an external law, valid for the TV-series Murder, She Wrote.
Jessica could not and would not argue that the sheriff should release his prisoner
on the grounds that he is the first to be put behind bars in this case.
However, we cannot dismiss all the external examples presented. The exter-
nal clue that the arrested person is an old friend of Mrs Fletcher’s or the external

12 I anticipate two objections to this way of formulating the reader’s effect expectation. The
term “information” cannot be applied to fiction, and a more accurate phrasing of the second part
of the sentences would be “therefore the expects to be informed that X is released.” To the first
objection I say: use “information” metaphorically with respect to fiction. And to the second: the
reformulation is acceptable, but I prefer the more direct way of putting it because it stresses the
similarity between what happens during reading and direct, non-representational, experience.
218   Göran Rossholm

clue that that the murder victim is a highly unsympathetic person may be inter-
nally relevant. The hallmark of external stimulus information is that it includes
references to the narrative work, the audience, etc., or that it is motivated by infor-
mation including such references. The last mentioned case may be motivated by
the fact that Murder, She Wrote is a cosy crime series, far from evoking pity and
fear, and that its audience doesn’t mind much when unsympathetic characters
pass away. But both cases may be reinterpreted as internal clues. The immediate
trigger is external and non-causal, but the afterthought constructs a causal link.
We may judge Mrs Fletcher to be such an expert on human character that she
would probably not choose a potential murderer as a friend, and we may regard
unsympathetic persons as more likely to be killed because they probably have
made more enemies.
There are also cases in which internal stimuli promote expectations with dis-
putable causal status. For instance, the reader may expect x to happen because
the character s/he is reading about expects it. The trigger is not external: there
is no code that claims that everything the character expects will come true. The
reader may consider this a non-causal expectation, but s/he may also interpret it
as causal. In the latter case we have an example of what Gendler (2010, passim)
calls “contagion”: the character’s expectation infects the reader. Like externally
based expectations, it does not present any rational ground for a causal infer-
ence, but it can be given a causal reinterpretation: “The character expects this to
happen; most probably she has some good reason to do so.”
Does this mean that internal stimulus information is information that plays
the same role in our real life? No, for there are certainly instances of highly unre-
alistic internal stimuli. When someone puts on a specific ring in Tolkien’s trilogy,
we expect that this person will be invisible. We have learned from the fictional
narrative that this is what happens when you put this ring on your finger: it is a
piece of internal information that triggers a causal assumption.

6 The Causal Expectation Model and Counter

There is much more to be said about causality in crime narratives in general and
Murder, She Wrote in particular, but I stop here. I have used this material to illus-
trate how our causal expectations are evoked when we are reading or watching
a narrative, not as an argument for the causal expectation model. Crime stories,
most of them at least, are probably extreme examples of discourses by which
our causal expectations are both amply evoked and generously met. The weak
 Causal Expectation   219

causality model would work well here, too. What we have to demonstrate is the
virtues of the causal expectation model in the counter examples mentioned and
see how our model works in relation to other candidates for narrativity.
First, Currie’s example, the death in the car accident: “[A] story of man who
successfully battles a life-threatening disease but who is killed in an unrelated
car crash the day after his discharge from the hospital.”
This is not a narrative but a synopsis of a narrative. The fact that the protag-
onist dies immediately after being successfully treated in the hospital makes it
probable that the reader will experience the information of his violent death with
a shock or, at least, surprise. And surprise is an indication of the presence of a
belief, in this case a belief of what will come to pass and what will not, that is,
an expectation. After having successfully survived a life threatening disease, the
patient is not expected to abruptly die. So eluded expectation results in surprise,
an effect that could not have been accomplished without a causal expectation.
Surprise results in turn in intensifying the reader’s attention to the narrative, thus
making him or her more alert to what the theme of the story might be and also to
its point: fortune comes, fortune goes, the weights of good and evil will in the end
be balanced. Causal expectation has done its job by being eluded.13
In this context I would like to rectify a number of details in the account of
Murder, She Wrote. It is not true that all the external “laws” are without excep-
tions. Sometimes the victim is sympathetic, and sometimes the murderer is one of
Jessica’s friends, and so on. The intention is probably to surprise the habituated
audience. Internal stimuli may also misfire: for instance, when the spectator of
Murder, She Wrote watches a raised arm holding an adjustable spanner behind
the back of an unsympathetic character in a dark alley, s/he expects a mortal/
fatal blow to the head of that character; but as matter of fictional fact, waving
with the spanner may just be a salutation. The Murder, She Wrote connoisseur
is well aware of all this, and consequently both expects that the effect will occur
and that it won’t. Both options are, to use Currie’s phrase, salient possibilities. In
other words, the fact that we expect both the probable effects and the frustration
of this expectation is no argument against the Causal Expectation Model. The
cognitive and emotional effect on us depends on our anticipations.

13 It should be noted that this synopsis is organized in accordance with the requirements of
the weak causal model: if the man hadn’t fallen seriously ill, he would probably not have been
taken to the hospital and treated as he was; and if he hadn’t been treated as he was, he probably
wouldn’t have got well when he did; and if hadn’t got well at that time, he probably wouldn’t
have left the hospital at the time he did; and if he hadn’t done so, he would not have been killed
in an auto accident at the time he did. The process is an unbroken chain of factual necessary
220   Göran Rossholm

The story of Mitys’ statue also being only a synopsis, it is not clear what
role causal expectation would play. However, we might imagine movie versions
in which the murderer approaches the statue. He might be ignorant of the pres-
ence of the image of his victim, or he might look scornfully or with anguish at
the sculpture. In all three cases the spectator would probably form some kind of
expectation of punishment, even when s/he has no idea of how this punishment
would be carried out; what seems to be (but is known not to be) an encounter
between offender and offended would probably give rise to some such anticipa-
tion. At the same time, the knowledge that the statue is nothing but a piece of
dead material, incapable of intentional action, would tell the spectator that it
could not reasonably do the murderer any harm by itself. But it does. One reaction
is certainly surprise, but as sketched here, another reaction is the satisfaction of
having one’s expectation fulfilled. Again: causal expectation has done its job by
meeting one expectation and creating one surprise by eluding another—a nega-
tive expectation of what could not happen.14

7 The Causal Expectation Model and Narrativity

In both examples surprise is accomplished as a consequence of frustrated causal
expectation. In the first case we assumed that narrative point, or tellability, is
exhibited. William Labov introduced narrative point in the narratological arena
in the following way:

Pointless stories are met (in English) with the withering rejoinder, “So what?” Every good
narrator is continually warding off this question; when his narrative is over, it should be
unthinkable for a bystander to say, “So what?” Instead, the appropriate remark would be,
“He did?” or similar means of registering the reportable character of the narrative. (Labov
1972, 366)

The question “He did?” in contrast to “So what?” usually indicates surprise in a
causal context: “He did? I wouldn’t have expected that!”
According to many theorists, narrative point and surprise are two of the factors
that make a discourse narrative. I mentioned more such candidates earlier: char-
acters, chronology (the order of the telling mirrors the order of the told), narrative

14 Werner Wolf applies a similar approach to literary interpretation, demonstrating how chance
in narratives operates as “privileged ‘doors’ that permit access to hidden depths of implied world
views.” (2008, 167)
 Causal Expectation   221

closure, conflict, complications, suspense, experentiality and, following Prince,

disnarration (Prince 2005, 2008). All of these terms, except chronology, refer more
often than not to causal relations. When disnarration presents plausible but false
clues that seemingly, but only seemingly, contribute to the causal web, it plays the
role of prompting causal expectation without fulfilling those clues. Currie (2010,
ch. 10) has stressed the ambiguity of the term “character,” and both senses apply
to literary characters. Briefly put: characters have character. The second meaning
of the word has a clear causal and anticipatory role in most narratives: we explain
and predict the doings of a character (in the first sense, i. e., a person) by judging
his or her character (in the second sense, i. e., moral and psychological properties
and dispositions). Narrative closure is usually an effect that is the terminus of a
causal process: conflict is normally presented as both an effect and the dynamo
of further causal actions; complications, in Aristotle’s sense, are causal turns of
fortune or misfortune; suspense, as the tension felt in facing the future, often
involves weighing the probabilities of different outcomes of the present situation
and thus is a clear case of causal expectation; and experientiality, interpreted as
implying the fact that narratives typically represent the character’s experiences,
is usually a causal affair.15 Chronology, finally, does not by itself belong to the
category of causation. But in order for internal effect expectation to function
during reading, chronological representation is presupposed. And it functions
extremely well: the reader’s inclination to project causal links between events
chronologically rendered is well known. Several commentators have noted that
Forster’s illustration of a story (“The king died. Then the queen died.”) is easily
given a causal interpretation.16 The example of a non-causal piece of information
from “Little Red Riding Hood” presented above may prompt a causal interpreta-
tion: the hood, given by the grandmother “who loved her most,” is most probably
made of a more exclusive and expensive material than wool or cotton.
Much of what Velleman says about the emotional cadence and its felt links
fits in here, too. The narrative motif of improbable reunions is an example of
a surprising event, at odds with rational expectation, and joke-telling is often
built upon something in conflict with ordinary causal anticipation that often
ends in resolution.17 The feelings of “anxiety relieved” and “hopes dashed” are

15 The term “experientality” refers to Monika Fludernik’s (1996) idea about what constitutes
narrativity. My formulation is no effort to catch the essence of the complex notion. I wish only to
point to one important corollary of her thesis. For a discussion of experientiality, see Caracciolo
16 John Pier has written an essay about this topic with the telling title “After this, therefore be-
cause of this” (Pier 2008).
17 See Hurley et al. (2011) on the IR Model of humor.
222   Göran Rossholm

usually connected with causal expectations, and the cadence of Oedipus Rex is a
sequence of emotions motivated by the causal process in the fictive process, not
only appearing in the emphatic audience. (More about this in the final section

8 Cadence and Expectation

Causal expectation is a promising candidate for a narrative universal: it explains
why causality is so prominent, but not ubiquitous, in narratives, and it also pro-
vides common ground for other narrative factors such as those mentioned in the
previous section. The most general background of the role of causal expectation
in the reading of narratives is the fact that humans are, to quote Hurley et al.
(2011, 251), “anticipation machines.” We are in a constant state of expectation,
as set out in the broad sense stipulated above. The narrator’s instrument is not
words (narratives can be non-verbal), but the reader’s readiness to form expec-
tations and the evaluation and interpretation of what happens relative to these
expectations, be they causal or otherwise.
As already mentioned, some stimuli of events to come in the course of the
story are purely external, and the relation is thus not causal. Some other non-
causal expectations have nothing to do with oncoming story events but are of
other kinds, for example, externally triggered expectations of what feelings are
awaiting us. But very often the outcomes of external expectations are interwoven
with the causal structure accomplished by causal expectation. It strikes me as
likely that the substance of Velleman’s observations about the emotive uptake
of (many) narratives has to do with this: combinations of causal and non-causal
expectations. The emotions favoured in Velleman’s cadences are all clearly
related to the psychology of expectation. When we become puzzled we usually
expect clarification, but when we experience grief—one of his examples of final
feelings—we are less prone to entertain expectations about what may come to
pass. The very idea of emotional cadences is an idea about anticipation and antic-
ipated endings with no further emotive strings attached. So described, cadential
expectations have nothing to do with expectations of causally connected events
in the story. There is a causal connection between the representation of story
events and the reader’s emotions, but no presumption that the reader’s emotive
expectations will be accompanied by any expectations about the unfolding of a
 Causal Expectation   223

causal plot.18 But this doesn’t seem right: the reader’s expectation is probably
much more complex. When the spectator of Murder, She Wrote expects that the
adjustable spanner will crush the skull of the presumed victim, s/he also expects
to experience a minor shock as well as a feeling of curiosity about who is holding
the spanner and why. In a longer perspective s/he expects that Mrs Fletcher will
present a convincing description of how and why all this has happened, and s/
he also foresees that the guilty party will be brought to justice as a consequence
of this and that, ultimately, these last two moments will induce the satisfying
feeling of closure. The expectations of the cadential and the emotional parts are
usually inseparable from causal expectations, and they are knit together by the
expectation that the causal events that are expected to occur will give rise to the
expected feelings.
Such composite expectations may also be part of the anticipatory repertoire
of the story of Mitys (“may”: keep in mind taht we don’t have the story, just a
synopsis). In addition to what I have suggested in the previous section, the
murder of Mitys may make the reader expect that the murderer will be punished
by someone other than Mitys himself: more exactly, that the killer will be, using
Flesch’s (2007) key phrase, altruistically punished because of the murder he has
committed. The reader also expects that, because of the realisation of this causal
expectation, she will experience relief, satisfaction and resolution, an emotion
that will foster no further emotional expectations. However, what happens is that
an event occurs which, even though (quoting Aristotle) “having the appearance
of design,” is not a punishment in the proper sense. The similarity to the conse-
quence of an intentional punishment evokes the expected feeling of resolution
(according to Velleman) even though the expected event never or only partly
occurs, a fact that modifies the final feeling so as to also include surprise. We
have, then, a split outcome: the causal component of the punishment is eluded,
but the physical effect is realised.19 Splits can be causal, and they can temporal
(the effect may be delayed or it may turn up too soon), and perhaps also other-
wise. Causal splits can be as in the story of Mitys (physical event right, causality
wrong), or they can be the other way around. When Aristotle praises surprising
yet causally integrated events (“Such incidents have the greatest effect on the

18 Regarding the causal link between event and emotion, Velleman says “a description of events
qualifies as a story in virtue of its power to initiate and resolve an emotional cadence in the au-
dience.” (2003, 18)
19 I don’t regard any expectation that is partly met and partly eluded as “split”: most outcomes
come true only in part. The example of the student in pyjamas given earlier may serve as an
illustration. The state of dress of the student is contrary to expectation, but not the fact that he
turned up. Still, this is no “split.”
224   Göran Rossholm

mind when they appear unexpectedly and at the same time in consequence of
one another”), he points to this latter category: events that meet the expectation
of causality, but not the expectation of the right event.

9 An Illustration
Equipped with this toolbox and with additional supplementary instruments of
his or her own making, the reader of this essay can set to work on a real narrative,
not a synopsis, so as to lay bare the anticipatory dynamics of narrative reading.
Just as the story about Mitys, this one can be read as being about divine interven-
tion, even though the final event may also be interpreted as pure happenstance.
The text below is taken from Mo Yan’s Nobel Lecture on 7 December 2012.20 Mo
Yan stressed that his profession was to tell stories, and he ended the lecture by
the following little story. I leave it, as Mo Yan did, without any further comment.

Bear with me, please, for one last story, one my grandfather told me many years ago: A
group of eight out-of-town bricklayers took refuge from a storm in a rundown temple.
Thunder rumbled outside, sending fireballs their way. They even heard what sounded like
dragon shrieks. The men were terrified, their faces ashen. “Among the eight of us,” one of
them said, “is someone who must have offended the heavens with a terrible deed. The guilty
person ought to volunteer to step outside to accept his punishment and spare the innocent
from suffering. Naturally, there were no volunteers. So one of the others came up with a
proposal: Since no one is willing to go outside, let’s all fling our straw hats toward the door.
Whoever’s hat flies out through the temple door is the guilty party, and we’ll ask him to go
out and accept his punishment.” So they flung their hats toward the door. Seven hats were
blown back inside; one went out the door. They pressured the eighth man to go out and
accept his punishment, and when he balked, they picked him up and flung him out the
door. I’ll bet you all know how the story ends: They had no sooner flung him out the door
than the temple collapsed around them.

20 The text is quoted from a link on the home page of the Swedish Academy.
sv_2012/nobel_lecture_2012_sv (Accessed 24 January 2014)
 Causal Expectation   225

10 The Pleasure of Surprise—A Physiological

Negative outcomes of causal effect expectations and the emotion accompany-
ing these outcomes, that is, surprise, have dominated this exposé of the Causal
Expectation Model. Why do expectations that are not realised play such a prom-
inent role in narratives, particularly in fictional narratives? Why does Aristotle
rank unexpected causally integrated events higher than expected causally inte-
grated events? Many answers will emphasize the importance of training one’s
sensitivity to apprehend causal structures, particularly intentional structures of
the reader’s co-species. To calibrate our intention-tracing nose, we should take
account not only of successful findings but also of mistakes.
I have no trouble with this way of reasoning. But in David Huron’s book Sweet
Anticipations: Music and the Psychology of Expectation there is a complement to
this that cuts deeper. Huron analyses musical anticipation, but he also explores
expectation in general in a biological and evolutionary perspective. In the first
two chapters he discerns two kinds of reactions to surprise, or misdirected
expectation, and they are both negative. One is the “response reaction” which
rewards successful anticipations and punishes unsuccessful ones with pleasant
and unpleasant feelings, respectively. The function is to encourage the forma-
tion of accurate expectations. Simultaneously, a first evaluation of the outcome
(unrelated to expectation) takes form: “the reaction response.” This very quick
neurological reaction, which is later supplemented and may be overridden by a
slow “appraisal response,” is basically pessimistic. It always assumes the worst
and thus detests surprises, all surprises. The function is obvious: when you are in
a dangerous environment you don’t have the time to make nuanced evaluations.
The most negative assumption is the best one if you want to stay alive. So fun-
damentally, surprises are bad for us and for all other animals. Still, in narratives
and in music, as in many other circumstances (e. g., playing peek-a-boo), we love
them. Why? Huron’s solution is physiological. In situations of stress, we release
endogenous opiates, which help us endure pain. The immediate negative attitude
towards any kind of surprise triggers such a defence reaction. A moment later,
the next instance, “the appraisal response,” informs us that what’s happening is
harmless or even intrinsically valuable. Huron calls this increase of pleasure felt
“contrastive valence”: the contrast between the initial negative response makes
the subsequent positive emotion stronger.21 Being under the influence of our

21 In Hurley et al. (2011, 253) this mechanism is called ”Huron’s backstage trampoline.”
226   Göran Rossholm

self-produced drug, we appreciate the situation more than we would without its
anticipatory pre-history. Surprise becomes enjoyable, and this is not necessarily
an instance of evolutionary adaption. It could just as well be seen as one of the
ways Homo sapiens has exploited its innate cognitive equipment—“the anticipa-
tion machine”—for new and joyful purposes.

Works Cited
Aristotle. 1952. “On Poetics.” Translated by Ingram Bywater. In The Works of Aristotle, vol. II,
681–699. Chicago: Encyclopaedia Britannica, Inc.
Caracciolo, Marco. 2014. The Experientiality of Narrative: An Enactivist Approach. Berlin and
Boston: De Gruyter.
Carroll, Noël. 2001. “On the Narrative Connection.” In N. C. Beyond Aesthetics: Philosophical
Essays, 118–133. Cambridge: Cambridge University Press.
Carroll, Noël. 1990. The Philosophy of Horror: Or, Paradoxes of the Heart. London: Routledge.
Currie, Gregory. 2010. Narratives and Narrators: A Philosophy of Stories. Oxford: Oxford
University Press.
Flesch, William. 2007. Comeuppance: Costly Signalling: Altruistic Punishment and Other
Biological Components of Fiction. Cambridge, MA: Harvard University Press.
Fludernik, Monika. 1996. Towards a ‘Natural’ Narratology. London and New York: Routledge.
Forster, E. M. 1962 [1927]. Aspects of the Novel. Harmondsworth: Penguin Books.
Gendler, Tamar Szabó. 2010. Intuition, Imagination and Philosophical Methodology. Oxford and
New York: Oxford University Press.
Genette, Gérard. 1997 [1982]. Palimpsests: Literature in the Second Degree. Translated by
Channa Newman and Claude Doubinsky. Foreword by Gerald Prince. Lincoln and London:
University of Nebraska Press.
Hurley, Matthew M., Daniel C. Dennett and Reginald B. Adams, Jr. 2011. Inside Jokes: Using
Humor to Reverse-Engineer the Mind. Cambridge, MA, and London: The MIT Press.
Huron, David. 2006. Sweet Anticipation: Music and the Psychology of Expectation. Cambridge,
MA, and London: The MIT Press.
Iser, Wolfgang. 1990. The Implied Reader: Patterns of Communication in Prose Fiction from
Bunyan to Beckett. Baltimore and London: The Johns Hopkins University Press.
Iser, Wolfgang. 1991. The Act of Reading: A Theory of Aesthetic Response. Baltimore and
London: The Johns Hopkins University Press.
Jauss, Hans Robert. 1982 [1977]. Toward an Aesthetic of Reception. Translated by Timothy Bahti.
Introduction by Paul de Man. Minneapolis: University of Minnesota Press.
Labov, William. 1972. Language in the Inner City. Philadelphia: University of Pennsylvania
Lewis, David. 1986. “Causal Explanation.” Philosophical Papers II, 214–240. Oxford: Oxford
University Press.
Mackie, John Leslie. 1965. “Causes and Conditions.” American Philosophical Quarterly 2:
Mackie, John Leslie. 1980. The Cement of the Universe. Oxford: Oxford University Press.
Marc-Wogau, Konrad. 1962. “On Historical Explanation.” Theoria xxviii: 213–233.
 Causal Expectation   227

Pier, John. 2008. “After this, therefore because of this.” In Pier and García Landa, eds.,
Pier, John, and José Ángel García Landa, eds. 2008. Theorizing Narrativity. Berlin: Walter de
Prince, Gerald. 2005. “Disnarrated, The.” In Routledge Encyclopedia of Narrative Theory,
edited by David Herman, Manfred Jahn and Marie-Laure Ryan, 118. London and New York:
Prince, Gerald. 2008. “Narrativehood, Narrativeness, Narrativity, Narratability.” In Pier and
García Landa, eds., 19–27.
Rossholm, Göran. 2007. “Story (first order predicate) logic.” Special issue: Semiotics of
literature. Semiotica 165 (1/4): 149–172.
Rossholm, Göran. 2012. “Narrative as Story Representation.” In Disputable Core Concepts of
Narrative Theory, edited by Göran Rossholm and Christer Johansson, 183–200. Bern: Peter
Sternberg, Meir. 1978. Expositional Modes and Temporal Ordering in Fiction. Indianapolis:
Indiana University Press.
Toolan, Michael. 2014 [2009]. “Coherence.” In Handbook of Narratology, 2nd edition, edited
by Peter Hühn, Jan Christoph Meister, John Pier and Wolf Schmid, vol. 1: 64–83. Berlin and
Boston: De Gruyter.
Velleman, J. David. 2003. “Narrative Explanation.” The Philosophical Review 112 (1): 1–25.
White, Morton. 1965. Foundations of Historical Knowledge. New York: Harper & Row.
Wolf, Werner. 2008. “Chance in Fiction as a Privileged Index of Implied Worldview: A
Contribution to the World-Modelling Function of Narrative Fiction.” In Pier and García
Landa, eds., 165–210.
Wolf Schmid (Hamburg)
Eventfulness and Repetitiveness:
Two Aesthetics of Storytelling

1 Jurij Lotman’s Concept of Event

The concepts of event and eventfulness open up a new area of inquiry in narra-
tology.1 However, these concepts are not peculiar to one of the many “new” or
“hyphenated narratologies” such as postcolonial, feminist, postmodern, trans-
generic narratology and others, but are rather at the core of “old” narratology. At
the same time, being culture-specific and historically changing phenomena of
narrative representations, event and eventfulness bridge the gap between “hard-
core” narratology and the “soft” history of culture and mentalities.
Central to this new dimension in narratology is the concept of event. What is
an event? As in the English word event, Ereignis in German, événement in French
and sobytie in Russian all denote an out-of-the-ordinary occurrence, something
which is not part of everyday routine.
Narratology employs a concept that has been used since the literature of
the early Italian Renaissance, when the novelty of a change of state began to be
regarded as a positive value. The leading genre of this novelty-based value was
the novella, and the leading author in this genre was Giovanni Boccaccio with the
Decameron. In this collection of novellas, the plot frequently involves the viola-
tion of a prohibition or the crossing of a boundary imposed by moral norms or by
the social order (Hühn 2014 [2009], 161). Such violations or boundary crossings
are characteristic of an event.
The first to speak of event as the defining property of the novella was Goethe.
In his conversations with Eckermann, Goethe asks, “What is a novella if not an
unheard-of occurrence that has taken place?”2 This question brings out both the
extraordinariness of the story told and its reality-based content.

1 Cf. Schmid (1992, 107–109; 2003; 2007; 2009a; 2010, 8–18; 2014, 12–30) and Hühn (2008; 2014
2 In German, as addressed to Eckermann, 25 January 1827: “Was ist eine Novelle anders als eine
ereignete unerhörte Begebenheit?”

DOI 10.1515/9783110555158-012
230   Wolf Schmid

However, no theory of event was developed either in the Renaissance or by

Goethe. The first to formulate a theory of events in literature3 was the Russian
literary scholar, semiotician, and cultural historian Jurij Lotman who was the
founder and head of the Moscow-Tartu school.
In his book The Structure of the Artistic Text (1977 [1970]), Lotman formulated
a theory of the sujet and of event that was inspired by the comparatist Aleksandr
Veselovskij and the formalist Viktor Šklovskij, but that is based more particularly
on Vladimir Propp’s functional approach in his Morphology of the Folktale (1968
[1928]; cf. Hauschild 2009).
The Russian term “sjužet” (sujet), which has manifold meanings in the writ-
ings of the three Russian proto-narratologists (cf. Schmid 2009b), is most often
translated as plot. However, plot here must not be understood in the sense of E. M.
Forster’s (1927) famous story-plot dichotomy. By “plot,” Lotman means an event-
ful action sequence with three components:

1) some semantic field divided into two mutually complementary subsets;

2) the border between these subsets, which under normal circumstances is impenetrable,
though in a given instance (a text with a plot always deals with a given instance) it
proves to be penetrable for the hero-agent;
3) the hero-agent. (1977 [1970], 240; original emphasis)

Lotman speaks of semantics in categories of space.4 His concept of space is

dichotomous and thus corresponds to binary semantic oppositions:

[As] a rule the principle of binary semantic opposition lies at the foundation of the internal
organization of textual elements: the world is divided up into rich and poor, natives and
strangers, orthodox and heretical, enlightened and unenlightened, people of Nature and
people of Society, enemies and friends. In the text, these worlds […] almost always receive
spatial realization: the world of the poor is realized in the form of a poor suburb, the slums
or attics, while the world of the rich is realized as Main Street, a palace, or the dress circle of
a theatre. Such notions arise as just and unjust lands, the antithesis of city and country, of
civilized Europe and uninhabited islands, of the Bohemian forest and paternal castle. The
classificatory border between opposing worlds assumes spatial features. (1977 [1970], 237)

3 The literary category of event is to be distinguished from the philosophical concept. For an
overview of the philosophical discussion on “events” as an ontological category, see Schneider
(2005), Casati and Varzi (2010) and Kaldis (2013).
4 Renner (1983) explains the concept of space, used metaphorically by Lotman, in terms of set
theory. For a critique of Lotman’s hermeneutic model and of Renner’s attempted inductive appli-
cation, see Meister (2003, 91–95). Lotman’s event categories are taken up and further developed,
in the sense of a formalization, by Krah (1999), Titzmann (2003) and Renner (2004). For a concise
overview of these approaches, see Hühn (2014 [2009], 172–173).
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   231

Lotman’s central definition of an event reads as follows:

An event in a text is the shifting of a persona across the borders of a semantic field. (1977
[1970], 282)

A seminal role in this definition is played by the concept of “border”:

As a rule […] all kinds of barriers in the text are concentrated at the border and structurally
always represent a part of the border. It is immaterial whether it be the ‘adversaries’ in a
fairy tale, the waves and winds and currents hostile to Odysseus, the false friends in a pica-
resque novel or the false clues in a detective novel, for in a structural sense they all have an
identical function: they make the movement from one semantic field to another extremely
difficult, and altogether impossible for everyone except the agent in a single given instance
[…]. (1977 [1970], 240–241)

The border can be topographical, but it can also be pragmatic, ethical, psycholog-
ical or cognitive. Lotman uses topological terms as the basis for his definition but
at the same time stresses the normative relevance of the definition by pointing
out that normative values tend to be described using spatial images and opposi-
tions. Thus, Lotman’s spatial semantics should be understood as a metaphor for
non-spatial, normative values. The normative character of the border is implied
in an alternative definition of the event in a plot:

[…] an event always involves the violation of some prohibition and is always a fact which
takes place, though it need not have taken place. (1977 [1970], 236)

On the basis of the presence or absence of an event, Lotman differentiates

between “plot texts” (sjužetnye teksty) and “plotless texts” (bessjužetnye teksty):

Texts without plots have a distinctly classificatory character; they establish a certain world
and its mode of construction. (1977 [1970], 236)

A second important trait of the plotless text is the fact that it insists on a definite order of
internal organization of this world. The text is constructed in a particular manner and it
does not permit its elements to move in such a way as to violate the established order. (237)

In contrast to plotless texts, which make borders fast, texts with a plot have a
“revolutionary” character insofar as they oppose established order.
On the basis of the opposition of texts with a plot and plotless texts, Lotman
gives an alternative definition of event:
232   Wolf Schmid

The movement of the plot, the event, is the crossing of that forbidden border which the plot-
less structure establishes. It is not an event when the hero moves within the space assigned
to him. (1977 [1970], 238; original emphasis)

Corresponding to the dichotomy of plot texts and plotless texts, two groups of per-
sonae can be distinguished: the immobile and the mobile. Immobile characters
are not permitted to cross the border whereas mobile characters are free to do so.
For the latter, Lotman gives four examples: Rastignac, a character from Honoré
de Balzac’s Comédie humaine making his way from the bottom of the social hier-
archy to the top; Romeo and Juliet crossing the border that divides their warring
houses; the hero breaking away from the home of his fathers to take his vows at a
monastery and become a saint; a hero breaking away from his social milieu and
going to the people to foment a revolution.
Lotman’s examples of plotless texts are calendars and telephone directo-
ries. But all kinds of descriptive texts also come within the type of plotless texts.
Another type of texts without a plot are mythological texts. Mythological texts do
not relate new developments in a changing world but represent the cyclical itera-
tions and isomorphisms of a closed cosmos, the order of which is fundamentally
affirmed by the text. What opposes them to narrative texts is that although they
represent changes of state, those changes are merely reiterated and thus do not
offer something really new.5
For Lotman, the plotless system is primary and can be embodied in an inde-
pendent text. The system with plot is secondary and always represents a layer
superimposed on a basic plotless structure. In a later article Lotman (1979 [1973])
argues that the modern “plot text” is the result of the interaction of the two basic
text types.
Lotman’s concept of event comprises one important point that we will have to
consider when it comes to historical or typological questions: what is regarded as
an event depends on context, world picture and point of view. As Lotman notes,

A plot is organically related to a world picture which provides the scale for determining
what constitutes an event and what constitutes a variant of that event communicating
nothing new to us. (1977 [1970], 234)

5 For the structure of mythical thought, cf. Cassirer (1955 [1925]), Lotman and Uspenskij (1973
[1970]) and Meletinskij (1976). For the manifestations of mythical thought in modern literature,
cf. Schmid (1987).
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   233

In support of this idea, Lotman gives an example: the case of a husband and
a wife who have quarreled about the value of abstract art and go to the police
station to have a report drawn up. The police inspector, having ascertained that
no assault, battery or any other transgression of civil or criminal law has taken
place, refuses to write up a report because from his point of view nothing hap-
pened. For a psychologist, moralist, social scientist or art historian, however, the
disputed fact constitutes an event.

The numerous arguments in the history of art concerning the comparative merits of various
plots arise because the same event represents something essential from one point of view,
something meaningless from another, and from yet another it does not exist. (1977 [1970], 234)

Discussing an example from Old Russian literature, where it is demonstrated

that even the death of a hero does not constitute an event in some texts, Lotman
touches on an important trait of events:

An event is that which did occur, though it could also not have occurred. The less probability
that a given event will take place (i. e., the greater the information conveyed by the message
concerning the event), the higher the rank of that event on the plot scale. (1977 [1970], 236)

In this somewhat incidental way, Lotman acknowledges the fact that events can
be characterized by different degrees along the scale of plot. By developing Lot-
man’s approach further, we can say that decisive factors for the plot-ranking of an
event are context and the ideological point of view of the beholder (be it a char-
acter involved in the action, the real author or the real reader). Context sensitivity
and dependence of the event on the perceiving subject are Lotman’s main points
for future discussions of the event as a hermeneutic phenomenon.

2 Event and Eventfulness

The concept of event should actually be somewhat broader than Lotman’s. An
event is not necessarily a violation of a norm. In the same way, the border need
not imply a prohibition. In place of border crossing, one should prefer the term
change of state.6

6 On the conditions constituting a change of state—1) a temporal structure with at least two
states; 2) an equivalence (similarity and contrast) between the states; 3) reference of the states to
one and the same acting or suffering subject)—cf. Schmid (2010 [2005], 2–5.
234   Wolf Schmid

Every event is a change of state, but not every change of state constitutes an
event. Event must be defined as a change of state that fulfills certain conditions.7
The first requirement of the event is that its associated change of state must
be factual or real (real, that is, in the framework of the fictional world). It follows
that changes of state that are wished for, imagined, or dreamed are not events
although the acts of wishing, imagining, or dreaming can qualify as events.
The second requirement of the event is resultativity, a correlate of the event’s
facticity. A change of state that constitutes an event must be resultative in that
it reaches completion in the narrative world. This means that a change of state
that constitutes an event is not inchoative (begun), conative (attempted), or dura-
tive (confined to an ongoing process) but rather is resultative in that it reaches
completion in the narrative world. This completion need not manifest itself in
concrete physical actions. For a mental event, it may suffice that a revision of
earlier opinions has taken place or that insight into certain circumstances has
been gained or lost.
Of course, the act of imagining, planning or similar acts can signal a mere
beginning or a faltering change in a character (Hühn 2014 [2009], 170). But in
each case the mental processes must conclude with a certain result.
Reality and resultativity are necessary conditions of an event. However, it is
clear that these requirements alone are not sufficient to turn a change of state
into an event, for they can both be fulfilled by trivial changes of state in a nar-
rated world. If a hero raises his hand, the change of state is both factual and
resultative. But as a rule this change will be inconsequential. In certain circum-
stances, however, the raising of a hand can be immensely meaningful. Imagine a
historical novel about ancient Rome. In the Colosseum a slight movement of the
emperor’s hand will decide the fate of many lives.
Events are more or less eventful. For this reason, eventfulness is a scalable
property of events. There are five features determining the degree of eventfulness:
1. Relevance. Eventfulness increases to the degree to which the change of state
is felt to be an essential part of the narrated world in which it occurs. Changes that
are trivial (in terms of the narrated world) rank low on the scale of eventfulness.
2. Unpredictability. Eventfulness increases in proportion to the extent to
which a change of state deviates from the doxa of the narrative (doxa corresponds
to what is generally expected in a narrated world). This does not mean that the

7 This is why Hühn’s (2014 [2009], 159) differentiation between “events I” (changes of state) and
“events II” (changes of a special kind) is not really helpful. The difference lies in the point of view
of the beholder taking a linguistic or hermeneutic stance rather than in the object itself. In the
narrated world, there are no borders delimiting events I from events II.
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   235

event must rest, as Lotman suggests, on the breaching of a norm or on the vio-
lation of a prohibition. Rather, the essence of the event can lie in the fact that it
breaks with expectations. A highly eventful change is paradoxical in the literal
sense of the word, for it goes “against expectation.”8
Doxa pertains to the story and its protagonists and their expectations and is
thus not equivalent to the reader’s script. A script or behavioral script (Schank
and Abelson 1977; Emmott and Alexander 2014 [2009]) is a sequence of expected
behaviors for a given situation. The classic script example is dining at a restau-
rant. The sequence of expected actions for this script begins with a hungry cus-
tomer entering the restaurant, ordering, eating, paying and then ends with the
customer leaving.
Doxa is what the protagonists expect from their lives, whereas script is what
the reader expects will occur in the action on the basis of certain patterns in lit-
erature or in the real world. A change of state that comes as a surprise to the
protagonists in a particular narrative world can be perfectly predictable for an
expert reader.
The wedding of the senior nurse with the chief physician of a hospital may
be a completely unforeseen event for both protagonists. But for the experienced
reader of doctor novels and hospital romances this happy ending is predictable
from the beginning. It follows that the reader’s script concerning the course of a
work and the protagonists’ expectations concerning the course of their lives must
be treated as distinct.
A change of state that can be seen to follow the normal rules of a narrated
world is predictable and thus will have a low level of eventfulness, even if it is of
great importance to the individual protagonists involved in it.
3. Persistence. The eventfulness of a change of state is in direct proportion
to its consequences for the thought and action of the affected subject within the
framework of the narrated world.
4. Irreversibility. Eventfulness increases with the irreversibility of the new
condition which arises from a change of state. That is to say, the more improbable
it is that the original condition can be restored, the greater the level of eventful-
5. Non-Iterativity. Repeated changes of the same kind, especially if they
involve the same characters, represent a low level of eventfulness, even if they
are both relevant and unpredictable with respect to these characters.
These five features are gradational and can be realized to varying degrees
(unlike binary features, which are unambiguously either present or absent).

8 Aristotle defines paradox as that which contradicts general expectation (Rhetoric 1412a 27).
236   Wolf Schmid

This means that events can have varying levels of eventfulness. There is no fixed
universal threshold of eventfulness that a change of state must cross in order to
become an event; conversely, no minimum level of eventfulness can be specified
below which events cannot exist.
At least two of the five features listed above, namely relevance and unpredict-
ability, are not objective, but rather depend on interpretation.
In a narrated world each of the protagonists can evaluate the relevance of a
change of state in different ways. But it is not only the protagonists in a narrated
world who can evaluate the relevance and unpredictability of a change of state in
different ways. The narrator and implied semantic instances such as the abstract
or implied author and the abstract or implied reader can also be perceived or con-
strued as evaluating the relevance of a narrated change of state in different ways.
Moreover, we need to take account of the fact that real readers may have their own
concepts of relevance and unpredictability that do not conform with those of the
fictive and implied instances.
It can be concluded that relevance and unpredictability are heavily dependent
on evaluation of the change of state by perceiving subjects. Each of the depicted,
narrating and reading instances is a subject in his or her own right within a par-
ticular social and axiological context that determines that individual’s norms and

3 Analyzing Eventfulness
How can eventfulness be analyzed? With every story, the reader will concentrate
primarily on the temporal links and their logic. This is why, in the interpretation
of a narrative text, the first thing to know is how the initial and final states of the
narrated world differ.
The ascription of meaning in the reading of narrative texts aims to identify
changes to the initial situation as well as the logic that underpins these changes.
However, not only the determining causes but also the changes themselves are
only rarely described explicitly and reliably and must therefore usually be recon-
structed. In order to reconstruct causes and changes, the reader is called on to
identify equivalences. Equivalence occurs with the presence of similarity and
contrast between two items. A change of state implies an equivalence between
the initial and final state of a story. In many cases, it is only non-temporal linking

9 Concerning the dependence of event on subject and context, cf. Schmid (2009a).
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   237

that brings temporal changes and their logic to the surface. For this reason finding
equivalences between episodes is a helpful tool for analyzing eventfulness (cf.
Schmid 2014).
Examples of events or temporal links being identified through the interplay
of non-temporal links are provided by the late stories of Čexov, which model the
life stories of their titular heroes as a chain of equivalent episodes: “The Grass-
hopper” (Poprygun’ja, 1892), “Ionych” (1898), “Darling” (Dušečka, 1899), “The
Lady with the Dog” (Dama s sobačkoj, 1899), and “The Betrothed” (Nevesta,
1903). The question in these stories is to know whether a full-fledged event—a
far-reaching change to someone’s situation—occurs or whether the same thing
merely repeats itself. In this regard “The Lady with the Dog” is the story most
disputed among interpreters. Does Gurov, the hero of this tale, really change,
converting from a notorious womanizer who sees women as an “inferior race” to
a sincerely loving man? This alternative can only be decided by confronting the
concealed similarities and contrasts between the episodes and the inner states
of the heroes. Temporal links therefore remain fundamental in a narrative work.
They are the aim of the reconstructive ascription of meaning, but it is often only
as a result of non-temporal links or equivalences that they reach a form accessible
to reconstruction.
It is not seldom that a change of state underlying a whole novel can be
tracked only in many small and seemingly inconspicuous steps. An example is
Thomas Mann’s novel Buddenbrooks (1901), in which the “decline of a family” is
portrayed, as stated in the subtitle of the German original. The changes between
the many steps, however inconsiderable they may seem, manifest themselves in
symptoms appearing not only in the characters and their behavior but also in
small details of the setting. In Mann’s Buddenbrooks such symptomatic details
form pairs of similarity and contrast that make the changes observable.
Another example of equivalence as a tool for reconstructing changes of state
is Jane Austen’s novel Pride and Prejudice (1813). At the heart of the novel are Eliz-
abeth Bennet—the second of the five unmarried Bennet daughters, twenty years
old, intelligent, lively, attractive—and Fitzwilliam Darcy, the wealthy owner of a
famous family estate. The course of Elizabeth’s and Darcy’s relationship is ulti-
mately decided when Darcy, who belongs to a higher social class than the Bennets,
overcomes his pride and Elizabeth her prejudice, leading them to surrender to
the love they feel for each other. Thus the novel’s central event is the twofold
mental change of state, the overcoming of an initial weakness of the heroes,
namely pride and prejudice. This drawn-out process can be conceived of on both
sides of the couple only by retracing the slightest changes in the heroes’ conver-
sations and reactions. But there is still another area where changes become man-
ifest. This can be seen in the portrayal of Elizabeth’s perceptions and reflections,
238   Wolf Schmid

marking Austen’s Pride and Prejudice as the beginning of the European novel of
consciousness. The focus of the plot is mental actions, and the central reflector of
these mental actions is Elizabeth, so that the reader sees the unfolding plot and
the other characters mostly from her position. Elizabeth’s perceptions are tinged
with her evaluative and linguistic point of view. Consequently, the mental events
forming the plot become accessible only by reconstructing Elizabeth’s changing
inner states. To do this, it is necessary to compare the forms and contents of Eliza-
beth’s external and inner speech, perceptions and shifting states of mind.
Once again, we can state that for analyzing event structures and forms of
eventfulness, it is helpful to draw on equivalences.
There is one simple guideline for reconstructing events: pay attention to the
beginning and the end of a story, and then look for all expressly stated or implied
similarities and contrasts between the initial and final states.
For an illustration of this analytical approach, we shall consider Aleksandr
Puškin’s tale “The Undertaker” (Grobovščik, 1831). At the beginning of the story,
the hero, an undertaker, is moving to a new house. But strangely enough,

As he approached the little yellow house that had enthralled his imagination for so long,
and that he had at last bought for a considerable sum, the old undertaker noticed with sur-
prise that his heart was not rejoicing. (Pushkin 1983 [1831], 87)

When the undertaker, waking up from his nightmare, learns that the merchant’s
widow, for whose death he could hardly wait, had not died the day before (i. e.,
on the day of his nightmare), he is, again unexpectedly, “much gladdened” (89).
Any interpretation of this tale must take account of the contrast between the par-
adoxical lack of joy at the beginning and the no less paradoxical presence of joy
at the end of the tale. The contrast between beginning and end gives the lie to
those interpreters (e. g., Ėjxenbaum 1987 [1919], 346) who maintain that in this
tale nothing happens and that nothing changes. In fact, something does happen,
but it is up to the reader to decide what.

4 Eventfulness in the History of Culture

Eventfulness began to be regarded favorably only in the literature of the Renais-
sance when philosophical and social thinking were freed from the bonds of a
closed doctrine. Under the reign of thinking in terms of salvation, it was con-
sidered that historical occurrences in the world were preordained, the need for
newness not being supported by doctrine.
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   239

The low esteem for newness in medieval thinking can be seen in Old Russian
literature (i. e., Russian literature up to the seventeenth century), which was
strongly influenced by religious thought. In Old Russian literature eventfulness
does not come forth as a positive quality. There is no unpredictability in hagiog-
raphy, the leading genre of the time. Of course, hagiographical texts do as a rule
represent changes of state, often culminating in miracles. Miracles, however, are
not genuinely surprising or unforeseeable in this textual world, for they follow
holy patterns modeled after the sacred, thus reaffirming the Christian world
order. In hagiography, miracles are a “plot need” (Lichačev 1987, 76) forming a
part of the script. Only miracles bring movement and development in the biogra-
phy of the saint compensating for the lack of psychological motivation (Lichačev
1987). Martyrdom, another subject of hagiography, or the conversion of a pagan
ruler, recounted in the Lives of the Princes, are formed according to beliefs in the
sacred and do not reflect genuine eventfulness. Essentially, Old Russian litera-
ture contents itself with small, relative, doxical changes of state or, when major
reversals are to be reported, it follows models that do not question the truth of
revelation, thus leaving little room for surprises.
A similar rejection or suppression of eventfulness can be observed under the
rule of more recent doctrines of salvation history or historical soteriology. Social-
ist realism appears at first glance to have been a development in which eventful-
ness thrived. The conversion of the doubter or the miscreant into a liberator of the
people who supports the right side in the struggle was one of the most popular
scripts in this kind of literature. On closer examination, however, this way of
thinking, with its similarities to salvation history, turned out to limit the possibil-
ity of border crossings just as much as the church literature of the Middle Ages.
Newness, unpredictability and surprise gained a positive value in the Tuscan
novella of the fourteenth century, a genre that highlighted a central event, a
turning point (cf. Pabst 1967, Polheim 1970, Thomé and Wehle 2000).
Eventfulness in this sense appears in Russia in some secular tales of the
seventeenth century that were influenced by western European novellas of the
Renaissance. These secular tales tell of morally dubious heroes and their border
crossings that are no longer punished at the end, as they were in religious tales.
The hero in The Tale of Frol Skobeev (between 1680 and the 1720s), for example,
is a poor nobleman who contrives to marry the daughter of a wealthy courtier.
He dresses as a woman in order to seduce the girl, marries her secretly and then
tricks her irate father into forgiving both of them. He is able to rise in society and
to become wealthy, all without any prospect of worldly retribution. For Lotman
this is an example of a mobile persona:
240   Wolf Schmid

The mobile persona is distinguished from the immobile persona because he is permitted to
act in certain ways forbidden to others. Thus Frol Skobeev’s norm of behavior is different
from that of those around him: they are bound to certain norms of morality which do not
restrict Frol Skobeev. From their point of view, Frol is a knave and a thief, and his behavior
“knavery” and “thievery.” From Frol’s point of view (and that of the seventeenth-century
reader of this tale) the behavior of his victims reveals their stupidity, and this justifies his
swindling them. The active hero conducts himself differently from the other personae, and
he alone possesses this right. (1977 [1970], 243)

The conditions for the rise of eventfulness in European literatures are anthropo-
centrism, freedom from thinking in categories of pre-ordained history, openness
to the new and joy of discovery.10

5 Repetitiveness
Now an uncomfortable question arises: if eventfulness and newness are so
important for narrative literature, is there only a negative role for familiar and
repeated things in stories? In other words, what part is played by the identical or
the uneventful implemented in stories? How does recognition, Viktor Šklovskij’s
antonym for defamiliarization,11 participate in aesthetic pleasure or, more pre-
cisely, in the pleasure of story reading?
Parents know that children never tire of listening to the same story again
and again and that they do not allow the slightest deviation in the sequence
of actions. Listening to the Brothers Grimm fairy tale “Mother Hulda,” children
insist that first the bread must be taken out of the oven and that only then can the
apple tree be shaken. Children tend to reject deviation or defamiliarization in the
telling of stories. Is that not a reason to assume a certain amount of pleasure from
familiarity and repetition?
Here is another example, this time from the world of adults. What fascinates
an experienced reader of hospital romances who knows in advance that the
heroine will marry the head physician?

10 For a short overview of the further history of eventfulness in Russian literature, cf. Schmid
(2010 [2005], 17–18. Hühn (2010) has submitted a study on eventfulness in British fiction.
11 “The purpose of art, then, is to lead us to a knowledge of a thing through the organ sight
[videnie] instead of recognition [uznavanie]. […] The life of a poem (and of an artifact) proceeds
from vision [videnie] to recognition [uznavanie], from poetry to prose […]” Šklovskij (1990 [1917],
6). For the various aspects of Šklovskij’s multifaceted concept of defamiliarization, cf. Schmid
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   241

The same question can be asked about crime series on television. What
surprise can arise here if, following the design of the series, it is clear from the
beginning that the criminal is played by the most prominent actor? If narration
exhausted itself in the satisfaction of the desire for the new and the unheard-of,
then what can explain the pleasure drawn from the familiar and the repeated?
Does the reader or spectator wind the clock back, so to speak, pretending not to
know the stereotype? Does he forget the script every time? Or does he concentrate
on other components of the work, for example “form”? This is perhaps true for
experienced readers rereading a piece of “high” literature, but it is hardly the case
for readers of hospital romances and viewers of serial thrillers.
With regard to the literature of the Middle Ages, one can ask what moved
readers within this ideological, closed system of thought that looked at the whole
of world history through the prism of the history of salvation and excluded all
paradoxical or unpredictable changes. On the one hand, the general human need
for surprise was satisfied by doxical changes that did not challenge or shake the
Christian world order while on the other hand, the aesthetics of repetition and the
pleasure derived from recognition dominated.
Obviously we have to take into consideration a particular kind of pleasure
that arises from the unfolding of an event even if its result is highly predictable
or if the same story is read or seen several times. We might even speak here of a
certain rituality, a mode that lets us perceive and appreciate the unfolding event
in all its stages and phases and in all its detail. In any case, the ritualistic percep-
tion of repeated occurrences is connected with immersion in the emotional reac-
tions of the characters exposed to positive or negative peripety. In this immersion,
the recipient can identify with the characters, plunging into their inner world to
experience the represented change of state as if it were his own.
What this means is that in addition to the aesthetics of deviation proclaimed
by the Russian formalists, we have to acknowledge an aesthetics of identification
and recognition. Of course, the two aesthetics have varying degrees of efficacy
in different spheres and epochs of culture and are experienced by recipients in
differing degrees.
This distinction can be compared to a distinction drawn by Jurij Lotman12
between an “aesthetics of opposition” (protivopostavlenie) and an “aesthetics
of identity” (toždestvo). To the latter belong works of art “whose structures are
given beforehand” and where “the audience’s expectations are met by the entire

12 First in his Lectures on Structural Poetics (1964, 171–181). Not translated into English, an
overview of this work is available in Shukman (1977, 65–68). German translation: Lotman (1972,
188–198), then in Lotman (1970, 349–359; 1977, 289–296).
242   Wolf Schmid

construction of the work” (Lotman 1970, 349; 1977, 289). Examples, according to
Lotman, can be found in the folklore of all nations, in medieval art, in commedia
dell’arte and in classicism. In all those genres and periods, works are constructed
according to certain scripts (as we would say today). In them, eventfulness, that
is, a relevant deviation from the script, would be regarded as a violation of the

If the author were to choose a situation that was “impossible” from the viewpoint of the
code rules within a given system of artistic perception, the structure expected by the audi-
ence would be destroyed, and as a result they would regard the work as inferior and the
author as unskilled and ignorant, or even a blasphemer of sinful audacity. (Lotman 1977
[1970], 289)

To the aesthetics of opposition belong works whose code is unknown to the audi-
ence even before the act of artistic perception begins. The audience is called on
to reconstruct the code from the text. Whereas the aesthetics of identity involves
simplification and generalization, the aesthetics of opposition tends toward
“complication” (usložnenie).
Although this second type is typologically and chronologically younger (it
received its greatest expression in realism), it is not superior to the first type axio-
logically. Both types have their virtues. Lotman underlines the historical relativity
of both types:

The observation of canons, norms, and clichés characteristic of the aesthetic of identity
does not annoy us or strike us as an artistic fault in the text of a folk epic or fairy tale. On
the other hand when we come across these same structural features in a modern social
novel we feel that we are dealing with an artistic failure, a falsification of the living truth.
The formula: “I predicted the whole structure” is deadly for a work of the second type [the
aesthetic of opposition], but has no effect on our evaluation of the first type [the aesthetic
of identity]. (1977 [1970], 294)

In our culture, works whose structure is predictable, such as hospital romances

or TV serial thrillers, tend to be regarded as having low aesthetic value. Never-
theless, such works may well provide consumers with a certain satisfaction that
relies not solely on entertainment.
Art tends to undermine the aesthetics of identity, not least by introducing
deviations from a script that seems to rule a work. This is one of the main devices
in Aleksandr Puškin’s Belkin Tales (1831). At first sight, these tales read like rep-
etitions of well-known sujets of sentimentalist or Romantic origin. Regarded as
imitations not worthy of their author, they were dismissed by the earliest critics.
But on closer examination, it appears that the stories deliberately sought to
undermine and refute the very scripts they appear to realize.
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   243

Nevertheless, we must admit that a certain portion of identity is contained in

every narrative, even in a sophisticated modern novel. There is no strict contra-
diction between the two aesthetics but rather a gliding scale or spectrum of forms
with different and frequently varying proportions of the two poles. Needless to
say, the relations between the two aesthetics will differ over epochs, cultures,
genres and types of readers. And clearly there are different kinds of pleasure to be
drawn from the two aesthetics.
The aesthetics of opposition, which is associated with a high degree of event-
fulness, produces a pleasure from newness, novelty, deviation and complexity,
thus challenging established schemata and stereotypes.
The aesthetics of identity, which enables easy recognition and prompts a ritu-
alistic form of reception, imparts a feeling of comprehensibility, straightforward-
ness and security.

Works Cited
Casati, Roberto, and Achille Varzi. 2010. “Events.” In The Stanford Encyclopedia of Philosophy,
edited by Edward N. Zalta.
(Accessed 14 December 2015)
Cassirer, Ernst. 1955 [1925]. The Philosophy of Symbolic Forms. Vol. 2: Mythical Thought.
Translated by Ralph Manheim. New Haven, CN: Yale University Press.
Ėjchenbaum (Eikhenbaum), Boris M. 1987 [1919]. “Boldinskie pobasenki Puškina.” In B. M. Ė. O
literature: Raboty raznyx let, 343–347. Moskva: Sovetskij pisatel’.
Emmott, Catherine, and Marc Alexander. 2014 [2009]. “Schemata.” In Hühn et al., eds., vol. 2:
756–764. Also available at: living handbook of narratology: http://www.lhn.uni-hamburg.
de/article/schemata (Revised 22 April 2014)
Forster, Edward Morgan. 1927. Aspects of the Novel. London: Edward Arnold.
Hauschild, Christiane. 2009. “Jurij Lotmans semiotischer Ereignisbegriff. Versuch einer
Neubewertung.” In Schmid, ed., 2009, 141–186.
Hühn, Peter. 2008. “Functions and Forms of Eventfulness in Narrative Fiction.” In Theorizing
Narrativity, edited by John Pier and José Ángel García Landa, 141–163. Berlin and New
York: Walter de Gruyter.
Hühn, Peter. 2010. Eventfulness in British Fiction. With contributions by Markus Kempf, Katrin
Kroll and Jette K. Wulf. Berlin and New York: De Gruyter.
Hühn, Peter. 2014 [2009]. “Event and Eventfulness.” In Hühn et al., eds., vol. 1: 159–178. Also
available at: (Revised 13
September 2013)
Hühn, Peter, Jan Christoph Meister, John Pier and Wolf Schmid, eds. 2014 [2009]. Handbook of
Narratology. 2nd edition, 2 volumes. Berlin and Boston: De Gruyter. Available online as
the living handbook of narratology at:
Kaldis, Byron. 2013. “Events.” In Encyclopedia of Philosophy and the Social Sciences, edited by
B. K. (Accessed 18 February 2014).
244   Wolf Schmid

Krah, Hans. 1999. “Räume, Grenzen, Grenzüberschreitungen. Einführende Überlegungen.”

Kodikas/Code 22 (1/2): 3–12.
Lichačev, Dmitrij. 1987. “Čelovek v literature Drevnej Rusi.” Izbrannye raboty v 3 tomax. T. 3,
4–147. Leningrad: Xudožestvennaja literatura.
Lotman, Jurij. 1964. Lekcii po struktural’noj poėtike.Vvedenie, teorija stixa (Trudy po znakovym
sistemam 1). Tartu: Tartuskij gosudarstvennyj universitet. Reprint: Providence, RI: Brown
University Press, 1968.
Lotman, Jurij. 1972 [1964]. Vorlesungen zu einer strukturalen Poetik. Einführung, Theorie des
Verses. Edited by Karl Eimermacher. Translated by Waltraud Jachnow. München: Fink.
Lotman, Jurij. 1977 [1970]. The Structure of the Artistic Text. Translated by Gail Lenhoff and
Ronald Vroon. Ann Arbor: University of Michigan.
Lotman, Jurij. 1979 [1973]. “The Origin of Plot in the Light of Typology.” Poetics Today 1 (1):
Lotman, Jurij, and Boris Uspenskij. 1973. “Mif – imja – kul’tura.” Trudy po znakovym sistemam
6: 282–303.
Meister, Jan Christoph. 2003. Computing Action. A Narratological Approach. Berlin and New
York: Walter de Gruyter.
Meletinskij, Ėleazar. 1976. Poėtika mifa. Moskva: Nauka.
Pabst, Walter 1967 [1953]. Novellentheorie und Novellendichtung: Zur Geschichte ihrer
Antinomie in den romanischen Literaturen. 2. Auflage. Hamburg: Cram, de Gruyter.
Polheim, Karl Konrad, ed. 1970. Theorie und Kritik der deutschen Novelle von Wieland bis Musil.
Tübingen: Max Niemeyer.
Propp, Vladimir. 1968 [1928]. Morphology of the Folktale. Translated by Laurence Scott, revised
by Louis A. Wagner. Austin: University of Texas Press.
Puškin (Pushkin), Alekandr. 1983 [1831]. “The Undertaker.” In A. P. Complete Prose Fiction,
translated with an Introduction and Notes by Paul Debreczeny, 87–93. Stanford: Stanford
University Press.
Renner, Karl Nikolaus. 1983. Der Findling. Eine Erzählung von H. v. Kleist und ein Film von G.
Moorse. Prinzipien einer adäquaten Wiedergabe narrativer Strukturen. München: Fink.
Renner, Karl Nikolaus. 2004. “Grenze und Ereignis. Weiterführende Überlegungen zum
Ereigniskonzept von J. M. Lotman.” In Norm – Grenze – Abweichung. Kultursemiotische
Studien zu Literatur, Medien und Wirtschaft. Michael Titzmann zum 60. Geburtstag, edited
by Gustav Frank and Wolfgang Lukas, 357–381. Passau: Karl Stutz.
Schank, Roger C., and Robert P. Abelson. 1977. Scripts, Plans, Goals and Understanding.
Hillsdale, NJ: Lawrence Erlbaum.
Schmid, Wolf. 1992. “Ereignishaftigkeit und ihre Einklammerung.” In W. S. Ornamentales
Erzählen in der russischen Moderne. Čechov – Babel’ – Zamjatin, 107–117. Frankfurt: Peter
Schmid, Wolf. 2003. “Narrativity and Eventfulness.” In What is Narratology. Questions and
Answers Regarding the Status of a Theory, edited by Tom Kindt and Hans-Harald Müller,
17–33. Berlin and New York: Walter de Gruyter.
Schmid, Wolf. 2005. “Defamiliarisation.” In Routledge Encyclopedia of Narrative Theory, edited
by David Herman, Manfred Jahn and Marie-Laure Ryan, 98. London: Routledge.
Schmid, Wolf. 2007. “Eventfulness as a Narratological Category.” In Amsterdam International
Electronic Journal for Cultural Narratology 4.
schmid.htm (Accessed 23 July 2013)
 Eventfulness and Repetitiveness: Two Aesthetics of Storytelling   245

Schmid, Wolf. 2009a. “Eventfulness and Context.” In Structuralism(s) Today. Paris, Prague,
Tartu, edited by Veronika Ambros, Roland Le Huenen, Adil D’Sousa and Andres
Pérez-Simón, 101–110. New York: Legas.
Schmid, Wolf. 2009b. “‘Fabel’ und ‘Sujet’.” In Schmid, ed., 2009, 1–45.
Schmid, Wolf. 2010 [2005]. Narratology. An Introduction. Translated by Alexander Starritt.
Berlin and New York: De Gruyter.
Schmid, Wolf. 2014. “Non-temporal Linking in Narration.” In Hühn et al., eds., vol. 2: 667–676.
Also available at:
(Revised 4 October 2013)
Schmid, Wolf, ed. 1987. Mythos in der slawischen Moderne (Wiener Slawistischer Almanach.
Sonderband 20). Wien: Gesellschaft zur Förderung slawistischer Studien.
Schmid, Wolf, ed. 2009. Slavische Erzähltheorie. Russische und tschechische Ansätze. Berlin
and New York: De Gruyter.
Schneider, Susan. 2005. “Events.” In Internet Encyclopedia of Philosophy. http://www.iep.utm.
edu/events (Accessed 14 December 2015)
Shukman, Ann. 1977. Literature and Semiotics. A Study of the Writings of Yu. M. Lotman.
Amsterdam: North Holland.
Šklovskij (Shklovsky), Viktor. 1990 [1917]. “Art as Device.” In V. S. Theory of Prose, translated by
Benjamin Sher, 1–14. Elmwood Park, IL: Dalkey Archive Press.
Thomé, Horst, and Winfried Wehle. 2000. “Novelle.” In Reallexikon der deutschen
Literaturwissenschaft, Neubearbeitung, edited by Harald Fricke et al., Bd. II: 725–731.
Berlin: De Gruyter.
Titzmann, Michael. 2003. “Semiotische Aspekte der Literaturwissenschaft: Literatursemiotik.”
In Semiotik/Semiotics. Ein Handbuch zu den zeichentheoretischen Grundlagen von Natur
und Kultur. A Handbook on the Sign-Theoretic Foundations of Nature and Culture, edited by
Roland Posner, Klaus Robering and Thomas A. Sebeok, vol. 3: 3028–3103. Berlin: Walter
de Gruyter.
Raphaël Baroni (Lausanne)
The Garden of Forking Paths:
Virtualities and Challenges for
Contemporary Narratology

1 A Forking Path for “Classical” and

“Postclassical” Narratologies
Nowadays, it has become almost impossible to deal with central issues in nar-
rative theory such as narrative sequence, plot, tellability, narrative interest, or
even narrativity without investigating how the story is embedded in a complex
network of virtual fabulas. By contrast, these questions were largely neglected
in formalist and structuralist theories. As stated by David Herman: “[Propp’s]
approach gave an overly deterministic coloration to narrative sequences […]. Part
of the interest and complexity of narrative depends on the merely probabilistic,
not deterministic, links between some actions and events” (2002, 94). Along the
same line, Hilary Dannenberg adds: “An analysis of narrative’s story tells us very
little about the true dynamics of plot and about the fascination of fictional worlds
for the reader; this stems from the fact that narrative does not simply tell one
story, but weaves a rich, ontologically multidimensional fabric of alternate possi-
ble worlds” (2004, 160).
With the emergence of the analysis of “alternate possible worlds” as a central
field of investigation, narratology has departed from its original formalist par-
adigm and overcome its methodological limitations. As pointed out by Emma
Kafalenos in an attempt to describe the evolution of contemporary narratology,
recent studies tend to highlight more the role of the reader and the indetermi-
nations of narrative: “What I see as new […] is the specificity of the analysis of
how readers’ decisions contribute to the construction of the narrative world. […]
Further developments along this path, if it occurs, will bring us an increasingly
precise account of sites where indeterminacy can enter a narrative representa-
tion, and of conditions that heighten the interactivity between representation
and reader in constructing narrative worlds” (2001, 114).
If this evolution can be seen as a general trend in narrative theory, and if
the analysis of “alternate possible worlds” appears, at first sight, to be a coher-
ent field of investigation for contemporary narratology, the unity of this field,
as well as the complementary or convergent nature of the theories dealing with
these questions, cannot be taken for granted. Reviewing the various ways nar-

DOI 10.1515/9783110555158-013
248   Raphaël Baroni

rative virtualities have been examined offers a good opportunity to describe the
current state of narrative theory. By doing so, I shall try to highlight the forces at
play, those encouraging the consolidation of the discipline or, conversely, those
pushing towards diversification.

2 Ts’ui Pên’s and Borges’s Garden of Forking

A short story by Borges entitled “The Garden of Forking Paths” will be used in
order to illustrate the various ways we can deal with the virtualities of narra-
tive. This fiction can be considered both a metanarrative and a spy story, and
both aspects will be useful for the following discussion. On the metanarrative
side, Borges refers to the work of Ts’ui Pên, a Chinese ancestor of the protag-
onist who has supposedly written a book titled The Garden of Forking Paths, a
book described as a “maze” and as a “labyrinth of time.” One character, Professor
Stephen Albert, explains that “In all fictions, each time a man meets diverse alter-
natives, he chooses one and eliminates the others; in the work of the virtually
impossible to disentangle Ts’ui Pên, the character chooses—simultaneously—all
of them. He creates, thereby, ‘several futures’, several times, which themselves
proliferate and fork” (Borges 1999 [1941], 125).
Ts’ui Pên’s book has often been associated with interactive storytelling, espe-
cially hypertext fictions and digital media.1 Indeed, in these kinds of narratives,
all alternatives have been programmed or written and fully belong to the struc-
ture of the work itself, even though the reader (or the player) actualizes only one
path, leaving the other paths unexplored. Still, we can notice that Ts’ui Pên’s
book, weird as this fiction is, is defined as a “novel.” Borges further highlights
the fact that “alternatives” belong to “all fictions.” Even though these alternatives
are not textualized, they can be imagined, both for the character, who is planning
his next move, and for the reader, who is wondering how the narrative is going
to develop. More generally, it is entirely possible to state that all narratives, since
they are representations of actions carried out by characters and addressed to
audiences, are interactive and comprise alternative “paths.”
On the spy story side of the tale, the narration recounts a fictional event
that occurred during the First World War. The story is in homodiegetic narra-
tion, purportedly the fragment of a deposition, “dictated, reread, and signed by

1 See for example Moulthrop (1991) and Ryan (2006a).

 The Garden of Forking Paths: Virtualities and Challenges    249

Dr. Yu Tsun, former professor of English in the Hochschule at Tsingtao” (119). At

the same time, Yu Tsun is a Chinese spy working for the Germans, and when the
story begins, he is about to be arrested by Captain Richard Madden, an Irish-
man working for the British Government. After discovering that the identity of his
accomplice, Viktor Runeberg, has been exposed, Yu Tsun loses all hope of saving
his life. Nevertheless, he decides to accomplish a final mission. He succeeds in
communicating the name of a place where the English artillery is located, a city
called Albert. In order to carry out his mission, Yu Tsun kills a man whose name is
Dr. Stephen Albert. The assassination of Dr. Albert is reported by the newspapers
and thus comes to the attention of the German authorities. Unfortunately for Yu
Tsun, his victim turns out to be a friendly man and a sinologist doing research on
the work of Ts’ui Pên, one of his ancestors. So the Chinese spy concludes his con-
fession by stressing that he has “abhorrently triumphed” (127) and he expresses
his “endless contrition” and “weariness” (128).

3 Possible World(s) Asserted by the Author

Borges’s story not only reflects on narrative virtualities, but it also actualizes all
kinds of virtual worlds that we can find in narrative. We begin our survey with
the kind of virtualities focused on by Thomas Pavel (1975) in his work on possible
worlds and modal logic. At this stage, we can consider the narrative world as a
whole, and not in its progressive actualization. Umberto Eco (1984, 235) insists
on the fact that, even if possible worlds are actualized through a series of stages,
all these moments ultimately belong to the same possible world as the author
planned it: a possible world in an assertive mode that must be contrasted with
other possible worlds imagined by the characters in the story or by the readers.
Here, modal logic provides useful tools for explaining how fictional worlds are
shaped, based on information we possess concerning the real world. Thus, from
the perspective of a phenomenology of reading, the “possible” must be consid-
ered a “parasite” of the “actual.” As Ryan puts it: “when readers construct fic-
tional worlds, they fill in the gaps in the text by assuming the similarity of the
fictional world to their own experiential reality. This model can only be overruled
by the text itself; thus if a text mentions a blue deer, the reader will imagine an
animal that resembles her idea of real deer in all respects other than the colour”
(2005a, 447).
Of course, we know that the fictional characters of the story by Borges are
human beings endowed with attributes that don’t have to be mentioned such as
the fact that people normally have two arms and two legs. But we also receive
250   Raphaël Baroni

relevant information about the characters. For example, we learn that the antag-
onists are Chinese and Irish and thus that they are working for enemies, i. e., the
Germans and the English. We can infer further information from the first lines of
the story:

On page 242 of The History of the World War, Liddell Hart tells us that an Allied offensive
against the Serre-Montauban line (to be mounted by thirteen British divisions backed by
one thousand four hundred artillery pieces) had been planned for July 24, 1916, but had to
be put off until the morning of the twenty-ninth. Torrential rains (notes Capt. Liddell Hart)
were the cause of that delay—a delay that entailed no great consequences, as it turned out.
The statement which follows—dictated, reread, and signed by Dr. Yu Tsun, former professor
of English in the Hochschule at Tsingtao—throws unsuspected light on the case. (119)

There is no reason to wonder whether the book by Liddell Hart truly exists or to
question what is truly written on page 242 of this book. What does matter here
is what we should infer from the mention of a history book, based on what we
know concerning the real world. Indeed, by mentioning a well-known historical
context, the incipit provides valuable information we need in order to fill in the
gaps of the storyworld. For example, we know that the story begins in the middle
of a war that broke out in Europe in 1914 and that Captain Liddell Hart’s book (a
controversial book that truly exists, by the way, published in 1930) represents a
version of the winners. We also know that in time of war stories are often told by
unreliable narrators or that they are based on unreliable sources such as propa-
ganda, for example. All of this information is important in order to understand
the fictional text, correctly infer some of its implicit elements and interpret the
text in a productive way.
It is interesting to notice that Borges suggests that we should call Captain
Liddell Hart’s version into question on the grounds that history books can never
be considered to constitute an exhaustive, completely reliable and definitive
version of what happened. Yu Tsun’s deposition might be a fictional document,
but in real life too, the way we understand history constantly changes. Borges
adds a further level of complication between two versions of the same fact. In
his deposition, Yu Tsun states that “Runeberg had been arrested, or murdered”
(119). However, an editor’s note comments on this hypothesis as follows: “A
bizarre and despicable supposition. The Prussian spy Hans Rabener, alias Viktor
Runeberg, had turned an automatic pistol on his arresting officer, Capt. Richard
Madden. Madden, in self-defense, inflicted the wounds on Rabener that caused
his subsequent death” (119). In this case, obviously, the editor seems to possess
more information than the narrator, a fact that he proves by revealing the real
name of his accomplice. So we must admit that Runeberg was probably killed,
and not arrested. But for the reader, how Runeberg was killed remains undeter-
 The Garden of Forking Paths: Virtualities and Challenges    251

mined. Yu Tsun’s hypothetical version might be partial, but the editor’s version
may have been manipulated by official sources in order to enable the agent to
plead self-defense. I think that the fabula allows us to choose between the two
versions—self-defense and murder—with the result that no assertion, fictional or
factual, can be left undisputed, even though some assertions seem to be more
credible than others.
This last point illustrates a property of possible worlds semantics: the fact
that, in some cases, it is possible to establish a hierarchy between opposing state-
ments concerning the fabula, while in other cases we must accept the existence of
multiple alternative worlds, since they cannot be reduced to mutually exclusive
versions. This plurality of worlds can also be related to the history of narrative
forms. As stated by Françoise Lavocat: “if the work stipulates the existence of
many possible worlds, their modalities of engendering and configuration differ
according to historical periods” (2010, 8; translation mine). Hilary Dannenberg
also suggests that “An analysis of the historical development of plot shows how,
with the rise of the novel, more sophisticated plots develop involving the tempo-
ral orchestration of alternate world versions: more than one version of the past or
future is suggested as a possibility by the text” (2004, 161).2 In this case, we could
state that the indeterminate elements in Borges’s story partly result from the fact
that it was written in the middle of the twentieth century and that it is a parody
of the paranoid genre of spy stories. Indeed, Dannenberg adds that “Fictional
genres across the board, whether realist, semi-realist (fantasy, science fiction) or
anti-realist (metafictional), all use alternate possible worlds, but with differing
forms of ontological hierarchy” (2004, 161).
Here again, we can stress that, at this level, we are concerned with “possible
worlds” asserted by the fiction: worlds that can be multiple or single, indeter-
minate or finite, indecisive or hierarchical. Either we compare the fiction with
an external referent—with infinite possibilities but at the same the uncountable
constraints of the real world—or we aim to contrast different versions of the same
fact inside the possible world of the fiction. And in the whole process, either we
find a clear hierarchy between the alternatives or we don’t. Now, I shall try to be
more specific on the nature of other kinds of virtualities that can be actualized,
implicitly or explicitly, by narrative fictions.

2 Here, we can historicize what Brian Richardson defines as “unnatural narratives”: those “rep-
resentations that contravene the presuppositions of nonfictional narratives, violate mimetic con-
ventions and the practices of realism” (2015, 3).
252   Raphaël Baroni

4 Virtualities Expressed by the Discourse

The first kind I shall comment on, and probably the most obvious one from a
formalist perspective, is explicit description by the narrative of events that don’t
belong to the fabula. In 1988, Gerald Prince introduced the category of the “dis-
narrated” in order to deal with this narrative modality that departs from mere
assertions. Prince’s definition runs as follows: “the category of the disnarrated
covers all the events that did not happen but, nonetheless, are referred to (in
a negative or hypothetical mode) by the narrative text” (1988, 2). Here, we see
that we are dealing with virtualities that fully belong to the text. The fundamen-
tal criteria are based on the explicit nature of the virtuality and non-assertive
tone of the discourse. The disnarrated can be either: (a) an unrealized possibil-
ity imagined by a character; (b) a forking path in the realm of the possible out-
lined by the narrator; or (c) a narrative possibility not chosen by the creator of the
textual universe but which is mentioned by an authorial voice. In “The Garden
of Forking Paths,” we can find examples of such disnarration actualized by the
main character. The first example expresses a false guess about the antagonist,
the other a hope about the future:

For one instant, I feared that Richard Madden had somehow seen through my desperate
plan, but I soon realized that that was impossible. (Borges 1999 [1941], 122)
I told myself that my duel had begun, and that in dodging my adversary’s thrust—even by
forty minutes, even thanks to the slightest smile from fate—the first round had gone to me.
I argued that this small win prefigured total victory. I argued that the win was not really
even so small, since without the precious hour that the trains had given me, I’d be in goal,
or dead. I argued (no less sophistically) that my cowardly cheerfulness proved that I was a
man capable of following this adventure to its successful end. (121)

In both cases, we see a sharp discrepancy between the hope of the character and
the knowledge expressed by the narrator, since the latter knows retrospectively
that things will end up tragically. But the narrator can also express his opinion
about alternatives in the past, as in the following statement: “Madden’s presence
in Viktor Runeberg’s flat meant the end of our efforts and (though this seemed to
me quite secondary, or should have seemed) our lives as well” (119).
 The Garden of Forking Paths: Virtualities and Challenges    253

5 Virtualities Shaped by the Characters and

Belonging to the Fabula
Marie-Laure Ryan has proposed a different way of dealing with virtualities. She is
not concerned with the problem of describing all kinds of virtualities expressed
by narrative discourse, but rather with analyzing the links between the “tellabil-
ity” of the fabula and its logical complexity:

[T]ellability is rooted in conceptual and logical complexity, and […] the complexity of a
plot depends on an underlying system of purely virtual embedded narratives. Embedded
narratives […] are the story-like constructs contained in the private worlds of characters.
These constructs include not only the dreams, fictions, and fantasies conceived or told by
characters, but any kind of representation concerning past or future states or events: plans,
passive projections, desires, beliefs […]. Among these embedded narratives, some reflect
the events of the factual domain, while others delineate unactualized possibilities. (Ryan
1991, 156)

As summarized by Dannenberg, Ryan (in contrast to Prince) “admits counterfac-

tual worlds into this category only if they are a product of the speculative activity
of a character” (Dannenberg 2004, 172). Indeed, Ryan argues that “the disnar-
rated [produced by the narrator] could be deleted from the text without conse-
quence for the logical coherence of the narrative events” (1991, 169). Additionally,
she also includes implicit virtualities that must be reconstructed by the reader, as
for example when a character’s intention is not specified. In the example already
mentioned, Yu Tsun makes a supposition concerning the past based on infor-
mation that he has just received: “Madden’s presence in Viktor Runeberg’s flat
[…] meant that Runeberg had been arrested, or murdered” (Borges 1999 [1941],
119). These suppositions (even though some of them are proven to be false) are
extremely important in order to understand the course of the narration, since the
future actions of the protagonist are based on this diagnosis concerning the past.
The second example shows an explicit intention of the protagonist, the narrator
explaining that it won’t be followed by action:

Something—perhaps the mere show of proving that my resources were non-existent—made

me look through my pockets. I found what I knew I would find: the American watch, the
nickel-platted chain and the quadrangular coin, the key ring with the compromising and
useless keys to Runeberg’s flat, the notebook, a letter which I resolved to destroy at once
(and never did). (120, emphasis added)

The last example refers to a “plan” that is mentioned, though not explicitly, since,
to preserve narrative interest, it is not yet revealed how the goal will be achieved:
254   Raphaël Baroni

If only my throat, before a bullet crushed it, could cry out that name so that it might be
heard in Germany. But my human voice was terribly inadequate. […] I vaguely reflected that
a pistol shot can be heard at a considerable distance. In ten minutes my plan was ripe. The
telephone book gave me the name of the only person able to communicate the information:
he lived in a suburb of Fenton, less than a half hour away by train. (120)

In distinguishing “tellability” from “strategic point,” Ryan highlights the differ-

ence between narrative interest that relies on the complexity of the plan devised
by the character and interest that relies on discourse strategies, for example the
fact that the plan is momentarily unclear for the reader:

Narrative suspense derives, for instance, from the confrontation of characters of limited
foresight and a reader who anticipates—correctly or not—the situations into which they
should run. The reverse strategy is also an efficient way to capture the reader’s interest:
delaying the reader’s understanding of a sequence of actions by preventing access to the set
of embedded narratives that motivate the agent. While the plot sets up a field of possibili-
ties, the strategies of narrative discourse may guide the reader along certain paths. (Ryan
1991, 174)

Ryan stresses the differences between virtualities shaped by the characters and
those shaped by readers. This brings us to the next stage of our investigation.
Here, the focus will be on gaps in the narrative that elicit active participation
on the part of the interpreter by making inferential walks into the woods of the

6 Virtualities Shaped by the Reader and their

Esthetic Effects
Several narratologists have distinguished between two types of narrative inter-
est according to the temporal orientation of the virtual scenarios shaped by the
readers. The first was Tzvetan Todorov, who made this distinction with regard to
detective fiction:

The first can be called curiosity, it works from effects to causes: starting from a certain effect
(a corpse and some clues), we must find its cause (the culprit and what drove him to the
crime). The second form is suspense, and it works from causes to effects: first we are intro-
duced with the causes, initial data (some gangsters who prepare a mischief), and our inter-
est is elicited by the expectation of what will happen, in other words, the effects (corpses,
crimes, clashes). (Todorov 1971, 60; translation mine)
 The Garden of Forking Paths: Virtualities and Challenges    255

Of course, this distinction can be generalized far beyond the scope of detective
stories; accordingly, Dannenberg argues that “complex novels involve the inter-
weaving of possible versions of both past and future world” (2004, 160). Along the
same lines, authors such as Meir Sternberg (1978), Peter Brooks (1984) and James
Phelan (1989) have investigated the links between the affective engagements of
the reader and progression in the text, including analysis of the strategies used to
“generate, sustain, develop, and resolve readers’ interests in narratives” (Phelan
1989, 15). Among these strategies, Sternberg focuses especially on “expositional
modes and temporal ordering” that can be linked to suspense, curiosity and sur-
prise, considered as universal functions of narratives.
I won’t discuss here the differences between these authors, but I want to
stress that, overall, their theories are limited to considering the relations between
certain narrative techniques and the two major kinds of interpretive procedures
that they elicit in the reading process: prospection and retrospection. Other
studies have taken a more adventurous path by trying to provide a description
of how prospection and retrospection are transformed into concrete hypothetical
scenarios in the reader’s mind. Umberto Eco (1984) opened this field of investi-
gation up by linking the inferential walks of readers with modal logic, possible
worlds semantics, Peircean semiotics and the more general notion of encyclope-
dia. Eco was building on the pioneering works of Thomas Pavel (1975) and Lucia
Vaina (1977), but his model already comprised three levels: 1) the possible worlds
asserted by the author; 2) the possible worlds shaped by the characters; and 3)
the possible worlds inferred by readers based on “common” and “intertextual
frames” (Eco 1984, 32).
Since then, Bertrand Gervais (1990) and David Herman (1997), among others,
have adopted cognitive models developed in AI studies to describe how our knowl-
edge about the logic of actions can shape how we interpret narratives. Among
these frames we find notions like “scripts” and the schematization of intentional
actions. According to these studies, narrative interest results from destabilizing
a stereotyped situation,3 and its development generally involves the realization
of a more or less complex series of planned actions leading to a specific goal.
Emma Kafalenos (2006) has also used an abstract model, inspired by Propp’s
functions, in order to define the causal configurations that readers build pro-
gressively while they progress through the text. Building on Sternberg’s model,
Kafalenos especially highlights the effect of deferred or suppressed information
on this ever-changing reconstruction of the story logic. In other models, authors
have stressed the intertextual knowledge that orients the inferential walks of the

3 On stereotypes, see also Dufays (2010).

256   Raphaël Baroni

readers. Among them, John Pier has shown that “intertextual frames” (2004, 240)
contribute to narrative configurations on the basis of abductions concerning the
past, present and future of the fabula. In La Tension narrative, I have combined
these different perspectives: the rhetorical investigation of textual devices arous-
ing narrative interest together with a description of the anticipations entertained
by interpreters in the process of narrative configuration, including the full range
of “endo-narrative skills”4 (semantics of action) and “transtextual knowledge”
(Baroni 2007, chap. 3 and 4).
Let us see now how this approach can be applied to our case. From the per-
spective of Yu Tsun, his spying activity is a routine regulated by scripts. The rev-
elation of his identity opens a less predictable development and thus increases
the interest of the narrative progression. From this point on, the story revolves
around two parallel goals which involve the planning of a series of actions: the
first, seemingly impossible, is to escape the deadly chase by Captain Richard
Madden; the second, that will be successful, is to convey secret information to
a distant recipient.5 The difficulty of the task builds suspense in the chronolog-
ical unfolding of the narration. In addition, the strategic delay of the revelation
of the plan devised by the protagonist, as already mentioned, helps to increase
narrative tension (in this case curiosity) by urging the reader to make hypotheses
concerning how the goal might be achieved. The genre of “spy stories” also pro-
vides useful configurations which help the reader to explore the virtualities of
the narrative. For example, we recognize that the “transmission of a secret” and
the “chase” are stereotypes usually found in this narrative genre, enabling us to
refer to previous fictions in order to formulate hypotheses as to how the conflict
may develop or how it may be resolved. The genre also induces indeterminacies
in the virtual scenarios of the reader, since stereotypical spy stories involve the
paranoid assumption that no one should be trusted and that the final resolution
must be unexpected.
Ryan mentions yet another important aspect concerning the virtualities
shaped by readers, namely, their resistance to reiteration: “Even after the pos-
sible has been exhausted by the actualization of a certain course of action, the
interpreter revisits mentally the paths that have fallen into the domain of the
counterfactual, so as to assess the ethical or strategic decisions of the character,
as well as the aesthetic decisions of the author” (Ryan 2005b, 628). In an involun-

4 For my use of AI studies, see Baroni (2002). For the relation between intertextual and actional
frames, see Baroni (2005, 2016).
5 It can be noted that the “quest” of the hero is both successful and unsuccessful, adding coex-
tensive virtualities to the story.
 The Garden of Forking Paths: Virtualities and Challenges    257

tarily ironical statement, Doctor Albert says: “Time forks, perpetually, into count-
less futures. In one of them, I am your enemy” (Borges 1999 [1941], 127). And Yu
Tsun replies, just before killing his interlocutor: “But I am your friend” (127). I
think that the reader will have to admit that in the possible worlds of the fiction,
Stephen Albert and his guest were both friends and enemies at the same time
and also that, even if the crime has been committed (and the future is already
written), Yu Tsun could or should have spared his friend. This is precisely this
alternative that makes the crime of Yu Tsun an “abhorrent” triumph, causing
“endless contrition” and “weariness.” So an ethical judgment is always combined
with the comparison between what happened in the fabula and what else could
have happened.

7 The Theory of Possible Texts

A study of all forms of virtualities that can be associated with stories would be
incomplete if we only considered virtual paths that are intentionally incorporated
into the text by the author. Marc Escola sees the possibility for a new kind of
critic, called the “interfering critic” (critique interventionniste), inspired by the
works of Pierre Bayard, a scholar who wrote academic bestsellers such as Sher-
lock Holmes Was Wrong: Reopening the Case of the Hound of the Baskervilles, and
How to Improve Failed Literary Works.

The contributors to this Theory of Possible Texts have learned from P. Bayard that a reader
is under no obligation to adopt the conclusions which the author claims to be the final
ones, that literary criticism can be a continuation of the creation by other means, and that
ultimately there is not a big gap between the reading of a work and its reinvention or its
refection. (Escola 2012, 10; translation mine)

We can apply this “theory of possible texts” to Borges’s story in various ways.
For example, when we read that the first two pages of Yu Tsun’s deposition
are missing, we can suppose that this hidden part of the story might shed an
unexpected light on the events. Some creative readers may even try to write or
comment on these missing pages, and by doing so, invent an alternative version
of the facts. We might also focus on apparent inconsistencies in the story. For
example, we can cast doubt on the truthfulness of Yu Tsun’s confession by point-
ing out the unlikely coincidence that the victim, chosen at random in a telephone
book, turns out to be linked to his killer in a surprising way. The story would
then have to be rewritten according to a more credible scenario. After all, the
whole deposition of the narrator could be a lie, a complete invention in order to
258   Raphaël Baroni

protect some darker secrets (maybe Doctor Albert was an accomplice who had to
be killed for some other reason). Here, we are still talking about the virtualities of
the narrative, but the initiative has been transferred from the text (and the hypo-
thetical intention of an implied author) to an empirical reader, who can exploit or
comment on the text and make it take unexpected directions.

8 Is Contemporary Narratology a Garden of

Forking Paths?
Of course, the virtualities of narratives represent only one issue among many
others for contemporary narratology. At the same time, I think that the episte-
mological problems raised by the various ways that virtualities have been dealt
with casts light on a number of important issues faced by the discipline in its
recent history. As I have shown, research on narrative virtualities is itself a kind
of garden of forking paths. What looked, at first sight, like a simple problem, has
produced a number of distinct, and sometimes opposed, theories. Tensions may
occur when different paradigms describe similar phenomena, for example when
a disnarrated event is also an embedded narrative in the form of character dis-
course and a virtuality that the reader might ponder in the course of reading. In
some works, we can observe a kind of continuity in the history of narratology. By
expanding on the formalist notion of fabula, Ryan has only added to its complex-
ity by exploring the virtualities shaped by the characters. We can also consider
Prince’s concept of disnarration as a mere expansion of the taxonomy of narrative
“figures” developed by classical narratology (especially in Genette’s works). In
both cases, we find once again the old distinction between story and discourse
reinterpreted in the light of the virtualities of narration.
On the other hand, many works dealing with virtualities, including Ryan’s,
have crossed an additional threshold by linking narrative structures to the func-
tion they play in discursive interaction. Despite a proliferation of epistemologies
(functionalism, constructivism, rhetoric, cognitive science, etc.) in their midst,
postclassical narratologies converge by stating that in order to understand the
dynamics of plots, we must take into account the “dialectical interplay between
narrative and consciousness” (Herman 2007, 257). Here, two attitudes are possi-
ble: the first, advocated by James Phelan, is to admit that, from a rhetorical per-
spective, we are dealing with some kind of “authorial audience,” one that realizes
the intention of an “implied author.” In this case, there is no reason to turn our
back on earlier perspectives because the virtualities that belong to the logic of
the fabula, and those expressed by the narrative discourse, must in some way
 The Garden of Forking Paths: Virtualities and Challenges    259

shape the inferences of this ideal audience. To these perspectives must be added
the extratextual knowledge that comes into play when a reader fills the gaps of
the narrative, or when s/he tries to predict its unfolding by resorting to scripts,
the logic of intentional actions, stereotypes, intertextuality, generic knowledge
and so on. It must also be admitted that all readers are not equal when it comes
to mobilizing extratextual knowledge in an attempt to explore the virtualities of
a text.
The second attitude would be to focus on empirical audiences. Here, the
clash with old paradigms appears to be more obvious. As stated by Michael
Toolan, “The difficult and interesting question is not whether readers experience
surprise, evasion, predictability, etc., but pinpointing the conditions, generaliz-
able beyond a case-by-case annotation, conducive to those conditions. Are the
conditions specifiable in formal terms at all?” (2004, 220). This question is still an
open one, I think. Some, like Umberto Eco (1984) or Wolfgang Iser (1978 [1976]),
believe that the “model reader” or the “implied reader” is mostly a construction
of the text and also that this abstraction is a good approximation of statistically
homogenous empirical readers.6 Others have stressed the initiative of the empiri-
cal reader and the unpredictability of his or her interpretations.
Herman has stated that “In 1983, Shlomith Rimmon-Kenan […] expressed
hope that deconstruction, at that time a comparatively new development on the
theoretical scene, might enrich narrative theory rather than render it obsolete”
(1999, 1). My opinion is that, thirty years later, while postclassical studies have
embraced new perspectives far beyond the mere description of formal attributes
of narratives, the trench between two major currents of narrative theories still
depends on how we deal with the “challenge” of deconstruction. Is contemporary
narratology entering in a period of consolidation or diversification? It all depends
on the way we look at the garden of forking paths, what might be called one’s
“narratological posture.” In the first case, we can have a global look at the garden
as a whole and consider every path as an interesting heuristic way to enlighten
the innumerable virtualities of narration. Of course, in each part of the garden
the kind of virtualities we are dealing with looks quite different. But it is possible
to admit that these different paths are complementary perspectives on a complex
issue. From this standpoint, we need to take a long walk to explore every corner
of the garden, drawing a map to see where we’ve come from and where we stand
and to get an idea of the regions that still need to be explored. This is the posture
adopted by Hilary Dannenberg when she expresses the hope for “a comprehen-
sive model of the counterfactual in fiction” (2004, 172). I think Herman shares

6 Eco (1984, 261–262) contains an example of such a statistical verification of “model reading.”
260   Raphaël Baroni

the same view when he says “Rethinking the problem of narrative sequences can
promote the development of a postclassical narratology that is not necessarily
poststructuralist, an enriched theory that draws on concepts and methods to
which the classical narratologists did not have access” (1997, 1048–1049).
Others, however, prefer to stress the conflicting nature of the different per-
spectives. Sternberg has formulated radical criticisms concerning formalism and
cognitivism.7 According to him, these approaches are unable to reflect the true
complexity of narrative structures and the multiplicity of their possible effects on
empirical readers. He also considers that the functionalist paradigm is not simply
a new layer in a continuous narratological history, but a completely different way
of approaching narrative phenomena: “I do not want to overdramatize matters,
but it comes to an either/or choice. Indeed, the trouble with some people who
have taken up my approach, to this day, is that they want to hold on somehow to
the bad good old formalism, to the old French structuralism” (Sternberg 2011, 43).
For Emma Kafalenos, when we distance ourselves from a formalist descrip-
tion, we inevitably have to deal with the complexity of empirical interpreta-
tions: “I emphasize the instability of a fabula as it grows and expands during
the process of reading, and I propose that contradictions […] permit readers of
narratives to participate in an endless play of signification. Such a view supports
a theoretical position that narrative is not a univocal mode of communication”
(1999, 60). We can also point to the fact that the “possible texts” theory developed
by French critics such as Pierre Bayard or Marc Escola clearly exceeds the scope
of both formal and functionalist narratologies, since it is clearly oriented toward
the infinite production of commentaries and alternative stories.
When dealing with non-literary texts, things might look quite different. It is
clear that in digital media, especially in videogames where the player has the
choice of several alternatives for continuing the story, the status of what belongs
to the fabula, and of what is counterfactual, changes completely. Moreover, trans-
fictional worlds (cf. Saint-Gelais 2011) such as the world of Starwars, created by
George Lucas and now proliferating in a wide variety of products including car-
toons, comics, novels, videogames, fanfictions, etc., open up new perspectives
by integrating into a single storyworld a potentially infinite range of “possible
narratives.” Whether the form and functioning of these storyworlds constitute
something which is radically new or whether they perpetuate traditional forms

7 For a critique of formalism, see Sternberg (2011, 43), and for a critique of cognitivism, see
Sternberg (2003).
 The Garden of Forking Paths: Virtualities and Challenges    261

of narrative in a new guise is still an open question.8 Here are clearly a few new
challenges for contemporary narratology.
In conclusion, I think that the richness of narrative theory can be attributed
to the coexistence of contradictory pathways, thus providing the discipline with
the right balance between forays into previously unexplored landscapes and the
possibility of capitalizing on a long tradition in narrative studies and theory as
well as with the continual necessity of engaging in discussion concerning the
modalities of exploration, the validity of epistemological perspectives and meth-
odologies. In any case, Thomas Pavel captures one of the common principles that
all contemporary narratologists would probably agree with, whatever their dif-

To fully account for our relationship with fiction, it is not enough to identify what is, but it
is also particularly important to consider the inferences caused by fiction. These inferences,
like those of everyday life, unfold in a space of values, norms, and possible actions. (2010,
312; translation mine)

Works Cited
Baroni, Raphaël. 2002. “Le rôle des scripts dans le récit.” Poétique 129: 93–114.
Baroni, Raphaël. 2005. “Compétences des lecteurs et schèmes séquentiels.” Littérature 137:
Baroni, Raphaël. 2007. La Tension narrative. Suspense, curiosité et surprise. Paris: Seuil.
Baroni, Raphaël. 2016. “Virtualities of Plot and the dynamics of Rereading.” In Narrative
Sequence in Contemporary Narratology, edited by Raphaël Baroni and Françoise Revaz,
87–103. Columbus: The Ohio State University Press.
Bayard, Pierre. 2000. Comment améliorer les œuvres ratées? Paris: Minuit.
Bayard, Pierre. 2008 (2008). Sherlock Holmes Was Wrong. Reopening the Case of the Hound of
the Baskerville. Translated by Charlotte Mandel. New York: Bloomsbury.
Borges, Jorge. 1999 “The Garden of Forking Paths.” In J. B. Collected Fictions, translated by
Andrew Hurley, 119–128. London: Penguin.
Brooks, Peter. 1992. Reading for the Plot. Design and Intention in Narrative. Cambridge, MA and
London: Harvard University Press.
Dannenberg, Hilary P. 2004. “Ontological Plotting: Narrative as a Multiplicity of Temporal
Dimensions.” In Pier, ed., 159–189.
Dannenberg, Hilary P. 2008. Coincidence and Counterfactuality: Plotting Time and Space in
Narrative Fiction. Lincoln and London: University of Nebraska Press.
Dufays, Jean-Louis. 2010. Stéréotype et lecture. Bruxelles, etc.: Peter Lang.
Eco, Umberto. 1984. The Role of the Reader. Bloomington: Indiana University Press.

8 For Ryan (2006a), such fictions are a continuation of the traditional forms.
262   Raphaël Baroni

Escola, Marc, ed. 2012. Théorie des textes possibles. Amsterdam and New York: Rodopi.
Gervais, Bertrand. 1990. Récits et actions. Longueuil: Le Préambule.
Herman, David. 1997. “Scripts, Sequences, and Stories: Elements of a Postclassical
Narratology.” PMLA 112 (5): 1046–1059.
Herman, David. 1999. “Introduction.” In Herman, ed., 1–30.
Herman, David. 2002. Story Logic: Problems and Possibilities of Narrative. Lincoln and London:
University of Nebraska Press.
Herman, David. 2007. “Cognition, emotion, and consciousness.” In D. H. The Cambridge
Companion to Narrative, 245–259. Cambridge: Cambridge University Press.
Herman, David, ed. 1999. Narratologies: New Perspectives on Narrative Analysis. Columbus:
The Ohio State University Press.
Herman, David, Manfred Jahn and Marie-Laure Ryan, eds. 2005. Routledge Encyclopedia of
Narrative Theory. London and New York: Routledge.
Iser, Wolfgang. 1978 [1976]. Act of Reading: A Theory of Aesthetic Reception. Baltimore and
London: The Johns Hopkins University Press.
Kafalenos, Emma. 1999. “Not (Yet) Knowing: Epistemological Effects of Deferred and
Suppressed Information in Narrative.” In Herman, ed., 33–65.
Kafalenos, Emma. 2001. “Editor’s Column.” Narrative 9: 113–114.
Kafalenos, Emma. 2006. Narrative Causalities. Columbus: The Ohio State University Press.
Lavocat, Françoise. 2010. “Avant-propos.” In Lavocat, ed., 5–11.
Lavocat, Françoise, ed. 2010. La Théorie littéraire des mondes possibles. Paris: CNRS Éditions.
Moulthrop, Stuart. 1991. “Reading From the Map: Metonymy and Metaphor in the Fiction
of ‘Forking Paths’.” In Hypermedia and Literary Studies, edited by Paul Delany et al.,
119–132. Cambridge, MA and London: The MIT Press.
Pavel, Thomas. 1975. “Possible Worlds in Literary Semantics.” The Journal of Aesthetics and Art
Criticism 34 (6): 165–176.
Pavel, Thomas. 2010. “Univers de fiction: un parcours personnel.” In Lavocat, ed., 307–313.
Phelan, James. 1989. Reading People, Reading Plots: Character, Progression, and the
Interpretation of Narrative. Chicago: The University of Chicago Press.
Pier, John. 2004. “Narrative Configurations.” In Pier, ed., 239–265.
Pier, John, ed. 2004. The Dynamics of Narrative Form: Studies in Anglo-American Narratology.
Berlin and New York: De Gruyter.
Prince, Gerald. 1988. “The Disnarrated.” Style 22: 1–8.
Richardson, Brian. 2015. Unnatural Narrative: Theory, History, and Practice. Columbus: The
Ohio State University Press.
Ryan, Marie-Laure. 1991. Possible Worlds, Artificial Intelligence, and Narrative Theory.
Bloomington: Indiana University Press.
Ryan, Marie-Laure. 2005a. “Possible-Worlds Theory.” In Herman et al. eds., 446–450.
Ryan, Marie-Laure. 2005b. “Virtuality.” In Herman et al. eds., 627–629.
Ryan, Marie-Laure. 2006a. “From Parallel Universes to Possible Worlds: Ontological Pluralism
in Physics, Narratology, and Narrative.” Poetics Today 27 (4): 633–674.
Ryan, Marie-Laure. 2006b. Avatars of Story. Minneapolis: University of Minnesota Press.
Saint-Gelais, Richard. 2011. Fictions Transfuges. Paris: Seuil.
Sternberg, Meir. 1993. Expositional Modes and Temporal Ordering in Fiction. Bloomington:
Indiana University Press.
Sternberg, Meir. 2003. “Universals of Narrative and their Cognitivist Fortunes (II).” Poetics
Today 24 (3): 517–638.
 The Garden of Forking Paths: Virtualities and Challenges    263

Sternberg, Meir. 2011. “Reconceptualizing Narratology. Arguments for a Functionalist and

Constructivist Approach to Narrative.” Enthymema 4: 35–50.
php/enthymema/article/view/1186 (Accessed 25 November 2015)
Todorov, Tzvetan. 1971. Poétique de la Prose. Paris: Seuil.
Toolan, Michael. 2004. “Graded Expectations: On the Textual and Structural Shaping of
Readers’ Narrative Experience.” In Pier, ed., 215–235.
Vaina, Lucia. 1977. “Les mondes possibles du texte.” Versus 17: 3–13.
Jan Alber (Aachen)
The Representation of Character
Interiority in Film: Cinematic Versions of
Psychonarration, Free Indirect Discourse
and Direct Thought
For decades, film theorists and narratologists have claimed that the medium
film is deficient with regard to the representation of character interiority. The
argument is usually that movies cannot depict fantasies, thoughts and feelings
as adequately or as convincingly as novels and short stories can. George Blue-
stone, for example, argues that “the rendition of mental states—memory, dream,
imagination—cannot be as adequately represented by film as by language” (1973
[1957], 47). Geoffrey Wagner even postulates that in film, “we cannot see what we
cannot see; in fiction we can” (1975, 183). Jakob Lothe also implies a hierarchy
between prose and film when he points out that “a film cannot convey a char-
acter’s thoughts, feelings, and so forth in the way fictional literature can” (2000,
By contrast, the more recent analyses by film scholars such as Leah Anderst,
Edward Branigan, Matthias Brütsch, Gilles Deleuze, Jens Eder, Markus Kuhn,
Maike Sarah Reinerth, Jan-Noël Thon and George Wilson, among others, have
shown that various ways exist in which movies can represent the inner lives of
their characters. My paper seeks to contribute to this new direction of research
by inductively developing a list of the numerous different ways in which movies
depict the mental states of storyworld inhabitants. In addition, I will show that
these cinematic types of consciousness representation bear significant structural
resemblances to novelistic techniques such as psychonarration, free indirect dis-
course and interior monologue, and that these similarities have hitherto been
overlooked. Finally, I will also comment on the ideological underpinnings of
filmic representations of character interiority.
This article builds on Alan Palmer’s argument that “the constructions of the
minds of fictional characters by narrators and readers are central to our under-
standing of how novels work because, in essence, narrative is the description of
fictional mental functioning” (2004, 12). From my perspective, Palmer’s state-
ment applies to films as well: we as recipients primarily understand cinematic
narratives by trying to grasp the intentions and motivations of the represented
minds. In addition, his distinction between an internalist and an externalist per-

DOI 10.1515/9783110555158-014
266   Jan Alber

spective on the mind plays a crucial role in my analyses. Palmer describes this
distinction as follows:

1. An internalist perspective […] stresses those aspects that are inner, introspective,
private, solitary, individual, psychological, mysterious, and detached.
2. An externalist perspective […] stresses those aspects that are outer, active, public,
social, behavioral, evident, embodied, and engaged. (2010, 39)

In this paper, I first assume an externalist perspective on the minds represented in

film. More specifically, I discuss external simulations of internal states (through
facial expressions, bodily positions, voice qualities, the metaphorical associa-
tion of a character with a different entity, and the use of music) which are struc-
turally similar to instances of psychonarration in prose narratives. In a second
step, I deal with dual-perspective shots which fuse Palmer’s externalist and his
internalist perspective. As I will show, in cases of quasi-perceptual overlays, the
camera does not only confront us with the subjective vision of a character, but
it also merges the figure’s point of view with an ‘objective’ one, thus creating
dual-perspective shots. This technique is reminiscent of passages of free indirect
discourse in novels and short stories. In a third step, I follow Palmer’s internalist
perspective on minds represented in film, and refer to more immediate cinematic
ways of rendering inner lives (through subtitles, captions, enacted mindscreens
and interior monologues at the auditory level) which are structurally similar to
stretches of direct thought in prose texts.1

1 The External Simulation of Internal States

Even though many theoreticians claim that films are deficient with regard to the
representation of character interiority, one can find many obvious examples in
which a cinematic narrative suggests or represents the internal state of a story-
world inhabitant. To begin with, films frequently allude to character interiority

1 I have excluded POV shots in which the camera assumes the spatial position of a character to
show the entities he or she sees because such shots typically only concern the figure’s vision and
not so much his or her thoughts or feelings. An exception might be the POV shots in Alfred Hitch-
cock’s Vertigo (1958) which simulate the acrophobia of John “Scottie” Ferguson (James Stewart)
through vertiginous shifts in perspective: in this film, dolly-zoom shots of staircases convey a
sense of the character’s dizziness and fear. Thon also refers to POV shots in terms of “the least
subjective of the pictorial strategies of subjective representation” (2014, 73). For detailed analy-
ses of POV shots, see Branigan (1984) and Choi (2005).
 The Representation of Character Interiority in Film   267

through facial expressions, bodily positions or the tone of his or her voice. In
The Shawshank Redemption (1994), for example, the middle-class banker Andy
Dufresne (Tim Robbins) is convicted of murdering his wife and her lover (even
though he did not kill them) and descends into the hell of the Shawshank State
Prison. When he walks into the prison at the beginning of the movie (see figure 1),
his facial expression, his insecure bodily position, and the context within which
they are used suggest that he is terrified by the screaming inmates who shake the
fence of the arrival area and also because he does not know what to expect inside
the prison (see also Alber 2007, 135).

Figure 1: Dufresne’s face and bodily position convey the internal states of fear and insecurity

Toward the end of the film, Dufresne does not only manage to escape from the
prison. He also exposes the injustices of Warden Norton’s (Bob Gunton) brutal
regime, and even becomes rich by taking money from the warden’s corrupt scams.
At this point in the film, his facial expression and his upright bodily position (see
figure 2) convey a sense of triumph and self-satisfaction.
268   Jan Alber

Figure 2: Dufresne’s face and bodily position convey the internal states of triumph and self

The film The Bourne Identity (2002) also offers us external clues to the inner life of
Jason Bourne (Matt Damon), a former CIA special agent who suffers from amnesia
and does not remember who he is. During a car ride to Paris, for instance, Marie
Kreutz (Franka Potente) repeatedly asks him what kind of music he likes until he
yells at her, “I don’t know!” The aggressive tone of his voice conveys that he is in
a state of despair because of his identity crisis: he does not know who he is and
he does not even remember his personal preferences or tastes, which is of course
extremely disconcerting.
Movies use not only facial expressions, bodily positions, or the tone of a
character’s voice to allude to the internal states of storyworld inhabitants; they
also often convey thoughts or feelings by (metaphorically) associating figures
with different entities (see also Alber 2011, 219–220). Toward the end of David
Lynch’s film The Elephant Man (1980), for instance, the deformed Englishman
John Merrick (John Hurt) is chased by a bullying crowd until he collapses near a
urinal. The juxtaposition of John Merrick with a urinal alludes to this character’s
emotional state: since the central protagonist has a deformed face and looks dif-
ferent from everyone else, he feels like a despicable ‘abject’ society wants to rid
itself of. The sensitive John Merrick feels like an outcast who is excluded from
society. David Lynch’s Lost Highway (1997) uses a similar strategy to convey the
feelings of one of its characters. In this film, the car mechanic Pete Dayton (Balt-
hazar Getty) is infatuated with his lover Alice Wakefield (Patricia Arquette), his
obsession bordering on self-destruction. At one point, she tells him that she won’t
 The Representation of Character Interiority in Film   269

be able to see him, and in the scene which follows, Pete is full of despair. In this
sequence, the film cuts from a close-up of Pete’s face to a shot of moths inside a
ceiling light, where they die in their attempt to fly into a light bulb. Furthermore,
the film juxtaposes Alice’s face with a close-up of a black-widow spider. These
two associations convey Pete’s impression that he is gradually destroying himself
through his relationship with Alice. On the one hand, he perceives himself in
terms of a moth in his desperate attempts to reach, have, or possess Alice. On the
other hand, he perceives Alice in terms of a black-widow spider, an insect which,
according to popular belief, kills its partner after mating.
Moreover, films may convey internal states through the use of particular types
of music or sound effects. Nobody’s Fault, the first part of Christine Edzard’s 1987
film adaptation of Dickens’s Little Dorrit, for example, uses romantic music to
highlight the emotional state of the character of Arthur Clennam (Derek Jacobi).
When Clennam watches Amy Dorrit (Sarah Pickering) in a restaurant, the film
deploys Verdi music which drowns all the other sounds (see also March 1993,
251). The slow and romantic string tunes illustrate that Clennam loves the girl
and, since all the other sounds of the represented world are drowned by the string
music, that he considers Amy Dorrit to be much more important or valuable than
everyone (and everything) else. A similar strategy is used in Lost Highway in the
scene in which the saxophone player Fred Madison (Bill Pullman) has sexual
intercourse with his wife Renée (Patricia Arquette). In this sequence, the film
deploys numerous sound effects to convey Fred’s feelings. More specifically, the
sex scene between the two characters is accompanied by lugubrious string sounds
which suggest that there is something fundamentally wrong between the two
figures. We can then hear a gong which is followed by a distant choir: it is almost
as though angels were singing somewhere in the background. At this point, Fred
feels that the lost unity between him and his wife is restored—at least for a brief
moment. However, Fred experiences a premature ejaculation and is quite obvi-
ously unable to satisfy Renée. When her hand then touches his back, we are con-
fronted with a disconcerting sound effect which is reminiscent of screeching tires
and alludes to Fred’s predominant impression that there are serious problems
in their relationship. While Nobody’s Fault uses romantic string tunes to convey
Arthur Clennam’s love, Lost Highway employs disconcerting music and sound
effects to give us a sense of Fred’s uneasiness.
In the examples discussed in this section, we as recipients do not get imme-
diate access to the characters’ thoughts or feelings. In some cases, external
signs (such as facial expressions, bodily positions, and voice qualities), which
are “outer, active, public, social, behavioral, evident, embodied, and engaged”
(Palmer 2004, 39), in combination with the context in which they occur, enable
us to determine the mental states of the characters. In other cases, the cinematic
270   Jan Alber

discourse provides information about the characters’ private feelings by meta-

phorically associating them with specific entities or through the use of particular
types of music or sound effects. This kind of representation also operates on the
basis of external simulations, i. e., without directly looking into the minds of the

2 Dual-Perspective Shots: Quasi-Perceptual

In this section, I look at dual-perspective shots which, as I will show, fuse the
point of view of one of the characters with an objective one.3 Following Jan-Noël
Thon (2014, 75), I use the term “quasi-perceptual overlay” to refer to shots in
which a film “simulates (quasi-) perceptual aspects of character’s consciousness
without […] assuming his or her spatial position.” In other words, we see a char-
acter from an external (third-person) perspective while the images we see simul-
taneously replicate this character’s world view.4
Christine Edzard’s 1987 film version of Dickens’s novel Little Dorrit (1855–57),
for instance, consists of two parts (Nobody’s Fault and Little Dorrit’s Story) which
present us with the quasi-perceptual overlays of two different characters. More
specifically, Nobody’s Fault focuses on the world view of Arthur Clennam, while
Little Dorrit’s Story deals with Amy Dorrit’s world view. Most of the individual
shots of these two films involve two perspectives because we see the charac-
ters from an external point of view (i. e., as a third-person observer would see
them) while the worlds through which they move are tinged with their subjective

2 Thon would refer to these examples, which concern the external simulation of internal states,
in terms of “intersubjective strategies of representation” that contribute to the “representation
of subjectivity” (2014, 69–71).
3 In this context, Edward Branigan points out that “objectivity and subjectivity interact in a nar-
rative film by being alternated, overlapped, or otherwise mixed, producing complex descriptions
of space, time, and causality” (1992, 160). Similarly, Celestino Deleyto argues that “subjectivity is
often expressed in a film without the complete disappearance of the external focalizer as a dis-
tinct agent from the character whose vision or mind we are made to share” (1996 [1991], 224). In-
deed, in this section, I try to show how objective and subjective perspectives can be intertwined
or combined in dual-perspective shots.
4 Both Leo Spitzer (1961 [1922]) and Per Krogh Hansen (2009) refer to quasi-perceptual overlays
as being “pseudo-objective.” Furthermore, as Thon (2014, 75) explains, they correspond to Ed-
ward Branigan’s “projection[s]” (1984, 90), George Wilson’s “subjectively inflected impersonal
shots” (2006, 81) and Jens Eder’s “subjectivized external views” (2008, 613).
 The Representation of Character Interiority in Film   271

impressions. That is to say, even though we see the characters from the outside,
the shots simultaneously imitate their respective perceptions of the world.
James Monaco speaks of a “‘third-person’ point of view” in the numerous
instances in film where we look at the characters from an external perspective
and contrasts it with POV shots that represent the perspective of one of the char-
acters (2000 [1977], 211). The examples I discuss fuse Monaco’s third-person point
of view with the character’s subjective world view. For instance, in Little Dorrit’s
Story, the room at the Marshalsea debtors’ prison is bigger and brighter than the
room we see in Nobody’s Fault. According to Joss March,

[…] the walls of the set have been bodily moved out by several feet; the set has been
repainted, redressed in slightly brighter colors; potted plants blossom […]; [William] Dorrit’s
bare chair grows a cover, and his dressing gown sprouts tendrils of embroidery. (1993, 255)

These two markedly perspectives on the prison and, by extension, the world
in general, recreate Arthur Clennam’s and Amy Dorrit’s subjective perceptions.
While Arthur Clennam has a pessimistic world view and feels oppressed in prison,
Amy Dorrit has become accustomed to the Marshalsea and has a more optimis-
tic world view. What Edzard’s film adaptation does is to blend the respective
world views of the characters with a third-person perspective: in both Nobody’s
Fault and Little Dorrit’s Story, we see the characters from the outside, i. e., as a
third-person observer would see them, but they move through worlds which look
like the world as they experience it.
Another example of a quasi-perceptual overlay can be found in Danny
Boyle’s film The Beach (2000), which is about backpackers who set up a beach
community on an island somewhere in Thailand. Toward the end of the film, the
American Richard (Leonardo DiCaprio) decides to live in the island’s forest in
order to defend the community against potential intruders. At this point, Richard,
who is rather fond of computer games, feels like being part of such a game, and
the film shows us Richard from an external perspective while the world through
which he moves imitates his subjective impression of the world (see figure 3). To
put this point somewhat differently, we see him as a third-person observer would
see him, but he moves through the world as he experiences it (namely as a com-
puter game) (see also Kuhn 153).
272   Jan Alber

Figure 3: We see Richard from a third-person perspective as he moves through a world which he
imagines to be a computer game

A particular type of quasi-perceptual overlay can be found in films in which a

character suffers from delusions or other psychophysical disturbances and the
filmic discourse then presents us with this character’s deranged world view as
though it were real. In other words, we as viewers are led into believing that we
are presented with an objective perspective on the film’s world, and it takes us
quite a while to realize that we are actually sharing the character’s deranged
vision or hallucination. In David Fincher’s Fight Club (1999), for example, the
nameless protagonist and voice-over narrator (Edward Norton) meets Tyler
Durden (Brad Pitt) on a plane, and the two set up a fight club for men as well as
an anti-materialist and anti-corporate organization called “Project Mayhem.” It
takes us as viewers until the end of the film to find out that Tyler Durden does
not actually exist, and that the nameless protagonist is merely imagining him as
some kind of alter ego. Most parts of the film involve two perspectives because we
see the unnamed protagonist from the outside as he interacts with Tyler Durden
as though he were an actual person. However, the latter turns out to be nothing
but a figment of the former’s imagination.
Similarly, in A Beautiful Mind (2001), John Nash (Russell Crowe), a student
of mathematics at Princeton University, begins to work for the US Department of
Defense. He is supposed to discover secret patterns in magazines and newspaper
articles to prevent a Soviet attack. However, toward the end of the film, we learn
that the people from the Defense Department do not actually exist and that Nash
is merely imagining them as a result of his schizophrenic episodes. Many scenes
in A Beautiful Mind involve two distinct points of view because we see Nash from
 The Representation of Character Interiority in Film   273

an external perspective, even though the world through which he moves is actu-
ally a hallucinated one.5

3 More Immediate Cinematic Ways of Rendering

Inner Lives
Finally, there are many more direct or more immediate cinematic ways of simu-
lating mental states. For instance, movies may use captions or subtitles to show
us what a character thinks or feels. In the famous balcony scene from Woody
Allen’s Annie Hall (1977), for example, we are presented with subtitles that render
the actual thoughts of Alvy Singer (Woody Allen) and Annie Hall (Diane Keaton)
as they are having a conversation (see figure 4). This scene is hilarious because it
exposes the discrepancy between what people say and what they think—which is
presumably something we are all familiar with. For example, at one point during
this conversation, Singer says to Hall, who tries to become a professional pho-
tographer, that “photography’s interesting ’cause you know it’s a new art form,
and a set of aesthetic criteria has not emerged yet” while the subtitles inform us
that he is actually wondering what she looks like naked:

5 Markus Kuhn discusses such forms of dual-perspective as types of “mental metalepsis” (2011,
156) while Per Krogh Hansen (2009) refers to them as cinematic examples of unreliable narra-
tion. From my perspective, both of these classifications are ill-advised. To begin with, there is
no actual crossing of narrative levels to be observed: the ‘jumping’ of fantasy figures into the
primary world of the film only takes place in the (deranged) mind of characters like Nash. Strictly
speaking, we are not confronted with metaleptic jumps. Furthermore, Nash is a focalizer who
misperceives the world he inhabits; he is not the primary narrator of the film. According to Volker
Ferenz, focalizers like Nash cannot be unreliable: they “cannot be held accountable for distort-
ing the fictional world simply because they do not narrate it” (2005, 140). Nash does not mis-
represent the world of A Beautiful Mind because he does not even try to narrate or represent it
in the first place; he only inhabits it. Similarly, it would be odd to call the third-person narrator
of Virginia Woolf’s novel Mrs Dalloway (1925) unreliable because she represents the deranged
world view of Septimus Warren Smith: he is a reflector-character who suffers from schizophrenia
following World War I, and his consciousness dominates large stretches of the novel. Woolf’s
narrator represents his deranged world view, but this hardly makes her unreliable.
274   Jan Alber

Figure 4: Subtitles give away Alvy Singer’s actual thoughts about Annie Hall

Similarly, A Study in Pink (2010), the first episode of the new television series
Sherlock, presents its viewers with captions which inform them about the con-
sulting detective’s deductions concerning a murder victim. At first, Sherlock
Holmes (Benedict Cumberbatch) assumes that the murdered woman scratched
the German word “Rache” (revenge) into the wooden floor (see figure 5). As in
Annie Hall, we as viewers can read a character’s thoughts:

Figure 5: Sherlock Holmes’s thoughts about the murder victim

 The Representation of Character Interiority in Film   275

Later on, however, he realizes that she actually tried to write “Rachel,” the
name of her daughter, but died before finishing the scrawl. As he continues to
investigate the victim, Sherlock Holmes also finds out that the victim repeatedly
removed her wedding ring, and he concludes that she must have been unhappily
married for over ten years (see figure 6).

Figure 6: Further deductions by Sherlock Holmes

In A Study in Pink, the terse, dictionary-like captions provide access to Sherlock

Holmes’s thought processes: the style conforms to the detective’s logical train of
The enacted mindscreen constitutes yet another cinematic way of simulat-
ing internal processes. The term was first used by Bruce Kawin, who argues that
mindscreens “belong to, or manifest the workings of specific minds” (1978, 12).
Moreover, the originator of the mindscreen exists at the intradiegetic level, and
the enactment of his or her thoughts constitutes a hypodiegetic level: the enacted
mindscreen is a type of interior monologue in which a character remembers
something or has a daydream, and we are then presented with a visual slice of
this character’s cogitating mind (Fleishman 1992, 27). In other words, we witness
the thoughts of this reminiscing character in terms of “mental images,” i. e., as
276   Jan Alber

“a dramatic rendering of narration [i. e., the character’s thoughts, J. A.] that has
itself been dramatized” (78, 15).6

Figure 7: The Deadly Viper Assassination Squad in a mindscreen/memory sequence

In Quentin Tarantino’s Kill Bill I (2003) and Kill Bill II (2004), for example, Beatrix
Kiddo (Uma Thurman) engages in a number of revenge killings. Before each
killing, the movie presents us with a mindscreen/memory sequence that informs
us about what the particular person she is about to slaughter had done to her. The
brutality and violence with which the other characters treated Beatrix Kiddo give
us a sense of the fury or anger she must feel. The following photograph (figure
7) is part of such a memory sequence. The shot presents the four characters Elle
Driver (Daryl Hannah), Vernita Green (Vivica A. Fox), Budd (Michael Madsen),

6 With regard to such sequences, Markus Kuhn speaks of cases of “mental metadiegesis” (or hy-
podiegesis): they constitute a character’s dreams, memories or hallucinations and do not repre-
sent anything that happens in the film’s primary storyworld (2011, 152). Thon, on the other hand,
speaks of “full-fledged ‘representation[s] of internal worlds’” (2014, 67) while Branigan refers
to such shots in terms of “mental processes” (1984, 90) and Wilson as “subjectively saturated
shots” (2006, 81). For further ways of simulating memory in film, see Reinerth (2009).
 The Representation of Character Interiority in Film   277

O-Ren Ishii (Lucy Liu), who are all former members of the Deadly Viper Assassi-
nation Squad and had all helped Beatrix Kiddo’s former lover Bill (David Carra-
dine), an elite assassin, in the attempt to kill her.
Films can also use interior monologues that convey a character’s thoughts,
feelings or motivations at the auditory level. In such cases, we can hear what a
character is thinking or feeling. Alfred Hitchcock’s Psycho (1960), for example,
is about Norman Bates (Anthony Perkins), who had murdered his mother and
her new partner. Bates now dresses up as his mother when he kills women (such
as Marion Crane [Janet Leigh]) he is sexually attracted by. When Bates is finally
arrested, the camera moves closer and closer to the prisoner until we are con-
fronted with an extreme close-up of his grinning face. As the camera moves
toward him, his interior monologue sets in. Shockingly, we do not hear Bates’s
voice but rather the voice of his mother:

It’s sad when a mother has to speak the words that condemn her own son. I couldn’t allow
them to believe I would commit murder. They’ll put him away now as I should have years
ago. He was always bad and in the end he intended to tell them I killed those girls and that
man, as if I could do anything except just sit and stare—like one of his stuffed birds. Oh,
they know I can’t even move a finger and I won’t. I’ll just sit here and be quiet just in case
they do… suspect me. They’re probably watching me. Well, let them. Let them see what kind
of a person I am. I’m not even going to swat that fly. I hope they are watching… they’ll see.
They’ll see and they’ll know, and they’ll say, ‘Why, she wouldn’t even harm a fly.’

This disconcerting technique illustrates what happens inside Bates’s mind: his
personality is entirely dominated by his dead mother who inhabits the body
of her ‘lost’ son. Furthermore, we learn that Bates assumes the identity of his
mother, and as his mother, he exculpates her from all responsibilities and instead
puts the blame on himself.
A similar technique can be found in the eighth season of the television series
Two and a Half Men (2009) when Chelsea (Jennifer Bini Taylor), Charlie Harper’s
(Charlie Sheen) girlfriend, tries to comfort her friend Gail (Tricia Helfer) whose
boyfriend Brian had left her. The chauvinist Charlie, on the other hand, does not
follow the conversation in the restaurant at all and instead muses about things
that interest him. His thoughts are represented in terms of stretches of direct
thought at the auditory level:

Hard to believe somebody got tired of doing her [Gail, J. A.]. [A waiter passes carrying French
fries, J. A.] Oh, French fries look good. Why did I order the baked potato? […] Now is this the
one who’s the old college roommate? I should really listen more. […] Boy, I bet when they
were roommates, they did some experimenting. Why is it that girls can experiment with
other girls and not be gay, but if a guy so as much kisses another guy … whoop, bad memory!
Drink! Drink, drink, drink! […] [Chelsea tells Charlie that “Brian (Gail’s ex-boyfriend, J. A.)
278   Jan Alber

took Rufus (the dog, J. A.)”]. Who’s Rufus? For that matter, who’s Brian? Life of Brian—that
was pretty funny. Brian’s song not so much. James Caan was great in it, though. That’s ‘cause
he’s a great actor. Godfather, Rollerball … Yeah, I can so see these two kissing in a dorm.

The point of this interior monologue is to demonstrate that the egocentric and
sexist Charlie lives in his own sphere, which is cut off from the rest of the world. He
does not care about other people’s problems at all and instead concerns himself
with what matters to him, i. e., things such as Gail’s attractiveness, French fries,
sexual experimentation, movies, actors and so forth. It is also interesting to note
that he tries to drown the memory of his homoerotic episode in alcohol. Charlie
presumably does this because homoeroticism contradicts his idea of what it
means to be a ‘real’ man.7
As I have shown in this section, subtitles, captions, enacted mindscreen
sequences and interior monologues at the auditory level do allow us direct access
to the thoughts and feelings of the characters in film: in the first case, we can
read them (as in Annie Hall and A Study in Pink); in the second case, we can see
enactments of them (as in Kill Bill I and II); and in the third case, we can literally
hear them (as in Psycho and Two and a Half Men).

4 Conclusions: Cinematic and Novelistic Types of

Consciousness Representation
While Dorrit Cohn in Transparent Minds illustrates that prose texts frequently use
psychonarration, free indirect discourse (what she calls “narrated monologue”)
and interior monologues (what she calls “quoted monologue”) to represent the
minds of their characters, it is my goal to demonstrate that despite claims to the
contrary, films actually have numerous ways of simulating character interiority
at their disposal. The medium film is clearly not deficient with regard to the rep-
resentation of the thoughts and feelings of storyworld inhabitants. Let me con-

7 Some critics argue that interior monologues at the auditory level were first used by Alain
Resnais in Hiroshima mon amour (1959). However, I am familiar with an even earlier example.
The British prison film Yield to the Night (1956) uses interior monologues at the auditory level to
familiarize us as viewers with the thoughts and feelings of the inmate Mary Price Hilton (Diana
Dors). The technique is also used in more recent television series such as the British Peep Show
(2003) and the American House of Cards (2013), where the plot comes to a halt whenever the
central protagonist Frank Underwood (Kevin Spacey), a Democrat from South Carolina, speaks
his thoughts into the camera.
 The Representation of Character Interiority in Film   279

clude by addressing the question of how the novelistic types of consciousness

representation discussed by Cohn are related to the cinematic techniques that I
have presented in this paper.
To begin with, as regards prose texts, the term “psychonarration” denotes a
narrator’s report of a character’s mental states in the narrator’s own words (Flud-
ernik 1993, 297). Dorrit Cohn (1978, 49) uses the following example of psychonar-
ration from D. H. Lawrence’s Women in Love (1920): “All this Gudrun knew in her
subconsciousness, not in her mind.” Here the narrator informs us about the mind
of the character Gudrun, and he does so in his own way of expressing himself.
Similarly, the cinematic discourse (or hypothetical filmmaker)8 may inform us
about character interiority by using certain cinematic strategies (such as close-
ups of facial expressions; specific bodily positions; particular types of voice;
the [metaphorical] association of a figure with a specific entity; or certain types
of music or sound effects). In such cases, the filmic discourse does not provide
direct access to character interiority; rather, it describes or alludes to the thoughts
or feelings of a character through external features.
Second, in relation to novels and short stories, the term “free indirect dis-
course” refers to a third-person rendering of thoughts, which remains close to the
character’s own oral syntax and diction. In the words of Monika Fludernik, “free
indirect discourse preserves some of the expressive elements of direct discourse
as well as its syntactic independence, but shares with indirect discourse the tem-
poral and referential consonance with the quoting instance” (1993, 74). Gram-
matically speaking, free indirect discourse is reported speech or thought without
introductory verb: “he thought that he should go now” is reported thought, while
“he should go now” is free indirect thought. Examples of free indirect discourse
can be found in the following quotation from D. H. Lawrence’s novel The Rainbow,
which moves from Tom Brangwen’s impressions (in free indirect thought) to the
implied speech of the girl (in free indirect speech):

Afterwards he glowed with pleasure. By Jove, but that was something like! He stayed the
afternoon with the girl, and wanted to stay the night. She however, told him that this was
impossible: her own man would be back by dark, and she must be with him. He, Brangwen,
must not let on that there had been anything between them. (1988 [1915], 23; emphasis added)

8 The term “hypothetical filmmaker” denotes “the viewer’s speculations about the conscious or
unconscious motivations actuating the group of professionals that were responsible for the mak-
ing of the film in question” (Alber 2010, 167). From my perspective, it does not matter whether
we speak of cinematic discourse or of the hypothetical filmmaker so long as we acknowledge the
fact that we speculate about the intentions behind the making of the film.
280   Jan Alber

To my knowledge, Mikhail Bakhtin and Valentin Vološinov were the first theore-
ticians to describe free indirect discourse in terms of a “dual voice” (see McHale
2014 [2009], 815, and Schmid 2010 [2005], 137–139). Roy Pascal likewise argues
that free indirect discourse “embeds the character’s statement or thought in the
narrative flow, and even more importantly in the narrator’s interpretation, com-
municating also his way of seeing and feeling” (1977, 75). Free indirect discourse
is thus typically seen as “a dialogue, a contest, between figural and narratorial
idiom, a merging or juxtaposition of voices, of the narrator and the character
respectively” (Fludernik 1993, 323). Quasi-perceptual overlays (as in Edzard’s film
version of Little Dorrit, The Beach, Fight Club or A Beautiful Mind) may be seen
as cinematic versions of free indirect discourse because they are—in a specific
sense—also dual-voiced: in such cases, we see the characters from a third-person
perspective while the images we are confronted with remain close to these char-
acters’ subjective perception of the world.9 With reference to Pier Paolo Pasolini,
Gilles Deleuze also classifies such shots as “free indirect subjective.” He explains
that in such instances, “the camera does not simply give us the vision of the char-
acter and of his world; it imposes another vision in which the first is transformed
and reflected” (2009 [1983], 74). Leah Anderst describes such dual-perspective
shots in a similar way: “the viewer simultaneously sees what the character sees
and sees the character in the act of seeing” (2011, 363).
Finally, with regard to prose texts, the term “interior monologue” denotes
longer passages of uninterrupted direct thought—usually without any narrato-
rial mediation, as in the “Penelope” chapter from James Joyce’s Ulysses (1922).
Generally speaking, interior monologues try to represent the chaotic, irregular
and associative character of human thought processes (such as those by Molly
Bloom in Ulysses). The use of subtitles, captions, enacted mindscreens or interior
monologues on the auditory level in film function analogously: such cinematic
strategies also present us with the random thoughts of characters, and depend-
ing on which technique we are confronted with, we can either read them, see
them, or hear them.
One does obviously not have to use the terminology of literary studies to
describe film techniques that concern the representation of character interiority.
Nevertheless, the overlaps between novelistic and cinematic strategies of con-
sciousness representation are interesting and striking, and they have hitherto
been overlooked. From my perspective, these overlaps are as fascinating and

9 In the case of film, the equivalent of direct discourse would be a purely subjective POV-shot,
i. e., a shot from the perspective of the character in which we see the world from his or her spatial
perspective. In addition, we would have to see the world exactly as he or she sees it.
 The Representation of Character Interiority in Film   281

remarkable as the transparent brain of Homer Simpson, where we can see the
workings of his mind in the shape of a monkey that tries to get Homer to finally
listen to his wife (see figure 8).

Figure 8: Homer Simpson’s transparent brain

This figure is interesting not only because we can directly look into Homer Simp-
son’s brain. It is also worth noting that the monkey, which represents Homer’s
thought processes (or perhaps the lack thereof), urges the male Homer to listen to
the female Marge. One might argue that most of the films I have discussed in this
paper could use such a monkey as well: as far as their ideological underpinnings
are concerned, they clearly have a masculinist bias because they focus on the
inner lives of their male characters and pay significantly less attention to the inte-
riority of their female characters. With the exception of Amy Dorrit and Beatrix
Kiddo, all of my sample films and television series concentrate on the ‘standard’
heterosexual male, and I believe that this is a general tendency in mainstream
films and television series. In addition to the striking similarities between novel-
istic and cinematic types of consciousness representation, I would therefore also
like to highlight that when it comes to the representation of character interiority,
most films foreground a masculinist subject position.
282   Jan Alber

Works Cited
Primary sources
Annie Hall. Dir. Woody Allen. United Artists, 1977.
A Beautiful Mind. Dir. Ron Howard. Universal, 2001.
The Beach. Dir. Danny Boyle. Figment Films, 2000.
The Elephant Man. Dir. David Lynch. Brooksfilms, 1980.
Fight Club. Dir. David Fincher. Twentieth-Century-Fox, 1999.
Kill Bill I and II. Dir. Quentin Tarantino. Miramax, 2003–2004.
Lawrence, D. H. 1988 [1915]. The Rainbow. Edited by Harold Bloom. New York: Chelsea House
Little Dorrit. Part I: Nobody’s Fault. Dir. Christine Edzard. Sands Film, 1987.
Little Dorrit. Part II: Little Dorrit’s Story. Dir. Christine Edzard. Sands Film, 1987.
Lost Highway. Dir. David Lynch. Asymmetrical Productions/Ciby 2000 [1997].
Psycho. Dir. Alfred Hitchcock. Paramount, 1960.
The Shawshank Redemption. Dir. Frank Darabont. Castle Rock Entertainment, 1994.
Two and a Half Men: Season 8. Dir. James Widdoes. Chuck Lorre Productions, 2009.
Sherlock Holmes: A Study in Pink. Dir. Paul McGuigan. BBC, 2010.
Vertigo. Dir. Alfred Hitchcock. Paramount, 1958.

Secondary sources
Alber, Jan. 2007. Narrating the Prison: Role and Representation in Charles Dickens’ Novels,
Twentieth-Century Fiction, and Film. Youngstown, NY: Cambria Press.
Alber, Jan. 2010. “Hypothetical Intentionalism: Cinematic Narration Reconsidered.” In
Postclassical Narratology: Approaches and Analyses, edited by Jan Alber and Monika
Fludernik, 163–185. Columbus: The Ohio State University Press.
Alber, Jan. 2011. “Cinematic Carcerality: Prison Metaphors in Film.” The Journal of Popular
Culture 44 (2): 217–232.
Anderst, Leah. 2011. “Cinematic Free Indirect Style: Represented Memory in Hiroshima mon
amour.” Narrative 19 (3): 358–382.
Bluestone, George. 1973 [1957]. Novels into Film. Berkeley: University of California Press.
Bordwell, David, and Kristin Thompson. 2008. Film Art: An Introduction. New York: McGraw Hill.
Branigan, Edward. 1984. Point of View in the Cinema: A Theory of Narration and Subjectivity in
Classical Film. Berlin: Mouton.
Branigan, Edward. 1992. Narrative Comprehension and Film. London and New York: Routledge.
Brütsch, Matthias. 2011. Traumbühne Kino: Der Traum als filmtheoretische Metapher und
narratives Motiv. Marburg: Schüren.
Choi, Jinhee. 2005. “Leaving It Up to the Imagination: POV Shots and Imagining from the
Inside.” The Journal of Aesthetics and Art Criticism 63 (1): 17–25.
Cohn, Dorrit. 1978. Transparent Minds: Narrative Modes for Presenting Consciousness in
Fiction. Princeton, NJ: Princeton University Press.
Deleuze, Gilles. 2009 [1983]. Cinema 1: The Movement-Image. Translated by Hugh Tomlinson
and Barbara Haberjam. Minneapolis: University of Minnesota Press.
 The Representation of Character Interiority in Film   283

Deleyto, Celestino. 1996 [1991]. “Focalisation in Film Narrative.” In Narratology, edited by

Susana Onega and José Ángel García Landa, 217–233. London: Longman.
Eder, Jens. 2008. Die Figur im Film: Grundlagen der Figurenanalyse. Marburg: Schüren.
Ferenz, Volker. 2005. “Fight Clubs, American Psychos and Mementos: The Scope of Unreliable
Narration in Film.” New Review of Film and Television Studies 3 (2): 133–159.
Fleishman, Avrom. 1992. Narrated Films: Storytelling Situations in Cinema History. Baltimore,
MD: The Johns Hopkins University Press.
Fludernik, Monika. 1993. The Fictions of Language and the Languages of Fiction: The Linguistic
Representation of Speech and Consciousness. London and New York: Routledge.
Hansen, Per Krogh. 2009. “Unreliable Narration in Cinema: Facing the Cognitive Challenge
Arising from Literary Studies.” Amsterdam International Electronic Journal of Cultural
Narratology 5. (Accessed 1 September 2014)
Kawin, Bruce. 1978. Mindscreen: Bergman, Godard, and First-Person Film. Princeton, NJ:
Princeton University Press.
Kuhn, Markus. 2011. Filmnarratologie: Ein erzähltheoretisches Analysemodell. Berlin and New
York: De Gruyter.
Lothe, Jakob. 2000. Narrative in Fiction and Film: An Introduction. Oxford: Oxford University
McHale, Brian. 2014 [2009]. “Speech Representation.” In Handbook of Narratology, 2nd ed.,
edited by Peter Hühn, Jan Christoph Meister, John Pier and Wolf Schmid, vol. 2: 812–824.
Berlin and New York: De Gruyter. Also available at:
article/speech-representation (Revised 8 April 2014)
March, Joss. 1993. “Inimitable Double Vision: Dickens, Little Dorrit, Photography, Film.” Dickens
Studies Annual 22: 239–282.
Monaco, James. 2000 [1977]. How to Read a Film: The World of Movies, Media, and Multimedia.
Language, History, Theory. New York: Oxford University Press.
Palmer, Alan. 2004. Fictional Minds. Lincoln and London: University of Nebraska Press.
Palmer, Alan. 2010. Social Minds in the Novel. Columbus: The Ohio State University Press.
Pascal, Roy. 1977. The Dual Voice: Free Indirect Speech and its Functioning in the Nineteenth-
Century American Novel. Manchester: Manchester University Press.
Reinerth, Maike Sarah. 2009. “Spulen, Speichern, Überspielen: Zur Darstellung von Erinnerung
im Spielfilm.” In Probleme filmischen Erzählens, edited by Hannah Birr, Maike Sarah
Reinerth and Jan-Noël Thon, 33–58. Münster: Lit.
Schmid, Wolf 2010 [2005]. Narratology: An Introduction. Translated by Alexander Starritt. Berlin
and New York: De Gruyter.
Leo Spitzer. 1961 [1922]. “Pseudoobjektive Motivierung bei Charles Louis Philippe.” In L. S.
Stilstudien II: Stilsprachen, 166–207. Munich: Hueber.
Thon, Jan-Noël. 2014. “Subjectivity across Media: On Transmedial Strategies of Subjective
Representation in Contemporary Feature Films, Graphic Novels, and Computer Games.” In
Storyworlds across Media: Toward a Media-Conscious Narratology, edited by Marie-Laure
Ryan and Jan-Noël Thon, 67–102. Lincoln: University of Nebraska Press.
Wagner, Geoffrey. 1975. The Novel and the Cinema. Rutherford, NJ: Fairleigh Dickinson
University Press.
Wilson, George. 2006. “Transparency and Twist in Narrative Fiction Film.” Journal of Aesthetics
and Art Criticism 64 (1): 81–95.
Małgorzata Pawłowska (Cracow)
Intermedial Transposition:
From Verbal Story to Music.
Narrative in Musical Works Based on
Romeo and Juliet

1 Music, Narrative, Intermedial Transposition:

Opening Remarks
There is an ongoing lively discussion, both in general narratology and in musico-
logical writings to date, as to whether music can be narrative.1 Among the various
media in which the concept of narrative is usually referred to, music seems the
most controversial. Nevertheless, narrative theory is being applied increasingly
in music analysis (e. g., by Tarasti, Grabócz, Hatten, Monelle, Micznik, Almén and
others). Musical narratology is currently undergoing a phase of dynamic devel-
opment but at the same time is still in the process of defining itself, taking ele-
ments of theory from both classical and post-classical narratology and seeking
to “bridge the gap between them,” to borrow an expression from Nora Berning’s
contribution to this volume. Apart from adopting various tools from literary
theory, musical narratology is also developing its own concepts and methods of
In this paper, I focus on a case of “intermedial transposition” (cf. Wolf 2008a
[2005] and Rajewsky 2005), in which the story of Romeo and Juliet is transposed
from literature to music, or at least serves as a source of inspiration for compos-
ers. Claude Bremond wrote:

Story is independent of the techniques that bear it along. It may be transposed from one to
another medium without losing its essential properties. (1964, 4; quoted in Chatman 1978,

1 I present an account of this discussion, as well as of musical narratology in general, in my ar-
ticle “Musical Narratology: An Outline” (Pawłowska 2014). For the debate as to whether musical
works can be considered as narratives, see Ryan (2004, Part 4) and Wolf (2005, 2008b [2005]).
For the discussion in musicology, see: Abbate (1991); Agawu (1991); Almén (2003, 2008); Berger
(2000); Cone (1974, 1982, 1985); Grabócz (1999, 2008, 2009); Kerman (1992); Maus (1988, 1991,
1997, 2001, 2005); Micznik (2000, 2001); Newcomb (1984, 1988, 1994, 1997); Ratner (1980); Taras-
ti (1991, 1994, 1995, 2004, 2008).

DOI 10.1515/9783110555158-015
286   Małgorzata Pawłowska

Intermediality in its broadest sense applies “to any transgression of boundaries

between media” (Wolf 2008a [2005], 252). However, there are many approaches
and categories related to the term. As Irina O. Rajewsky writes, medial transposi-
tion is the subcategory of intermediality in which “the intermedial quality has to
do with the way in which a media product comes into being, i. e., with the trans-
formation of a given media product (a text, a film, etc.) or of its substratum into
another medium” (2005, 51).
William Shakespeare’s drama Romeo and Juliet has long served as a point of
departure for composers taking up the subject of Romeo and Juliet, and it thus
provides many examples of intermedial transposition. With this in mind, I will
consider the following questions in this chapter: What elements of Shakespeare’s
literary narrative can be transposed to music, and in what ways? What has been
changed in order to express the theme in music? In what ways does the choice
of musical genre affect the narrative? And what are the purely musical means of
suggesting narrative? Surely some of these questions go beyond a “case study” of
the Romeo and Juliet story, for they concern more than the content and structure
of the story, touching upon a special case of intermedial transposition, namely,
the transposition of certain formal devices from one medium to another—from
literature to music.
In Wolf’s typology, intermedial transposition belongs to “extracompositional
intermediality.” The author explains: “as is typical of extracompositional inter-
mediality in general, in all of these cases the intermedial quality is primarily
located in the space between the two works: in the process of gestation, but not
in the end product” (2015, 462; cf. Wolf 2008a [2005], 254). However, it seems the
musical works based on Romeo and Juliet legend are also characterised by some
degree of “intracompositional intermediality.” Adopting Steven Paul Scher’s
triadic distinction between “literature in music,” “music and literature” and
“music in literature” (cf. Wolf 2015), we can classify Tchaikovsky’s instrumen-
tal programmatic piece Romeo and Juliet as “literature in music” while Bellini’s
and Gounod’s operas, consisting of text and music, are examples of “literature
and music.” No doubt most of the musical compositions inspired by an extra-
musical subject touch or transgress in one way or another the boundaries of the
musical medium, as happens not only with programme music, opera, musical
and Singspiel, but also in ballet or film music. To make things more complex,
we are dealing here with plurimedial phenomena, leading us to trace narrative
devices specific to individual media as well as to transmedial devices.
From this perspective, it is worth considering the presence of transmedial
elements—i. e., non-media-specific elements—in both literary sources and music
compositions. As I attempted to explain in my article on musical narratology
(Pawłowska 2014), backed up with a number of authors working on the question
 Intermedial Transposition: From Verbal Story to Music    287

(Berger 2000, Micznik 2001, Almén 2008, etc.), it is possible to view narrative as
a concept established before the means are specified. If we accept a broad defi-
nition of narrative, we can say that narrative is possible in music as well as in
language and that it is not a secondary or derivative phenomenon taken from
literature, since narrative can manifest itself through the interaction of musical
elements. Byron Almén indicates that the definition of narrative is the source of

Because narrative was first conceptualized in relation to literature, we have largely failed
to recognize the distinction between narrative proper and narrative as manifested in liter-
ature. (2008, 12)

Among the existing definitions, he distinguishes between those based on

descendant models and those based on sibling models. Almén writes:

To use a genealogical metaphor, I prefer a sibling model rather than a descendant model
for articulating the relationship between musical and literary narrative. The descendant
model presupposes a conceptual priority for literary narrative, while the sibling model dis-
tinguishes between a set of foundational principles common to all narrative media and
principles unique to each medium. (2008, 18)

Although transposing a story from literature to music explicitly fits the descend-
ant model, I will also take a step away from the story proper and reflect on the
narrative affordances of music as compared with those of literature.

2 The Career of the Romeo and Juliet Theme in

Before examining specific narrative aspects of intermedial transpositions in the
corpus involved, it is useful to take a look at the scope and dimensions of Romeo
and Juliet theme in music.
Shakespeare’s drama Romeo and Juliet2 has inspired many composers (for a
list detailing the most important musical works inspired by the story of Romeo
and Juliet, see the annex). Many composers have felt deeply touched by Shake-

2 The story of Romeo and Juliet has existed in Europe in several epic versions dating back to
antiquity (Xenophon of Ephesus, Masuccio da Salerno, Luigi da Porto, Matteo Bandello, Arthur
Brooke). Shakespeare was the first to put the story into a dramatic genre (1596), and his play has
inspired countless works in theatre, literature, music, ballet, the fine arts, film, etc.
288   Małgorzata Pawłowska

speare’s art and considered this play especially prone to musicalisation. For
instance, Tchaikovsky observed that “this great work of the arch-genius is well
adapted to inspire a musician” (to N. F. von Meck, M. Tchaikovsky 1924 [1905],
304) and Berlioz wrote: “What a subject! How well made for music!” (quoted by
Rushton 1994, 10).
Musical works inspired by the Romeo and Juliet myth3 consist either of text
and music (as in opera and Singspiel) or simply of music (programme music such
as symphonic poems or ballets). Thus, works with text include a vocal element
(solo voices and/or chorus), while those without text are purely instrumental
(although they may include plurimedial elements such as dance in the case of
ballet). The dramatic symphony Roméo et Juliette by Berlioz is a unique case,
including sections written with the use of text and vocal parts alongside purely
instrumental segments.
The range of genres is vast, including operas (e. g., Bellini, Gounod, Dusapin),
a symphonic poem (Tchaikovsky), an orchestral suite (Kabalevsky), a symphonie
dramatique (Berlioz), a ballet (Prokofiev), a Singspiel (Benda), a musical (e. g.,
Bernstein) and even a chamber oratorio (Blacher). Moreover, the slow movement
of Beethoven’s String Quartet Op. 18 No. 1, Adagio affettuoso et appassionato, was
inspired by the tomb scene from Romeo and Juliet (cf. Micznik 2000, 51–54). Nor
must film music be forgotten (e. g., the music composed by Rota for Zeffirelli’s
film also functions separately from the film). However, the majority of works
inspired by the Romeo and Juliet myth are operas, representing the genre most
naturally suited to the transposition of drama into the sphere of music.
The first known musical interpretations of the story date from 1776. In the
course of the nearly two centuries that had passed since Shakespeare wrote his
play, the shape and character of the myth of Romeo and Juliet had varied sig-
nificantly. In its first musical interpretations, the composers and librettists used
altered versions and adaptations of Shakespeare’s play which were sentimen-
tal and melodramatic in character, deprived of the Shakespearian sense of the
comedic and of anything “trivial, exaggerated and redundant to the action,”
as the author of such an adaptation, Felix Weiße, is reported to have said (cf.
Komorowski 1990, 58). These adaptations, however, contain an extensive scene

3 The Romeo and Juliet story, because of its scope, meaning, content and structure, has all the
features of a modern myth, one which was present in Europe, in various versions, since before
Shakespeare. Since his drama, the myth has spread through the collective consciousness of Eu-
ropean culture. Alongside the myths of Faust, Orpheus and Eurydice and Tristan and Isolde, it
has become one of the most frequently used themes—a story which “has been told over and over
again with variations”—one of the basic qualities of a myth (Walker 2008 [2005], 239–330; cf.
Weinrich 2004 [1973]).
 Intermedial Transposition: From Verbal Story to Music    289

depicting Juliet’s burial (and funeral procession) which was added to Shake-
speare’s original play. In accordance with the tendency of the Classical era,
some of the earliest-known musical works based on the myth introduce a happy
ending4 (e. g., an opera by Schwanberger from 1776, a Singspiel by Benda from
1776 and an opera by Steibelt from 1793). In the first three mutually interrelated
Italian operas5 (Zingarelli 1796, Vaccai 1825, Bellini 1830), one can find, in addi-
tion to Shakespeare’s drama, pre-Shakespearean sources6 (Luigi da Porto, Ban-
dello) as well as later ones (e. g., the adaptation of Shakespeare’s drama in French
pseudo-classical style by Jean-François Ducis in 1772). In the early days of the
drama’s theatrical life, Juliet’s role was played, following the acting practices of
the time, by men (until the 1660s), whereas in the early stagings of the opera the
role of Romeo was given to women. In Zingarelli’s opera, Romeo’s role was ini-
tially written for a castrato, but by the 1820s it was being sung by prima donnas
whereas in the works of Vaccai and Bellini, the role was written for a female per-
former from the outset.
The nineteenth century brought a breakthrough in the reception of Shake-
speare, who had effectively reached cult status, as proven by the fascination
which Romantic-era composers had with Shakespeare and his works—particu-
larly Berlioz or d’Ivry, who travelled to Verona and England to better understand
the context of Romeo and Juliet (Dollinger 2007). It must be added that Romantic
composers often discovered Shakespeare through other sources. For example, the
production of Romeo and Juliet seen by Berlioz in 1827, when “sudden […] revela-
tion of Shakespeare overwhelmed” him (Berlioz 1966, 66), was based on Garrick’s
adaptation. It was only Tchaikovsky who was directly affected by Shakespeare’s
drama, resulting in his observation that in the existing operas by Bellini and
Gounod “Shakespeare is mutilated and distorted until he is hardly recognizable”
(Tchaikovsky 1924 [1905], 304). In various nineteenth-century musical interpre-
tations, which move closer to or farther from the original, there is no room for
the Shakespearean polarisation of romanticism and anti-romanticism, tragic and
comic elements; romantic and tragic elements dominate.

4 For instance, in the famous opera Orfeo ed Euridice by Christoph Willibald Gluck from 1762,
there is a happy ending; hence the conclusion of the myth is changed.
5 Felice Romani, who wrote the librettos to Vaccai’s and Bellini’s operas, based them on the
libretto written by Giuseppe Maria Foppa for Zingarelli. Bellini was Zingarelli’s student. One of
the scenes from Vaccai’s opera is often included in productions of Bellini’s opera.
6 The works of da Porto and Bandello may have been known in Italy; there are probably echoes
of them in Zingarelli’s opera, and thus in Vaccai’s and Bellini’s. Only Zandonai’s opera (1922) is
based entirely on these pre-Shakespearean sources.
290   Małgorzata Pawłowska

It was only twentieth-century musical interpretations that restored Shake-

speare’s polarisation of romanticism and anti-romanticism, on the one hand, and
tragedy and comedy, on the other, in new musical idioms (e. g., in Prokofiev and
Bernstein). Prokofiev, through his “masculine” reserved lyricism and “sharply
contoured” melodies, introduced the antidote to the early and late sentimentality
that the plot had acquired in the course of the nineteenth century .
Simultaneously, in the twentieth century we observe the gradual movement
of the myth towards the sphere of popular music, beginning with Bernstein’s
musical West Side Story (1957). This, along with several popular musicals, show
how conducive to modernisation this narrative is. Following the appearance of
the plot in the cinematic medium, the topic became popular in film music (pri-
marily films by Zeffirelli and Luhrmann). Many popular music artists also refer to
the myth, among them Radiohead, Dire Straits, the Supremes, Bruce Springsteen,
Taylor Swift, Tom Waits and Lou Reed.
Even a brief glance at the titles of compositions inspired by the myth indi-
cates a shift of accent relative to Shakespeare’s work. Often, Juliet is made the
principal subject of the narrative, which can be suggested by inverting the title
(Zingarelli, Vaccai, Torriani, Matuszczak).
It is significant that the latest known interpretation of the myth in (non-pop-
ular) music dates from 1988 (Dusapin’s opera). One might assume that contempo-
rary composers avoid the overly specified semantics and potential banality of the
subject, perhaps thinking that the topic has already been thoroughly exploited.
Contemporary opera—to mention one genre only—often explores the complexi-
ties of a particular character’s psyche; hence its favourite characters from Shake-
speare’s repertoire are Hamlet, Ophelia and King Lear.
And now: exactly what elements from Shakespeare’s play can be transposed
to music, and (since “what music lacks is vocabulary”) (Rosen 1971, 38) how?
I shall consider the issue of this intermedial transposition on three levels suc-
cessively: structure, content and formal narrative devices. I shall then consider
transmedial features shared by literary narratives and musical narratives based
on Romeo and Juliet.
 Intermedial Transposition: From Verbal Story to Music    291

3 Structure
When constructing his five-act drama Romeo and Juliet, Shakespeare relied on
existing literary sources. Nevertheless, it was he who constructed a narrative
which amazes the reader with its use of antithesis on every level and in which the
action has been intensified in relation to both time and quality by embedding it in
a goal-oriented structure using a dramatic crescendo. Shakespeare’s most direct
source was Arthur Brooke’s epic poem The Tragical History of Romeus and Juliet
(1562), which was based on earlier Italian epic versions of the legend. As Geoffrey
Bullough writes:

Brooke’s poem is a leaden work which Shakespeare transmuted to gold. […] In Brooke
Shakespeare found his subject well laid out and ready for quick dramatization, but told
with a turgid emotionalism and pedestrian repetitiveness […]. The surprising thing is that
Shakespeare preserved so much of his source in vitalizing its dead stuff […] (1957, 278)

Primarily, Shakespeare changed the indirect epic mode, with its long descriptions,
into direct drama (although with some epic interpolations, such as the Prologue
and Epilogue or Mercutio’s Queen Mab speech). The presentation of events is also
intensified in Shakespeare’s drama: in Brooke’s poem, “a week or two” passes
between the ball and the lovers’ meeting; in Shakespeare’s drama, they meet on
the same night. Brooke puts “a month or two” between the wedding and the duel;
Shakespeare makes them happen consecutively. Through the direct “collision” of
love and hate scenes, Shakespeare maximized the contrast between them. Shake-
speare was also the first to introduce comic aspects into the myth by using puns
and wordplay, especially in the scenes featuring the Nurse and Mercutio.
The plot of Shakespeare’s play, “divided” into five acts (which are in turn
divided into scenes) can be considered from the angle of Freytag’s Pyramid,
which leads the action “upwards” through the exposition, complication and
climax in the third act (the double apogee of hate and love), whence things take a
tragic turn and the “fall” of the main characters begins: reversal of the action and
catastrophe (see figure 1).
292   Małgorzata Pawłowska

Figure 1: Shakespeare’s Romeo and Juliet according to Freytag’s Pyramid

As illustrated by this figure, the narrative gradually thickens throughout the syn-
tagmatic axis, and the tragic misunderstandings make it tighten like a Gordian
knot leading to the inevitable and repeatedly anticipated catastrophe. The pro-
tagonists of the drama often speak with a sense of foreboding.7 As early as in the
first act, for instance, Romeo says to his fellow revellers:

I fear too early, for my mind misgives

Some consequence yet hanging in the stars
(Romeo and Juliet, I.iv. 107–108)8

7 Todorov distinguishes transformation of supposition in narration: when certain suppositions

or when certain assumptions or premonitions of the protagonists presented in the course of nar-
ration come true later in its course (1990 [1978], 32).
8 All quotations from Romeo and Juliet are taken from the Penguin Books edition edited by G. B.
Harrison (London 1994). His edition is based on the Second Quarto (with certain recommenda-
tions in brackets from the First Quarto).
 Intermedial Transposition: From Verbal Story to Music    293

The misgivings in the statements of various characters accumulate when the plot
speeds up in the second half of the play. For example, Juliet says, while bidding
farewell to Romeo after their wedding night:

O God, I have an ill-divining soul,

Methinks I see thee now, thou art so low,
As one dead in the bottom of a tomb (III. v. 53–55)

The turn of fate “from happiness to misfortune” (the Aristotelian metabole) inher-
ent in tragedy, shown in the pyramid as a climax, is the moment at which the
lovers, who have until now consciously shaped their fate, become powerless,
entangled in the actions of others and in a web of coincidences. The comedic
quality, so important in the two first acts, disappears almost completely; this also
emphasizes the change initiated by Tybalt’s death. In this phase of the narration
appear—in a way characteristic, according to Aristotle, of the reversal of circum-
stances of the fateful phase—the following: peripeteia (dramatic shock), anagno-
risis (recognition) and pathos (violent impression).
How can the structure of Shakespeare’s play be transposed to music? Music
itself possesses, to some extent, the potential for evoking certain extramusical
ideas (I will explore this in the next section). As for the text or programme, music
can try to illustrate it or complement it or it can to serve merely as a medium
“carrying” the text.
We can consider the question of intermedial transposition of the structure
of Shakespeare’s play into music through the prism of the following aspects: the
chronological structure of events (plot), the deeper structure (Freytag’s Pyramid),
the presence of strong direct antitheses and the goal-oriented composition (con-
nected with causality).
Some musical works, especially dramatic genres such as opera, Singspiel and
musicals, are to some degree constructed like drama. They are made up of acts
and scenes, although they are even further broken down into musical numbers
(aria, recitative, chorus, instrumental). In the case of these genres, the libretto
gives us a very clear clue as to which scenes from Shakespeare correspond to
individual fragments. Nor is there any doubt as to which scenes are assigned to
the numbers in Prokofiev’s ballet. Despite its not possessing a layer of verbal text,
each of this ballet’s 52 consecutive numbers are titled by the composer. In some
instrumental music, especially music composed in a single movement without
clear-cut sections, as in Tchaikovsky’s symphonic poem, the chronological struc-
ture of “events” cannot correspond to Shakespeare’s drama. At first glance, it
is mainly the “emotional content,” general atmosphere or general ideas of the
theme of Romeo and Juliet that is transposed to Tchaikovsky’s piece, rather than
294   Małgorzata Pawłowska

the structural elements. However, we can still determine whether or not there is
some degree of correspondence to other structural features.
Having analysed many of the musical works inspired by the story of Romeo
and Juliet (Pawłowska 2013a), I can say that most of them seem, on a general
level, to follow the Freytag’s Pyramid structure, as Shakespeare’s drama does.
At some point around the middle there is a metabole, after which the musical
language also changes. There is an increasing number of ostinato motifs (often
played on gran cassa) and a stretto towards the end, constituting a accumula-
tion and juxtaposition of musical themes.9 Music can evoke more than one phe-
nomenon simultaneously, superimposing multiple layers of meaning at various
moments during the performance. In the works of music inspired by Romeo and
Juliet, one can observe a tendency to accumulate musical themes not only consec-
utively but also vertically, in the second half of the composition.
As in Shakespeare’s narrative, these musical narratives inspired by the story
of Romeo and Juliet correspond to the dramaturgic crescendo. The organic devel-
opment and transformations of musical themes make music more of an abstract
process, directly experienced and felt, rather than a tale with precise conceptual
meanings. In musical works based on Romeo and Juliet, there is a tendency to
apply anticipation, as well as to surprise the listener with explosions of sounds.
This seems to reflect the spirit of Shakespeare’s drama (“bad feelings” and the
final catastrophe).
In the narrative trajectory of Shakespeare’s play, we observe the transforma-
tion of the protagonists, especially the deep transformation of Juliet from a naïve
young girl to a suffering mature woman. Music can express these changes through
development and transformational techniques of the musical material, as asso-
ciated, for instance, with characters. Thus, Prokofiev, in his ballet, uses musical
motifs and themes representing Juliet, Romeo and Mercutio. Juliet’s themes are
profoundly transformed in the course of the narrative trajectory; they occur at
the end in a completely different musical context than at their first presentation.
In musical interpretations, the antitheses and contrasts present in Shake-
speare’s narrative are usually emphasized, since musical structures make use
of contrasts in a very distinctive way. The semic opposition of love (Romeo and
Juliet) and hatred (the feud) seems to have suggested the idea to Tchaikovsky of
composing his work in sonata form—a form for which the contrast of two main
themes is a crucial factor.

9 A musical theme is understood as recognizable and coherent musical material (especially a
recognizable melody).
 Intermedial Transposition: From Verbal Story to Music    295

If we were to compare the chronology of events in Shakespeare’s play with

that in the most representative musical works inspired by the story of Romeo and
Juliet, we would find that the closest to Shakespeare’s can be found in Prokof-
iev’s ballet and Gounod’s opera, even though in the case of the latter, different
elements of the plot are emphasized. Interestingly, Bernstein’s musical West Side
Story is very close to Shakespeare’s chronology, merely transferring the place
and time to Manhattan (New York) of the 1950s. Bernstein’s musical mirrors the
Shakespearean plot, finding contemporary equivalents for the situations and
characters. Benda, in his Singspiel Romeo und Julie, and Bellini, in his opera I
Capuleti e i Montecchi, both begin the action after the key duel, that is, after the
first half of Shakespeare’s narrative.
In the works of Berlioz and Tchaikovsky, the structure of Shakespeare’s
drama is deconstructed. Berlioz narrates some events twice, first using literary
words, then by purely instrumental means. Tchaikovsky extracts and abstracts
basic narrative units, primarily in the juxtaposition of love and hatred, but also
with the character Friar Laurence;10 in the course of the narrative, we can also
hear evocations of death, fate and tragedy.
Surprisingly, only Berlioz and Gounod included the episode with Mercutio’s
story about Queen Mab, presenting the peculiar metaphysics of Shakespeare’s
unreal worlds in scherzo-like music.
Farthest from Shakespeare, however, is the structure of works of so-called
contemporary classical music, from the second half of the twentieth century.
Interestingly, while popular musicals (or those on the border of popular and clas-
sical music) taking up the subject of Romeo and Juliet present the plot of the myth
more or less faithfully (even if transposed into different realities), “highbrow”
music has gradually come to evade the linear presentation of Romeo and Juliet’s
story. In the operas of Blacher (1950), Matuszczak (1970) and Dusapin (1988), one
can find references to an archetype: the first two use fragments of Romeo and
Juliet’s dialogue while the latter no longer has anything in common with Shake-
speare’s drama. Grisey in 1984 evokes the names of mythical lovers (Romeo,
Juliet, Tristan) in Chants de l’amour, a work based entirely on the phrase “I love
you” pronounced in 22 languages by computer-modified voices. Here, the loving
pair are subject to the madness of multiplication, which seems to be symptomatic

10 Tchaikovsky’s symphonic poem Romeo and Juliet is a composition devoid of verbal text or
titles of sections, but it includes certain musical themes corresponding to elements of the play’s
plot, such as the musical theme of love, the musical theme of hatred and the musical theme of
Friar Laurence. We find evidence concerning these meanings in Tchaikovsky’s correspondence
with Mily Balakirev.
296   Małgorzata Pawłowska

of its time: both Matuszczak and Dusapin multiply the number of main charac-
ters; in Matuszczak’s work there are three Romeos and Juliets; in Dusapin’s the
characters are doubled.

4 Content
Can the content of any story be transposed to music? How do we know what is
being told through music? The problem of the musical signifié is extensive (cf.
Pawłowska 2014).
Certainly music cannot express meanings in the way that literature does. The
affordances of these two media differ in significant ways. Not only does music
lack a vocabulary (Rosen 1971, 38), but it is also incapable of making propositions
(Micznik 2001, 218); moreover, there is no link between subject and predicate in
music. Hence, the nuanced Shakespearean plays on words and puns cannot be
transposed to music.
Nevertheless, music can suggest certain extramusical phenomena, especially
those connected with emotional content. It can convey emotions in a direct way.
As Piotr Podlipniak writes:

Music omits the process of a conscious and costly (in terms of time and energy) process
of information decoding and uses a shorter, direct way, circumventing consciousness—the
impression. (2007, 186; my translation)

In his analyses of music, the musicologist Eero Tarasti applies Greimasian modal-

Modalities are general human ways of evaluation […]. As a series of emotional states, modal-
ities account for the way the listener unites a musical text with human values. (Tarasti, 1991,
The prevalent modalities of music are ‘being’ and ‘doing’, in addition to the normal tempo-
ral process of music, which I call ‘becoming’. Being means a state of rest, stability, and con-
sonance; doing is synonymous with musical action: event, dynamism, and dissonance […]
The basic modalities of being and doing are sur-modalized by several others: will, the
so-called kinetic energy of music, its general direction, its tendency to move toward a goal;
know, the information conveyed by music, its cognitive moment; can, the power and effi-
ciency of music […]; must, the control exercised by the rules of genres and formal types […];
believe, the epistemic values of music […] (Tarasti 2004, 295‒296)

According to Susan Langer, musical forms bear a close logical resemblance to the
forms of human feelings. Music is a “presentational symbol” of a psychic process
and its tonal structures bear a close logical similarity to the forms of feeling,
 Intermedial Transposition: From Verbal Story to Music    297

“forms of growth and of attenuation, flowing and stowing, conflict and resolu-
tion, speed, arrest, terrific excitement, calm, or subtle activation and dreamy
lapses” (Langer 1953, 27).
This is why musical works inspired by the story of Romeo and Juliet can
present emotions and actions, enact them, imitate them. In the love scenes of
Berlioz’s and Tchaikovsky’s works we hear the dynamic of amorous passion; in
the fragments corresponding to duels we hear the iconic signs of the sounds of
swords. In Berlioz’s symphony and in Prokofiev’s ballet we can hear the imitation
of Romeo’s steps approaching Juliet’s tomb, including all his halts and hesita-
tions, conveying the morphology of the character’s psychic processes.
The problem of the affordances of instrumental music in conveying meanings
and of text-music relations has often been the subject of composers’ reflections.
Berlioz was especially preoccupied with the question of how he should compose
Roméo et Juliette: with a text (singing parts) or without? In the end, he chose to
include the vocal-text element only in certain parts of his symphony; most of it,
though, is expressed through instrumental music without text. In his introduc-
tion to the score, Berlioz wrote he believed that instrumental music, “richer, more
varied, less restricted” than words, is a better means of conveying the sublime
feelings of love and despair (Rushton 1994, 87–88).
The powerful capacity of music to convey emotional content and its limited
ability (or inability) to express the nuances and plays on words results in “dis-
torting” the content of Shakespeare’s drama when it is transposed to music. In
many musical interpretations, love and despair are emphasized while leaving out
details of the plot, especially its comic elements.
However, it is not only the emotional content or presentation of psychic pro-
cesses that can be conveyed by music. Musical narratologists agree there are
structures of signification even in so-called absolute music (music without text or
programme). Some musical motifs and themes possess a denotational quality. In
musicology musical themes are termed “topics.” As Kofi Agawu writes:

Topics are musical signs. They consist of a signifier (a certain disposition of musical
dimensions) and a signified (a conventional stylistic unit, often but not always referential
in quality). Signifiers are identified as a relational unit within the dimension of melody,
harmony, meter, rhythm, and so on, while the signified is designated by conventional
labels. (Agawu 1991, 49)

The indexical, iconic and symbolic nature of musical topics has been discussed
in musicology (on topic theory, see Ratner 1980, Agawu 1991, Hatten 1994, Tarasti
1994 and Monelle 2000). Certain ways of presenting extramusical themes have
become conventional in the course of music history.
298   Małgorzata Pawłowska

Comparison of musical compositions on Romeo and Juliet that differ in genre

and historic style confirms that there are similarities between fragments of works
conveying such narrative units as love, hatred, battle, or death.
The topics of court dances, representing a certain stylization and referring
to Renaissance Verona, are among the easiest to grasp. They appear in many
musical works inspired by the legend, exploited especially in Prokofiev’s ballet
and in Rota’s film music.
Funeral topics, characterised mainly by a minor key, slow tempo and march-
like rhythm, are also recurrent. In certain musical interpretations (Benda, Bellini,
Gounod, Prokofiev), there appears an extensive scene of Juliet’s funeral which
has no equivalent in Shakespeare’s drama (this being rather a result of Garrick’s
adaptation): its characteristic quality is the use of chiaroscuro, ombra and alla
breve topics. In both Benda’s and Gounod’s works (as well as in Gluck’s Orpheus),
these scenes are expressed in C minor.11 In the majority of musical works inspired
by the Romeo and Juliet story, the course of narration contains short, obsessive,
mainly narrow-frequency, “acute” motifs which can be read as semes12 of death.
The “last gasps” of the lovers in the tomb, or of the dying Mercutio, are usually
set to music with defragmented phrases, motifs and sighing gestures, often in
the pianto topic. A non-Shakespearean version,13 with the inclusion of a final dia-
logue before Romeo and Juliet’s death in the tomb (Juliet awakens while Romeo
is still alive), has found its way into many musical interpretations. After all, this
made it possible to include the final operatic duet (Bellini and Gounod), while
in Berlioz’s work it was a presentation of the instrumental “idea of a dialogue”
(Micznik 2000) in music. The tomb dialogue also appears in Benda’s work, where
the lovers are rescued from the oppression of death, changing the ending com-
pletely. Prokofiev, too, nearly fell prey to the temptation of a happy ending. As
he wrote in his Autobiography: “The reasons for this bit of barbarism were purely
choreographic: living people can dance, the dying cannot” (Prokofiev 2000

11 According to Ferdinand Hand’s catalogue (dating from 1837, quoted after Golianek 1998, 12),
C minor connotes sadness, mourning and lament; according to Paul Ertel’s catalogue (dating
from 1896, quoted after Golianek 1998, 12), C minor is the key of the funeral march (128). What
needs to be emphasized is that this key (no matter if it has a “natural,” acoustic explanation)
became conventionally and symbolically associated with the aforementioned meanings, and
therefore it was the choice of many composers to write the funeral scenes in the key of C minor.
12 On semes and classemes in music, cf. Grabócz (1996).
13 This version was popularised by Garrick. It is not known to what extent such a finale was in-
fluenced by the previous renderings of the myth by da Porto and Bandello, in which the dialogue
between the lovers before their death also appears.
 Intermedial Transposition: From Verbal Story to Music    299

[1960], 86). Eventually the composer changed his mind and remained faithful to
Shakespeare, because a happy ending did not sound true to him.
In any case, rewriting the literary source in some musical works is connected
with the affordances of the medium, as can be seen in the funeral topic, the final
singing duet of Romeo and Juliet in the tomb, or the happy ending.
The elements from the Romeo and Juliet myth that seem to be most consist-
ently transposed to music are simply the general ideas of love and hatred around
which the story is centred.
Shakespeare’s love dialogues, naturally prone to musicalisation, gain an
expressive reinforcement in all the works discussed here, either as sung duet
or as purely instrumental music. It is no wonder that music is the appropriate
medium for this, when we find such words in the dialogues as:

How silver-sweet sound lovers’ tongues by night,

Like softest music to attending ears!14 […]
… Let the rich music’s tongue
Unfold the imagin’d happiness15

As previously mentioned, Berlioz in his dramatic symphony consciously chose

to deprive the most significant love scenes of any words because he believed that
pure music was a more versatile and sensitive tool for expressing feelings than
music combined with text. Likewise, Dusapin decided not to use words in the
central part of his opera.
When comparing pieces based on Romeo and Juliet (of various styles, times
and genres), it is possible to discern similarities in the way they convey the idea
of love and of hatred. These had become conventional through music history,
although there may be a natural explanation for their characteristics. Hence, we
can also speak about the topics of love and hatred in music. However, it needs to
be emphasized that the qualities used by composers to convey the ideas of love
and hatred are not restricted to this signification. There can be absolute music
without an extramusical programme having the qualities described below, but
a composer aiming at giving a musical representation of love and hatred would
probably choose these topical qualities. We can find evidence of this in the text
layer (if the composition includes text), in the title, in the titles of respective
movements and in composers’ personal reflections.

14 Romeo’s words, balcony scene (II.ii.166‒67).

15 Romeo’s words, wedding scene (‒27).
300   Małgorzata Pawłowska

Therefore, hatred in these musical works is usually expressed through dis-

sonant harmony; short, abruptly ending motifs with dense structures and dotted
rhythms; repeated notes or rapid scale passages; sharp articulation; loud dynam-
ics; ensemble and instrumentation—tutti, with emphasis on sharp instrumental
timbres and sudden percussion strokes. This type of musical arrangement in the
works on Romeo and Juliet usually applies to the sections recalling the Capulet
and Montague families. Conversely, love in music is most commonly expressed
through cantilena-like melodies, consonant harmonies, legato articulation, and
the use of soft, warm instrumental timbres. Love as a musical topic has been cul-
turally (conventionally) grounded at least since the medieval troubadours.
It is quite natural that hatred is usually conveyed as disharmony, detach-
ment (short and sharp articulation), and movement “away from,” and love as
harmony, connection (legato articulation) and movement “towards.” Philosoph-
ically, love and hate are often combined as similar phenomena with opposing
spatial vectors. Ortega y Gasset (1941 [1939], 4–5) describes them as streams of
feelings or energy: love unites us with the object, whereas hate is a metaphysical
dissonance. Similarly, Descartes, in The Passions of the Soul, a study considering
feelings from the biological and psychological perspective, writes that whereas
love incites the soul “to join itself in volition to the objects,” hatred “incites the
soul to […] be separated from the objects” (1989 [1649], 62).
While hatred in musical expression is always a struggle for separation, in
the musical expression of love we can observe two modalities (cf. Pawłowska
2013b). One is connected with the lovers’ unity or symbiosis and is shown by
means of symbiotic symbols (such as singing in unison) and by creating a general
atmosphere of intimacy: a slow tempo, pp dynamics, the use of “soft” muted
instrumental timbres—con sordino strings, woodwind instruments, harp. Among
the expressive specifications given by composers we find amoroso, misterioso,
dolce and sensibile. The other modality is connected with desire, with an aspira-
tion towards and temporary achievement of ecstasy. Musically, it is manifested
through the use of a full range of means within a given style including melos
ascendens, ff dynamics and accelerando as well as directions such as appassion-
ato and espressivo.
The protagonists of Shakespeare’s play (especially Romeo and Juliet) are
symbolized in many works by recurring musical themes or motifs, easy to identify
and remember (for instance, in Prokofiev’s ballet). Such musical themes, adopt-
ing individual features and being subject to modifications and transformations
throughout the composition, but still preserving their identity, make it easy for
a listener to follow the “plot” of the musical narrative. While listening, one syn-
thesizes its consecutive phases so that it emerges as an integrated entity in which
something has “happened.” Given the title Romeo and Juliet, one is invited to
 Intermedial Transposition: From Verbal Story to Music    301

follow this particular plot in one’s imagination; however, the narrative qualities
of music do not require imagining a specific extramusical story (cf. Berger 2000,
Pawłowska 2013a, 2014).

5 Formal Narrative Devices

Wolf, in his article devoted to intermediality, mentions “formal intermedia imi-
tation,” a characteristic feature of which is “the attempt to shape the material of
the semiotic complex in question […] in such a manner that it acquires a formal
resemblance to typical features or structures of another medium” (2008a [2005],
255). “Literalisation” of music (as in programme music) is an example.
Indeed, in the musicological debate mentioned at the beginning of this
article, even those who are sceptical about regarding musical works as narra-
tives (e. g., Nattiez or Abbate) admit that music can resemble a narrative style or
imitate a narrative mode from literature.
It has been claimed that music bears more similarity to drama than it does
to epic and thus that it represents or enacts actions rather than relate them (see,
for instance, Maus 1988). In narratological musical interpretations, actants (or
agents) are often identified as, for example, two opposing forces represented by
musical themes or groups of instruments. This can be most readily observed in
the concerto genre. As Joseph Kerman explains:

While plenty of exceptions exist […] in general one knows exactly who is who in a concerto
and who is doing what. There is a soloist and an orchestra, and there is usually quite a sharp
sense of character, of “the powerful and multicolored orchestra and its weak but high-spir-
ited adversary,” as Tchaikovsky once put it. The agents exist in some kind of relationship,
and what is traced in a concerto is the course of a relationship. (Kerman 1992, 97‒98)

In a certain type of classical music—mainly from the eighteenth and nineteenth

centuries—musical themes acquire individual, personal features. The work of
music, integrated by the main theme that maintains its identity throughout the
transformations, can be understood as an expression of an enduring individual
moral “character” (Berger 2000, 200). The use of contrast and disruptions in the
musical “discourse” invites further comparisons with drama or with plays per-
formed on stage. In his narratological analysis of Beethoven’s Waldstein Sonata,
Tarasti makes the following observation:

A basic hypothesis of the present analysis […] is that a musical actor does not disappear
“from the stage” even though it is not “saying” something, just as theater performers in
dialogue do not disappear when the speaker changes. On the contrary, the actors remain
302   Małgorzata Pawłowska

present throughout the discourse, but in absentia, as destinatees who immediately interpret
and respond to the utterance of their interlocutor. What happens in music corresponds to
theatrical communication: a musical theme-actant might disappear from the score, say at
the moments when its opponent’s theme occurs in another register; but it does not disap-
pear from the listener’s mind. (Tarasti 1991, 100‒101)

Tarasti asserts that music can be an “arena” of simultaneously presented musical

actions (at different levels), such that Bakhtin’s concept of the polyphony of the
novel can also be applied to it. In any case, in Tarasti’s understanding of a musical
“event” results not only from the interaction between actors but also from sharp
contrasts of the spatio-temporal dimension such as slowing down/acceleration,
sudden changes of register, etc.
Direct presentation, showing and enacting of events (as in drama), has tra-
ditionally been identified with the mimetic mode of narration, as opposed to the
diegetic mode, which (as in an epic) corresponds to indirect storytelling, telling
and recounting of events (cf. Prince 2003 [1987], 52‒53; Schaeffer and Vultur
2008 [2005]; Shen, 2008 [2005]).16 The comparison of music with drama or plays,
together with the mimetic affordances of music discussed earlier in this article,
suggests that music has a potential for coming within the mimetic mode of nar-
Even so, differentiating between the two modes with reference to musical nar-
ratives can still prove useful. Some musical works, or their fragments, by seeming
to resemble literary epics, invite comparison with the literary diegetic mode. In
musicological writings, it is not only the concept of actants-agents-actors that
is applied to music, but also the concept of a narrator who tells the “story.” The
subject of musical narration is, in Tarasti’s opinion, “a kind of an inner narrator
of music, the ‘I’ who experiences various phases and changes between ‘being’
and ‘doing,’ but who cannot be identified with any real subject” (1995, 60).
This diegetic quality can manifest itself, for example, in narrative frames such
as the beginning and ending of certain works, where archaic-sounding themes
function as the opening and closing of a mythical world (cf. Tarasti 1979, 67),
as occurs in the slow introduction to Tchaikovsky’s symphonic poem (chorale
topic). We can compare this type of introduction in a musical story to the voice of
a narrator saying “once upon a time…” and, actually, to Shakespeare’s Prologue
in Romeo and Juliet, with the narrator’s voice saying:

16 The terms mimesis and diegesis also have other meanings and uses. Mimesis refers to the
more general imitation of reality in the arts, whereas diegesis may also refer to a fictional world
in which certain events and situations occur. Here I use these terms in a more narrow narrato-
logical sense.
 Intermedial Transposition: From Verbal Story to Music    303

Two households both alike in dignity,

(In fair Verona where we lay our scene)
From ancient grudge, break to new mutiny […]

In Zeffirelli’s film it is Rota’s archaic-sounding music (menuet), heard in the first

seconds, which transports the viewer into “mythical time.”
The mimetic mode of narration is apparent, for example, in the opening
Allegro fugato from the dramatic symphony by Berlioz, which places the listener
in medias res of the narrative. This purely instrumental passage illustrates the
initial street brawl from Shakespeare’s drama. At some point the fugue is inter-
rupted by a recitative of brass instruments (the composer provides the indication
“in the character of a recitative”); in terms of the programme, this passage corre-
sponds to the Prince’s intervention in Verona’s square. We can thus recognize dis-
tinct actants, even though no words are actually spoken. In the love or “balcony”
scene by Berlioz, expressed only through instrumental means, the idea of the
lovers’ dialogue, shown in the changeability of instruments or of melodic mate-
rial (cf. Micznik 2000), is presented. The “idea of dialogue” musically achieved
in this love scene has proven so convincing that it encourages some interpreters
to match specific words from Shakespeare to Berlioz’s musical phrases (Kemp
In musical narratology, the concept of a narrative “unsung voice” appears (cf.
Abbate 1991). This voice is defined by Abbate as “not literally vocal performance,
but rather as a sense of certain isolated and rare gestures in music, whether vocal
or non-vocal, that may be perceived as modes of subjects’ enunciations” (1991,
ix). Robert Hatten, in reference to such sections, talks about a narrative agent
(2004, 225–226). Sometimes such a role is performed by instrumental recitatives
which interrupt the musical plot in order to “comment” on it. This happens often
in Berlioz’s symphony as well as in Dusapin’s opera. We may even say that in
Dusapin’s work the solo clarinet becomes a significant actant. Its constant pres-
ence—closing phrases and movements, accompanying vocalists in sections
where the entire orchestra remains silent—contributes to its integrative and con-
trolling function, similar to that of a narrator. The composer himself admits in the
introduction to his score that:

The clarinet player is a character in his own right, and could perhaps be seen as a “wild
card.” He is watching, listening to and telling the tale, and also accompanying it. He is a
“healthy” character, with no psychological baggage. Words do not impinge on his musical-
ity. He sings, but through his instrument only. He has no doubts. (Dusapin 1988)
304   Małgorzata Pawłowska

The diegetic quality of music also lies in its ability to convey an illusion of tem-
poral distance. Tarasti, in his book Myth and Music (1979), claims that this can
be achieved not only via archaic topics or timbres but also through the apposite
use of musical motifs that are repeated in the course of the work so that the lis-
tener can compare later passages with earlier ones. According to Tarasti, if a com-
position begins with a distinct main theme and develops dramaturgically, and
if a number of musical events occur afterwards, then a recurrence of the theme
at the end can suggest that a great deal of time has passed in the musical story
(1979, 67–68). The listener, comparing passages, may have the impression that
“a lot has happened” since the theme’s first presentation, especially since such
repetitions of a theme at the end of a story have a special significance (different
augmentation, instrumentation, etc.). An example of this can be found in Prokof-
iev’s ballet. Here, several basic themes are assigned to particular characters and
situations (e. g., the fate theme, Juliet’s theme, the family theme). At the conclu-
sion, after a number of musical disruptions, the fate theme returns, exposed and
augmented. Juliet’s theme also returns, strongly processed, in a higher register
and is repeated up to five times in different keys. One may have the impression
that a certain stretch of time in the story has modified these themes.
In Dusapin’s opera there is a distinct tension between the diegetic and the
mimetic mode. There is a narrator (Bill), who is sometimes placed outside the
presented world and sometimes in the middle of it. The use of metalepsis17 in this
work leads us to the next section of the article: transmediality (and plurimedial-

6 Transmedial Features
As we have observed, there are a number of features of Shakespeare’s drama
Romeo and Juliet that can be transposed to music. However, in many cases the dif-
ferences between the affordances of literature and music result in drastic changes
to the original narrative. Musical works, compared to verbal narratives, are more
abstract processes. As Walter Pater writes:

All art constantly aspires towards the condition of music, because, in its ideal, consummate
moments, the end is not distinct from the means, the form from the matter, the subject from
the expression […] (1986 [1873], 135)

17 On metalepsis as a transmedial phenomenon, see Pier (2014 [2009], § 3.2.2) and Wolf (2005).
 Intermedial Transposition: From Verbal Story to Music    305

However, many scholars agree that, taking into consideration the broad defini-
tion of a narrative as a transmedial phenomenon, some musical works can be
What transmedial features are shared by the literary narrative of the Romeo
and Juliet story, and what musical narratives are inspired by the story?
Most of these features seem to lie at a very general, deep narrative level. In
order to see these transmedial features at this level, we need to pass through
many other levels of the text (i. e., to skip them somehow). It is not my intention to
simplify Shakespeare’s complex and multifaceted narrative, but simply to show
that it is at this general level that we can see most transmedially shared features.
As already mentioned, Shakespeare’s narrative is full of antitheses, oxymo-
rons and contrasts. This was one of the aspects that made Romeo and Juliet myth
so ripe for musicalisation, since the use of contrast and antitheses are indeed
features that cross media.
A narrative can be understood as a process in which the subject, characters,
situations or values undergo change. According to Tzvetan Todorov, narrative is
governed by two leading principles: succession and transformation (1990 [1978],
28‒30). The principle of transformation is related to the processual aspect of nar-
rative. Almén (2008) claims that even if musical narrative does not present spe-
cific characters or situations, the arrangement of qualities or values undergoes
change. He particularly draws on James Jakób Liszka’s work The Semiotic of Myth,
where the concept of transvaluation appears as crucial to narrative. Transvalua-
tion occurs as a result of changes in hierarchy and in the arrangement of values
presented in narrative. Almén’s definition is as follows: “Musical narrative is the
process through which the listener perceives and tracks culturally significant
transvaluation of hierarchical relationships within a temporal span” (2003, 12).
Conceived in this way, transvaluation seems to be applicable both to Shake-
speare’s narrative and to the musical works based on it. On a very general level, at
the end of a narrative we find qualities opposed to the original. The starting point
of the drama, namely, the feud between the families, is transformed into a state
of reconciliation. The birth of love (or, to use Dante’s metaphor, “the birth of new
life”)18 turns into death. Likewise, Friar Laurence’s willingness to help becomes,
through tragic misunderstanding, a cause of calamity; hence an actant helper

18 In Dante’s Vita nova, we read:

In that book which is my memory,
On the first page of the chapter that is the day when I first met you,
Appear the words, ‘Here begins a new life.’
(Dante 2012 [1295]).
306   Małgorzata Pawłowska

(in the actantial system of Greimas) turns into an actant opponent, and a reshuf-
fling occurs in the actantial system.19 The culmination of the tragedy is a double
climax: one of hatred and of love. In the course of the day, the two most militant
representatives of both families die in duels. The night joins two others, mad with
love. All of these could be termed “transformations through negations” (Todorov)
in which the original seme turns into its opposite, more of which we can find in
the pages of the drama. In the text itself, Capulet states:

All things that we ordained festival,

Turn from their office to black funeral;
Our instruments to melancholy bells,
Our wedding cheer to a sad burial feast,
Our solemn hymns to sullen dirges change,
Our bridal flowers serve for a buried corse,
And all things change them to the co