(Film Culture in Transition) Steffen Hven - Cinema and Narrative Complexity - Embodying The Fabula-Amsterdam University Press (2017)

FILM
CULTURE
IN TRANSITION
Cine m a a n d
Narrative
Com p l e x i t y
embodying the fabula
steffen hven
Cinema and Narrative Complexity
Cinema and Narrative Complexity
Embodying the Fabula
Steffen Hven
Amsterdam University Press

For Gefion
Cover illustrations: (front) Photo: Ruth Harriet Louise - Buster Keaton, The Cameraman
(1928), directed by Edward Sedgwick; Collection Christophel / RnB © Metro Goldwyn Mayer;
(back) frame from Memento (2000), directed by Christopher Nolan.
Cover design: Kok Korpershoek, Amsterdam

Lay-out: Crius Group, Hulshout
Amsterdam University Press English-language titles are distributed in the US and Canada by
the University of Chicago Press.
isbn 978 94 6298 077 8

e-isbn 978 90 4853 025 0
doi 10.5117/9789462980778
nur 670
© S. Hven / Amsterdam University Press B.V., Amsterdam 2017
All rights reserved. Without limiting the rights under copyright reserved above, no part of
this book may be reproduced, stored in or introduced into a retrieval system, or transmitted,
in any form or by any means (electronic, mechanical, photocopying, recording or otherwise)
without the written permission of both the copyright owner and the author of the book.
Table of Contents
I. Introduction 7
II. Cinema in the Interstices 25

Introduction 25
Cinema and its two Temporal Modes 26
The Cognitive Fabula and Linear Storytelling 29
Film-Philosophy and the Challenge of Contemporary Cinema 41
III. Narrative Ambiguity in the Classical Cinema 53

Introduction 53
Narrative Transgression in Stage Fright 58
Defamiliarization and Beyond 69
IV. Modern(ist) Cinema: Logic of the Encounter 83

Introduction 83
Cinema and the Modern 84
Hiroshima mon amour – A Cinematic Love Affair 93
V. Towards the Embodied Fabula 111

Introduction 111
Towards an Embodied and Enactive Approach to Cognition 113
Embodying the Narrative Experience 121
VI. The Complexity of Complex Narratives 137

Introduction 137
21 Grams – How Much Does Life Weigh? 141
The Fabula and the ‘Decomplexification’ of the Narrative Continuum142
A Problematic Combination: Classical Narratology and Com-
plexity Theory 150
‘Forking Paths’ – Complex Storytelling in Lola rennt 157
Are Forking Paths Linear? 161
Embracing the Narrative Rhythm of Lola rennt 166
VII. Memento and the Embodied Fabula 175

Introduction 175
Realism and Constructivism: A Narratological Deadlock 180
The Defamiliarization of Narrative Perception 193
VIII. Conclusions 205
The Embodied Fabula and Beyond 210
Bibliography 215
Notes 229
Acknowledgments 245
Index 247
I. Introduction
No method nor discipline can supersede the necessity of being forever on the
alert. What is a course of history, or philosophy, or poetry, no matter how well
selected, or the best society, or the most admirable routine of life, compared with
the discipline of looking always at what is to be seen?[...] Read your fate, see what
is before you, and walk on into futurity.
‒ Thoreau 1995, 72
The brilliance of Christopher Nolan’s Memento (2000) lies in its complex

narrative structure, which imposes a sensation of temporal disorientation
upon its viewers that mirrors the anterograde amnesia suffered by its main
character, Leonard Shelby (Guy Pearce). Thereby the film allows the specta-
tor to enact – rather than merely observe – the amnesia of what has become
‘the archetypal example of the character who suffers from a loss of memory’
(Elsaesser 2009, 28). Consequently, Memento stands out as one of the most
vivid representatives of a contemporary body of films that have challenged
the long-dominant opposition between classical Hollywood storytelling
and the tradition of (European) art cinema (cf. Kovács 2007, 33-48).1 Pulp
Fiction (Tarantino 1994), Lola rennt ([Run Lola Run] Tykwer 1998), Be-
ing John Malkovich (Jonze 1999), Fight Club (Fincher 1999), Amores
Perros ([Love’s a Bitch] Iñárritu 2000), Oldboy (Park 2003), 21 Grams
(Iñárritu 2003), Eternal Sunshine of the Spotless Mind (Gondry 2004),
2046 (Wong 2004), Inception (Nolan 2010), Source Code (Jones 2011), and
Coherence (Byrkit 2013) are but a few examples of the international surge
within the landscape of moving images to develop increasingly demand-
ing and challenging narratives.2 These ‘complex narratives’ (Simons 2008)
embrace non-linearity, time loops, and fragmented spatio-temporal realities
(cf. Buckland 2009a, 6) to demonstrate a contemporary interest in personal
identity, memory, history, trauma, embodied perception, and temporality
(cf. Elsaesser & Hagener 2010, 149). Located somewhere in the encounter
between film and spectator (cf. Deleuze 2005b; Engell 2005; Pisters 2012;
Brown 2013), the complexity of these complex narratives turns out to be a
complex phenomenon itself (cf. Simons 2008, 111).
Thus, perhaps not surprisingly, there are disputes on how to comprehend
‘complex narratives’, ‘puzzle films’ (Buckland 2009b), ‘mind-game’ movies
(Elsaesser 2009), ‘modular’ narratives (Cameron 2008), ‘forking-path’ narra-
tives (Bordwell 2002a), ‘hybrid’ films (Martin-Jones 2006), or ‘neuro-images’
8 Cinema and Narr ative Complexit y
(Pisters 2012) film-historically. The most detailed definitions of what shall

be referred to as the contemporary complex narrative are to be found in this
body of research, yet no agreement exists as to whether the films in question
constitute a break with, or should rather be perceived as an extension of,
the boundaries of Hollywood’s canonical storytelling format. The variety
of answers to this question is demonstrated by the lack of both consensus
in terminology and a clearly defined body of films.
Buckland’s (2009a) term ‘puzzle films’ refers to ‘a popular cycle of films
from the 1990s that rejects classical storytelling techniques and replaces them
with complex storytelling’ (1; see also Buckland 2014). Following Elsaess-
er (2009), the ‘mind-game films’ cross ‘the usual boundaries of mainstream
Hollywood, independent, auteur film and international art cinema’ (13) and
comprise films in which games are being played with a character, as well as
films which play with its audience (14). For Cameron (2008) it is central that
‘modular’ narratives offer ‘a series of disarticulated narrative pieces, often
arranged in radically achronological ways via flashforwards, overt repeti-
tion or a destabilization of the relationship between present and past’ (1).
Pisters (2012) argues that with the ‘neuro-image we have quite literally moved
into the characters’ brain spaces. We no longer see through characters’ eyes,
as in the movement-image and the time-image; we are most often instead
in their mental worlds’ (14). Ultimately, Simons (2008) maintains that ‘[i]n
spite of all this diversity and the different ways of approaching and assessing
this body of films, most theorists would agree to subsume these films under
the predicate “complex narratives”’ (111). This book argues that inquiries into
the nature of these films have been framed by overarching oppositions (e.g.
classical versus modern(ist) cinema, linear versus non-linear temporality)
that the films themselves have left behind (cf. Shaviro 2010, 2012).3
In her contribution to Hollywood Puzzle Films (Buckland 2014), Maria
Poulaki (2014b) maintains that narratives, as well as the narrative mode
of reasoning, prioritize wholes over the parts (36-37). Narrative events, she
argues in reference to the narratologist Donald E. Polkinghorne (1988),
make sense by forming a meaningful whole, where the events are perceived
to cause each other. Meanwhile, the narrative is itself embedded with the
expectation that it will eventually make sense according to a causal-linear
schema. Given that such a schema defines how we think about narratives,
scholars have tended to understand complex narration in terms of its devia-
tion from causal linearity. As Poulaki (2014b) goes on to argue, this approach
has lost its ability ‘to provide further insight into the complex and non-linear
structure of complex films, particularly at this point in time when the latter
seem to have established a new paradigm of cinematic storytelling’ (37).
Introduc tion 9
When it comes to narrative complexity, Poulaki states: ‘It is no longer

enough to show how complex films are not conventional narratives; the
need for a positive definition and description of their processes has become
apparent’ (37). According to her, the most valuable lesson of complexity is
its insistence on ‘processes of resonance between individual components or
units, and how the forms they create are never whole or complete, neither
in the beginning nor at the end’ (48). Along similar lines, this book argues
that in order to better comprehend contemporary complex cinema in its
own terms, the formulation of a new mode of spectatorship that enhances
cinematic perception by allowing spectators to ‘embody’ the narrative
universe is required. In contrast to most prevalent studies, I argue that the
complexity of contemporary cinema does not primarily rest in a complex,
entangled, or complicated syuzhet or dramaturgy but owes to a ‘will to
complexity’ – understood as an insistence on the mutual dependence of
cinematic dimensions that have traditionally been kept apart.
From this perspective, contemporary cinema not only calls for a re-
newed appreciation of what shall be referred to as the ‘linear-non-linear’
dichotomy, but also forces us to rethink the interrelation of the cognitive,
emotional, and affective circuits that constitute the cinematic experience.
Therefore, parallel to studying a variety of modes in which cinema elicits
spectators’ affective, emotional, or cognitive responses, I question favoured
interpretative strategies with the aim of formulating an alternative ap-
proach designed to open up rather than closing down, ‘straightening out’,
or ‘decomplexifying’ the narrative continuum.
In relation to this, the word ‘affect’ (l’affect or affectus) – as differentiated
from cognitive states and describing the bodies’ capacity to move and be
moved, to affect and be affected – becomes central. Within Deleuzian affect
theory, as Shouse (2005) explains, emotions are object-oriented and social
phenomena, whereas affect is prepersonal. Within cognitive theory, as
Plantinga (2009a) explains, ‘emotions are intentional in the sense that they
are directed toward some “object”’ (86), whereas affect is a broader category
that comprises any ‘felt bodily state, including a wide range of phenomena,
including emotions, moods, reflex actions, autonomic responses, mirror
reflexes, desires, pleasures, etc.’ (87). In brief, whereas the Deleuzian frame-
work tends to separate the ‘affective’ too abruptly from the cognitive and
emotional sphere, the cognitive framework tends to reduce the affective
to cognitive-emotional components. 4 As Seth Duncan and Lisa Feldman
Barrett (2007) point out, since Plato and Aristotle, thoughts and emotions
have been viewed as ontologically distinct, yet ‘[a]ny thought or action can
be said to be more or less affectively infused, so that there is no ontological
distinction between, say, affective and non-affective behaviours, or between

“hot” and “cold” cognitions’ (1202).
Perceived as indicative of a crisis in the spectator-film relation, the com-
plex narrative does not merely demonstrate that the traditional suspension
of disbelief or the classical spectator position as voyeur, witness, or observer
is no longer deemed compelling or challenging enough (Elsaesser 2008, 16),
it also demonstrates that our involvement and construction of a narrative is
as affective as it is cognitive. In this fashion, contemporary complex cinema
calls for a reconceptualization of the core analytical and narratological
concept of the fabula and thus I propose an expanded notion of this term:
embodied fabula. In order to develop this concept of embodied fabula, I
will turn to embodied cognition, complexity theory, cognitive and affective
neuroscience, and Deleuzian film-philosophy. While the contour of this
concept, as it is formulated here, is based on an ongoing dialogue with
complex narratives, the embodied fabula is not restricted to this particular
kind of cinema. Instead, the embodied fabula is here envisioned as a ‘proces-
sual’ concept whose lines are dynamic and subject to change (cf. Mullarkey
2009, xiv-xvi).5
In this book, complex narratives are not distinguished by their intrinsic
narrative complexity (cinema is by definition a complex phenomenon),
but by virtue of their ability to induce a rethinking of elements that have
commonly been thought of as separate in the tradition of classical science.
In doing so, these films call for a film-philosophical excavation designed
to render visible and to distinguish various modes of cinematic complexity
– whether classical, modern(ist), or contemporary. In this context, cinema
becomes philosophical insofar as the experience it gives rise to can be
described as a form of philosophical thinking in action (Mulhall 2008, 4).
However, my key concern is to argue that the kind of thinking involved in
the cinematic experience is ill-conceived from the monolithic perspective
of the analytically, cognitively, and temporally detached spectator, whose
thinking consists in organizing the cinematic material into a unified, linear,
and coherent story (e.g. Bordwell 1985a; Branigan 1992; Carroll 1996). Instead,
it shall be argued that complex cinema facilitates a reconceptualization
of the cinematic experience as embodied thinking in action, from which
film-philosophical excavations can examine how the cinematic experience
challenges the boundaries of our dominant conceptual frameworks and
traditional patterns of thought.
From a narratological point of view, the use of the word ‘complex’ can be
traced back to Aristotle’s differentiations between ‘simple’ and ‘complex’
plots. For Aristotle, simple plots are mimetic because they arrange events
Introduc tion 11
into a single, continuous action with a clearly defined beginning, middle,

and end. From such a perspective, complexity arises from an interweaving
of two causal lines into a single, unified plot line.6 Warren Buckland (2009a)
has questioned the aptitude of this mode of comprehending cinematic com-
plexity in relation to complex narratives. Against the cognitive-formalist
film scholar David Bordwell (2006), Buckland (2009a) argues that the con-
temporary complex narrative ‘is intricate in the sense that the arrangement
of events is not just complex, but complicated and perplexing; the events
are not simply interwoven, but entangled’ (3, emphasis in original).
For these reasons, complex narration cannot be reduced to the linear
trajectory of classical cinema – something Buckland criticizes Bordwell for
doing. However, Buckland commits to a widespread Bordwellian notion,
when he argues that contemporary complex narration ‘emphasizes the
complex telling (plot, narration) of a simple or complex story (narrative)’ (6).
Thus, although Buckland is critical of Bordwell’s cautious stance about the
novelty of complex narratives (cf. Bordwell 2006), the overall approach of
the anthology Puzzle Films is representative of much theoretical work on
complex narration, since it reinvigorates Bordwell’s (1985a) cognitive and
analytical distinction between syuzhet and fabula (Buckland 2009a, 7).
In the following, I set out to challenge this popular analytical tool for
examining cinematic complexity. While the distinction of the fabula (~
story) from the syuzhet (~ plot) can be extremely useful, its limitations once
confronted with cinematic complexity shall be demonstrated throughout
this book. I maintain that the major problem pertaining to this analytical
distinction relates to its adherence to a series of classical scientific principles
designed for the reduction of complexity. These principles – as traced out
by Edgar Morin in his article ‘Restricted Complexity, General Complex-
ity’ (2007) – are 1) the principle of universal determinism associated with
Laplace; 2) the principle of reduction, which ‘consists in knowing any
composite from only the knowledge of its basic constituting elements’;
and finally 3) the principle of disjunction, which ‘consists in isolating and
separating cognitive difficulties from one another’ (5).7 The cognitive and
classical narratological presumption that complexity can be seen as an
intrinsic value of the narrative itself can be traced back to these classical
principles for the reduction of complexity. In order to counter this pre-
sumption, I aim to demonstrate that complex narratives first and foremost
deserve the designation ‘complex’, because they make evident that ‘[c]
omplexity is not only a feature of the systems we study, it is also a matter of
the way in which we organize our thinking about those systems’ (Tsoukas
& Hatch 2001, 986; cf. Simons 2008, 116). That being said, it is important to
stress that I am not arguing against reduction per se, since complexity and
reduction are necessarily intertwined. I rather claim that the dominant and
systematic mode of reducing complexity is no longer a viable approach, since
it relies on a separation of those elements whose interrelation contemporary
complex cinema sets out to (re-)explore.
Consequently, it can be argued that the term ‘complex narratives’ is
problematic, because it falsely suggests that other types of cinema are not
complex – in fact, it may even appear to imply a general prevalence of
‘non-complex narratives’ within cinema. It is possible to avoid this problem
by taking the notion of complexity a step further than that usually found
in the study of complex systems. This means understanding cinema from
the perspective of what Morin (2007) has labelled ‘generalized’ rather than
‘restricted’ complexity. According to the latter, ‘complexity is restricted to
systems which can be considered complex because empirically they are
presented in a multiplicity of interrelated processes, interdependent and
retroactively associated’ (10). Since this perspective never questions the epis-
temological nature of complexity it ‘still remains within the epistemology
of classical science’ (10). Consequently, restricted complexity acknowledges
the non-linear, relational nature of complex systems, but seeks to tame it in
ways that reintroduce positivism and reductionism, whereby complexity is
ultimately acknowledged only by means of ‘decomplexification’.8
A move towards generalized complexity must thus involve an episte-
mological displacement encouraged by the invention of new conceptual
frameworks that do not seek to redeem complexity into the classical sci-
entific ideals of linearity, neutrality, objectivity, isolation, reduction, and
disjunction. Yet, this should not encourage a simple reversal of the relation
by means of an emphasis on those elements that have traditionally been
excluded, such as non-linearity, complex temporal processes, incom-
mensurable spaces, heterogeneity, and logic unruled by the principle of
non-contradiction (cf. Rodowick 2001, 49). Ultimately, what is required
instead are conceptual tools capable of embracing complexity, such as those
which emerge from the interrelation of the elements that have been kept
separated far too long (cf. Morin 2007).
In Deleuze, Altered States and Film (2007), Anna Powell contends that
experimental cinema does not invite the ‘problem solving’ associated with
cognitive-formalist approaches (cf. Bordwell 1985a, 2006; Branigan 1992;
Carroll 1996; Thompson 1988). Instead, such films ‘aim to derange the senses
and the mind’ (Powell 2007, 8). My goal is to make evident why such a
conception misconstrues the cinematic experience as an option of either-or.
In the 1980s, two very different approaches proposed alternatives to the
Introduc tion 13
psychoanalytical and semiotic studies that dominated film studies at that

time. Despite being united in their dissatisfaction with the manner in which
cinema had been used to confirm the theory applied, one could hardly
imagine two more opposed conceptions of the cinematic experience than
those proposed by cognitive film studies and Deleuzian film-philosophy.
Since then the divide between them has been ever expanding and is today
defining of film studies. At least until recently, when contemporary complex
narratives have begun to encourage scholars to think beyond the linear-
non-linear dichotomy underlying this divide (cf. Brown 2013; Engell 2005;
Fahle 2005; Mullarkey 2009; Pisters 2012).
The ‘linear’ segment of cognitive film science includes scholars such
as David Bordwell, Noël Carroll, Joseph Anderson, Murray Smith, Carl
Plantinga, Greg M. Smith, Ed Tan, and Torben Grodal. In the introduction
to the anthology Post-Theory: Reconstructing Film Studies (1996), Bordwell
and Carroll explain that ‘a cognitivist analysis or explanation seeks to
understand human thought, emotion, and action by appeal to processes of
mental representation, naturalistic processes, and (some sense of) rational
agency’ (xvi). Bordwell’s cognitive theory is particularly representative
of the linear side of the dichotomy, since he maintains that the classical
cinema has become the ‘standard’ film because its conventions of linear
causality are ‘cognitively optimal’.9 The popularity of classical cinema thus
lies in its natural correspondence with the systematic manner in which
human beings make inferences, test hypotheses, and apply interpretative
schemata in their everyday lives.
In his Henri Bergson-inspired film-philosophy, the French philosopher
Gilles Deleuze (2005a, 2005b) provides a framework that accentuates
the linear-non-linear dichotomy since it proposes two overarching im-
age regimes that roughly correspond to the opposition of classical and
(European) art cinema: the ‘movement-image’ (and, more precisely, the
‘action-image’) and the ‘time-image’.10 In the first, the narrative universe
is unified and its events are linked through ‘rational’ cuts and a style of
montage that present time in an ‘indirect’ manner according to a logic that
accommodates our everyday sensory-motor capacities. In the second image
regime, time is freed from its sensory-motor linkages, while the cuts have
become ‘irrational’, from which a ‘direct’ image of time appears. Unlike the
more classical narratological, film-theoretical, and formalist concerns of
cognitive film science, Deleuze’s interest in cinema should be understood
in relation to his philosophical undertakings, insofar as cinema grants
philosophy a renewed mode of thinking about movement and time. Thus, it
is important to keep in mind that Deleuzian film-philosophy and the more
traditional film-theoretically founded cognitive formalism are interested

in cinema for different reasons.
Perhaps exactly such lack of sensitivity towards their respective research
interests has contributed to the tension between the ‘Continental’ film
philosophers and the ‘Anglo-American analytical’ cognitive film science.
This tension is detectable in Raymond Bellour’s (2010) telling dismissal of
the cognitive stance:
There is always the fear that the film and the spectator are all the more
average, standardised, attuned to the dominant cinema, that one wants
to address a supposed truth of the film and its spectator in a sort of
monstrous, targeted freeze-frame. This is why, in their dogmatic applica-
tion of knowledge of the cognitive sciences, most cognitive theoreticians
of the cinema are, for example, inevitably attracted by Steven Spielberg’s
films and Hollywood blockbusters. (92)
Despite the reservations between cognitivists and Deleuzian film phi-

losophers, I argue that contemporary complex cinema can be perceived
as an encouragement to reconnect the linear, cognitive, and analytical
approaches of cognitive formalists with the affective, non-linear, and non-
representational attitude that defines Deleuzian film-philosophy. However,
this is not a straightforward task due to the discrepancies between the repre-
sentationalist, realist, and classical scientific presumptions of (Bordwellian)
cognitive film science and Deleuze’s anti-Cartesian, anti-representational
film-philosophy. My thesis, however, is that the incongruities begin to van-
ish once the computational assumptions that guide the frameworks of most
cognitive theories are replaced with those of embodied cognition. Not only
have the computational assumptions started to reveal their philosophical
or theoretical limitations (such as embodied cognition maintains), the rise
of contemporary complex cinema has rendered visible how these restrict
our comprehension of the cinematic experience, too.
Cognitive media scholars would immediately object and, rightly, argue
that hardly anyone (if anyone at all) in their field has explicitly promoted a
computational understanding of mind. When computationalism is debated,
cognitive scholars point out that their field has ‘followed cognitive sci-
ence’s gradual move from a focus on “cold” cognition (information-driven
mental processes described in terms of inferential and computational
models) to “hot cognition” (affect-driven mental processes)’ (Nannicelli &
Taberham 2014, 5). Jovially referring to himself as a part-time cognitivist,
Bordwell (2010) declares that ‘you don’t have to be a cognitivist 24/7’ (15).
Introduc tion 15
Thus, we should not expect the computational assumptions of cognitive

film theory to be explicitly stated. Instead, these assumptions, as I will
demonstrate, are expressed in analytical devices and interpretative strate-
gies that have their origin in, but are not restricted to, cognitive film theory.
Consequently, it is incisive to allow the films themselves to take an
active part in the reconfiguration of our analytical devices and interpreta-
tive strategies. This inductive approach of allowing the films to shape our
understanding of the cinematic experience guides all the examinations of
individual films to be found in this book. In an extension of this, I aim to
demonstrate that once we move beyond the linear-non-linear dichotomy
and start to think of the cinematic experience in terms of the complex
interplay between linear and non-linear elements, a field emerges from
which complexity theory, cognitive film science, Deleuzian philosophy, and
‘embodied cognition’ can be combined in a joint effort to reconceptualize
the cinematic experience as a genuinely cognitive-embodied experience.
However, the danger involved in this is to reintroduce an inverse dualism
that favours the body over the mind, or, as Brown (2013) asserts in relation
to the pioneer embodied-phenomenological work of Sobchack, Marks, and
Barker, the challenge of today is to ‘synthesize with the haptic, or affective,
elements of the cinematic experience the “higher” “brain” elements that in
fact form a continuum with them’ (141).
Therefore, to unite Bordwellian and Deleuzian ideas on the same con-
ceptual plane, I draw upon the recent developments within the cognitive
neurosciences, where certain scholars are coming up with an increasingly
embodied understanding of cognition (cf. Varela, Thompson, & Rosch 1992;
Clark 1998; Noë 2004; Wheeler 2005; Shapiro 2011). Given that these schol-
ars understand mind, thinking, and cognition as genuinely embodied
mental processes, this appears to be a particularly suited framework for
comprehending the reconstitution of the viewer’s affective, emotional, and
cognitive bonds in the complex narrative. Ultimately, I argue that complex
narratives question the classical narratological understanding of the fabula
by 1) embracing a non-representational and non-computational mode of
spectatorship 2) whose temporality may contain instances of, yet is not
predisposed to, causal linearity.
The cinematic capacity to stimulate viewers in a direct, corporal-affective,
and sensorial fashion has recently been the subject of growing attention
(cf. Barker 2009; Marks 2000, 2002; Shaviro 1993; Sobchack 1992, 2004).
Unfortunately, this body of work has been conducted in relative isolation
from studies that examine the narrative powers of cinema (cf. Bordwell
1985, 2007; Branigan 1992; Smith 1995). Studies of the experiential and the
narrative cinematic domains have thus up to now coexisted peacefully

as two distinct cinematic dimensions best studied apart from each other.
At the back cover of Vivian Sobchack’s influential Carnal Thoughts (2004)
the book’s intention is declared to emphasize our corporal rather than
our intellection stimulation with cinema. In The Skin of Film (2000), Laura
Marks explains that ‘haptic media encourage a relation to the screen itself
before the point at which the viewer is pulled into the figures of the image
and the exhortation of the narrative’ (187-188). More recently, the concept
of ‘affect’ has served a similar function as not only different in kind to
cognitive responses but radically isolated from such processes insofar
as affects are perceived as immediate and bodily autonomous responses
to the images thus detached from their representational and narrative
dimensions (cf. Clough & Halley 2007; Gregg & Seigworth 2010). A central
argument made by this book, however, is that embodiment does not simply
occur beneath or below cinematic narration; it actively demands a new
interpretation of cinematic narration. No longer understood as a mental
representation, the cinematic narrative must be perceived as an embodied
activity that emerges out of the assemblage of spectator and film.
The growing popularity of embodiment has rendered evident the limita-
tions of the fabula as a theoretical idealization that focuses exclusively on
the cognitive aspects of narrative construction (cf. Bordwell 1990, 108).
Yet, this key narratological concept remains to be revised according to the
enactive, emotional, affective, and embodied understanding of cinematic
spectatorship now prevalent (cf. Tikka 2008). Hence, I propose a conceptual
differentiation between the ‘analytical’ and the ‘embodied’ fabula. The
cognitive-formalist fabula thus pertains as the analytical fabula according
to which our cinematic perception is structured towards the construction
of a causal-linear story (cf. Bordwell 1985a). The embodied fabula, on the
other hand, is designed to open up for an exploration of the narrative as
our surrounding environment. Since the embodied fabula is a complex and
recursive concept involving constant feedback loops, it is not diametrically
opposed to the analytical fabula, which is rather to be understood as a
prominent dimension of the embodied fabula – i.e. the analytical fabula is
also an embodiment of the cinematic narration, but in a particular analyti-
cal manner.
In relation to contemporary cinema, the attention will be focused on four
characteristics that in combination have facilitated the mode of compre-
hending cinematic complexity that I propose here. Contemporary complex
cinema allows us to rethink and reconf igure 1) the linear-non-linear
dichotomy of film studies that harks back to the opposition of classical
Introduc tion 17
cinema and the tradition of (European) art cinema. In moving beyond this
divide, complex narratives 2) reveal a profound will to complexity, since
they force us to think about the interrelation of what has traditionally been
kept isolated, such as the linear and non-linear; the affective, emotional,
and cognitive investment of the audiences; the contingent from the causally
determined; the body from the mind; etc. It will be argued that contempo-
rary complex cinema 3) reconfigures our mode of experiencing narration as
the process of organizing events into a causal-linear order, i.e. according to
the analytical fabula. More precisely, these films demonstrate 4) that linear
cinematic perception (as cognitively structured around the construction of
a causal-linear story) coexists with other modes of ‘inhabiting’ the narrative
universe. It is in the context of capturing these dimensions of contemporary
complex cinema that the differentiation between the analytical and the
embodied fabula is suggested.11
I believe that the ability of contemporary complex narratives to ‘enfold’
or ‘embed’ us in their narrative universes has been a decisive factor for the
formative role that recent film-philosophical projects trying to rethink
the cinematic experience in the age of new media have granted this type
of cinema (e.g. Pisters 2012; Shaviro 2010; Bianco 2004; Rodowick 2007;
Brown 2013). Yet, throughout this book, I accentuate the importance of
refining our conceptual frameworks through a constant dialogue with
the challenges that arise from our encounters with various – i.e. not solely
contemporary – complex forms of cinema. Rather than primarily asking
what the films in question ‘mean’ or are ‘about’, I question the prevalent
string of arguments that have traditionally been invoked to make sense
of the moving images. In doing so, my focal points are the encourage-
ments, obstacles, resistances, or ruptures that the films in question exhibit
towards particular dominant modes of organizing and comprehending
experience.
Consequently, the analytical material chosen for this study belongs to
the sphere of well-known and much-debated examples. The advantages
and disadvantages of this choice are obvious. The main disadvantage is, of
course, that using established examples remains oblivious to new emerging
trends and experimental approaches to filmmaking that indeed deserve
more critical attention (I have chosen somewhat lesser known examples
from acclaimed directors such as Alfred Hitchcock and Alain Resnais).
Nevertheless, the selection of frequently studied examples has been a re-
quirement for me to perform a series of meta-analytical readings of the films
in question. Such readings help demonstrate how differences in theoretical
and philosophical assumptions shape our actual analytical procedures.
The reader should be aware that the concept of the embodied fabula
in its present form reflects the commitment of this book to known cin-
ematic examples and the foregrounding of contemporary complex cinema.
Nevertheless, I believe that the embodied fabula could prove useful for
comprehending a variety of changes currently occurring within the field
of cinema, which are not explicitly related to the complex narrative. The
concept could, for instance, be utilized to understand the film-philosophical
encounter facilitated by the experimental documentary film Leviathan
(Castaing-Taylor & Paravel 2012) or to conceptualize how 3D technology
has been implemented in Gravity (Cuarón 2013) to allow for an entirely
different embodied experience than traditional 2D cinema is capable of (I
explore the idea of expanding the notion of the embodied fabula in relation
to this cinematic ‘will to immersion’ in Chapter VIII). With such examples
the notion of the embodied fabula could be expanded beyond the scope
of the analytical material that I have selected. In this sense, each film-
philosophical encounter bears the promise of uncovering dimensions of
the embodied fabula that the encounters of this study have not brought to
light. By the same token, I believe that it is possible to use this concept to
comprehend aspects of the cinematic experience – the political, sociologi-
cal, national, economic, technological, etc. – that have not been my focal
points.12
That being said, I firmly believe that my examples demonstrate a simulta-
neously interesting and highly relevant contemporary embrace of cinematic
complexity. In this context, the films’ popularity has made it easier to study
the recursive nature of complexity that emerges once the films and the
prevalent conceptual tools, frameworks, and analytical assumptions of the
analyst and/or the spectator are allowed to mutually reflect back on one
another. In particular, the films I have chosen to study have been crucial to
the development of the concept of the embodied fabula and to the ongoing
reformulation of the cinematic experience that this concept entails. In this
manner, I have endeavoured to retain the open nature of the films to allow
them to inform us about the manner in which we participate in structuring
experience according to our prevalent metaphors, conceptual tools, and
along the lines of how we ‘normally’ structure perception – and thus to
study cinema between the lines.
This has required a careful selection of films that enable the examination
of different aspects of the conceptual tools and interpretative methods
that concern us here. Stage Fright (Hitchcock 1950), which is examined
in Chapter III, represents a classical instance of defamiliarization insofar
as the film upsets the automatic expectations aroused in the audience by
Introduc tion 19
the formal device of the flashback. Memento, the subject of Chapter VII,
departs from a comparable defamiliarization in order to allow us to sense
the habitual processes that usually operate unattended to structure our
cinematic perception of the narrative in causal-linear terms. Yet, Memento
takes defamiliarization a step further to include the very foundation of
what we traditionally think about as constituting the cinematic experience,
i.e. it defamiliarizes the stable ‘background’ against which everything has
traditionally been defamiliarized, thereby forcing us to reconceptualize
the very notion of defamiliarization.
Similarly, 21 Grams and Lola rennt – both f ilms are discussed in
Chapter VI – enable their audiences to explore the virtual, non-linear
dimensions (that, which could have, yet did not happen) of their narratives,
but in entirely different manners. The non-linear and fragmented narrative
structure of 21 Grams breaks down the smooth operation of spectators’ af-
fective, emotional, and cognitive circuits to establish instead a more direct
empathetic bond between characters and spectators. Lola rennt, on the
other hand, uses its multimodality and music-video aesthetics to make spec-
tators bond with the narrative rhythm(s) of the film and the kinaesthetic
of its main character. In Hiroshima mon amour (Resnais 1959), which is
discussed in Chapter IV, a comparable examination of cinema’s virtual
dimension can be detected. However, here this is achieved in a manner
typical of modern(ist) cinema, which is to say that it involves a criticism of
the linear organizational principles associated with classical storytelling.
The structure of this book – in which predominantly theoretical chapters
(Chapter II and V) are followed by chapters devoted to closer examinations
of films, their analytical treatments, and the development and application
of the embodied fabula (Chapter III, IV, VI, and VII) – is different from the
inductive and film-philosophical research process that lies behind it, where
these closely intertwine. I have nonetheless chosen this structure to render
it more visible how our philosophical presumptions influence our actual
analytical procedures, even in cases where the analyst believes to have left
these presumptions behind.
Consequently, this book is comprised of two parts. Chapters II, III, and
IV form the first part of the book in which the linear-non-linear dichotomy
is examined from several perspectives. This part establishes the theoretical
foundations for the argument that the ongoing replacement of computational
with more embodied approaches within the cognitive sciences may prove
an important cornerstone for bridging cognitive and film-philosophical
approaches to cinema (cf. Protevi 2010; Pisters 2012; Brown 2013; Chapter V).
Thus, the main argument found here is that in disputing the computational
and representational roots of cognitive film studies, it becomes possible to

perceive cognitivism and film-philosophy as complementary, rather than
opposing, positions. This would not only open cognitive studies up for the
more non-linear and aesthetic dimensions of cinema, but equally encourage
Deleuzian film-philosophy to connect with the more rigorous, analytical,
and empirical – yet extremely innovative – research, which is currently
being conducted within cognitive film science.
Hence, chapters V, VI, and VII form the second part of the book, which
is dedicated to the exploration of various aspects of the embodied fabula
in relation to contemporary complex cinema. In this part I draw upon
embodied cognition, Deleuzian film-philosophy, and complexity theory to
demonstrate that especially with regard to contemporary complex cinema
the analytical fabula must be supplemented with a concept of the fabula
that has been designed to capture the cinematic experience as embodied
and complex. Therefore, I propose the embodied fabula as an operational
tool that guides the viewers in their enactive and embodied engagement
with the narrative universes they no longer primarily ‘organize’, ‘linearize’,
or ‘straighten out’ but explore.
Chapter II argues that despite its constructivist nature, Bordwell’s theory
of narrative comprehension comes with the analytical presumption that
the narrative – though being the mental construction of the spectator – can
be analytically dissected as an intrinsic quality of the narrative itself. The
problem with this reasoning is that it harbours a misunderstanding about
the complexity of contemporary complex cinema. Following Morin (2007)
‘any system, whatever it might be, is complex by its own nature’ (10), and thus
complex narratives cannot be said to differ from other cinematic regimes
by virtue of their intrinsic complexity. From this perspective, the narratives
in question deserve the adjective ‘complex’ only by virtue of inducing a
transformation of our onto-epistemological conception of complexity – in
particular by facilitating a highly sophisticated interplay of linear and non-
linear cinematic elements. While all films from a generalized perspective
are complex in their own right, not all films call attention to the immanent
complexity of cinema.
In reference to Alfred Hitchcock’s Stage Fright, Chapter III argues
that while classical storytelling has often been associated with cinematic
linearity, this linearity is as much a product of the linear interpretative
methods that have been accepted as standards for dealing with classical
cinema. It is examined how the dominance of goal-oriented characters,
narrative resolutions, and linear spatio-temporal coherence has justified
analytical methods that in their classical scientific cause-effect principles
Introduc tion 21
have produced an overtly linear understanding of classical cinema. It is

especially examined how the cognitive-formalist conceptual framework
perceives the linearity of the classical film to correspond to a ‘natural’ kind
of filmmaking that can operate as a background to which all other regimes
can be understood. It is argued that the cognitive concept of ‘defamiliariza-
tion’ actually performs a ‘refamiliarization’ of narrative transgressions.
Ultimately, this chapter questions the restricted comprehension of the
classical paradigm of both Bordwell and Deleuze, and argues that it stems
from a lack of sensitivity for the inherent ambiguities that reside in the
cause-effect dramaturgy of classical cinema.
Chapter IV examines Hiroshima mon amour to carve out how the
film formulates a cinematic logic that arises from the encounter between
oppositions. As much as the Deleuzian concept of the time-image is capable
of shedding light on this poetic film, so does Hiroshima mon amour also
facilitate an understanding of the conceptual powers and limitations
of the time-image. It is argued that its main limitation rests upon the
methodological choice of separating and contesting cinema’s linear and
non-linear dimensions. While it will be demonstrated that this resonates
with the modern(ist) cinematic ideals, it is maintained that the opposition
of movement-image and time-image is no longer capable of capturing the
complex interplay between these dimensions in contemporary complex
cinema. Hiroshima mon amour becomes especially interesting because
it expresses an at the time entirely new cinematic logic, which can be
formulated with reference to Deleuze’s concept of the encounter. This
logic is opposed to the long-dominant linear manner of understanding
narrative, history, memory, and time, whose desirability is repudiated at
the end of Hiroshima mon amour. It is in this context that the film can
be taken as an acute expression of the modern(ist) paradox of ‘representing
the unrepresentable’, which is still haunting Deleuzian film-philosophy.
Therefore, the basis for understanding the narrative experience as
embodied and the development of the concept of the embodied fabula in
Chapter V is not a mapping of cognitive-formalist and film-philosophical
ideas on to each other. The aim is rather to reach a more comprehensive
understanding of cinema, and thus not to remain ‘true’ to the theories, but to
challenge and enrich their conceptual schemes. Patricia Pisters’s (2012) work
on the ‘neuro-image’ is crucial to this chapter, since she connects neurosci-
ence to film-philosophy and modern screen culture. Equally important is
John Protevi’s (2010) argument that embodied cognition would benefit from
adopting the Deleuzian tripartite ontological differentiation between the
virtual, intensive, and actual. Protevi’s text is visionary due to its insistence
on viewing Deleuzian philosophy as working across – rather than playing

a part in – the so-called ‘Analytic-Continental’ divide of philosophy. This
is essential for my proposal of an alternative account of the fabula that is
not based on the ‘classical sandwich model of perception’ (cf. Hurley 2002).
However, this primarily theoretical examination of the embodied fabula
must be complemented with a study that takes into account the obstacles
stemming from the complex narratives themselves. Chapter VI, therefore,
scrutinizes a variety of applications of the classical distinction between
fabula and syuzhet. I demonstrate that the distinction carries a series of
presumptions to which any uncritical use automatically commits. This is
done in relation to two complex narratives: Tom Tykwer’s ‘forking-path’
narrative Lola rennt (Run Lola Run 1998) and Alejandro González Iñár-
ritu’s ‘mosaic’ narrative 21 Grams (2003). Ultimately, just as the scientific
principles for the reduction of complexity have led to important and brilliant
advancements up to a point, ‘where the limits of intelligibility which they
constituted became more important than their elucidations’ (Morin 2007, 5),
so the rise of complex narratives makes a similar statement about classical
narratology possible. I argue that the problem is that narratologists have
seen it as their task to explain away or straighten out narrative complexity.
As an alternative to this, I draw upon the idea of embodying the fabula to
enhance our understanding of the multimodal and complex cinematic
experience that these films give rise to.
Finally, I examine Christopher Nolan’s Memento (2000) in Chapter VII
and argue that the film has become a site for rethinking the complexity of
the cinematic medium. In particular, this chapter is interested in explor-
ing how the film facilitates a renewed conceptualization of two concepts:
the fabula and defamiliarization. This requires a re-examination of the
complex interplay between the film’s linear and non-linear dimensions,
which I maintain is constituted in a reconfiguration of the feedback loops
of the cognitive, emotional, and affective registers. It is with reference to
Memento’s narrative feedback loops that the logic of the embodied fabula
finds its clearest cinematic expression.
On a final note, I am aware of the implications of referring to digital
cinematic works as ‘films’. However, following Brown (2013), I use the words
‘film’ and ‘cinema’ according to ‘what they can do as opposed to in terms
of what each word means’ (11). Thus, it is the terms ‘film’ and ‘cinema’ that
evolve to accommodate the products and the manner we use these, rather
than the products outstripping the terms (13). Following a similar line of
reasoning, my reflections about the cinematic experience depart from ideal
viewing circumstances, although I acknowledge that films nowadays are
Introduc tion 23
often seen on multiple platforms – such as on a laptop or a smartphone –

that may reduce their impact and ability to enfold or embed the spectator
in their universes. Yet, I am interested in carving out the potential of the
cinematic experience, and the fact that not all viewing circumstances are
optimal for narrative immersion, does not cause this cinematic potential to
disappear altogether (cf. Brown 2013, 9-12; see also Carroll 1996; Rodowick
2007). In answering the Bazinian question of ‘what is cinema?’, Dudley
Andrew (2010) maintains that ‘cinema, essentially nothing in itself, is all
about adaptation, all about what it has been led to become and may, in the
years to come, still become’ (140-141; cf. Brown 2013, 12). This study explores
how moving images are constantly expanding the potential of what the
cinematic experience might become.
II. Cinema in the Interstices
We are torn between a nostalgia for the familiar and an urge for the foreign and
strange. As often as not, we are homesick most for the places we have never known.
‒ McCullers 2012, Chapter ‘Look Homeward, Americans’
Since a paradigm of simplification controls classical science, by imposing a

principle of reduction and a principle of disjunction to any knowledge,
there should be a paradigm of complexity that would impose a principle
of distinction and a principle of conjunction.
‒ Morin 2007, 10
Introduction
One dominant approach to complex narratives has been to focus upon how
these ‘foreground the relationship between the temporality of the story and the
order of its telling’ (Cameron 2008, 1; cf. Buckland 2009b; Feagin 1999; Kinder
2002). Thus, for many scholars the complexity of contemporary cinema consists
in the complex telling of a more or less complex story. However, this chapter
suggests that the ‘complexity’ of complex narratives is better understood in
terms of how the films challenge prevalent conceptual tools and patterns of
sense-making. In particular, this chapter addresses the cognitive conception
of the fabula conceived as a mental representation. From this conception, the
fabula is often perceived to be an intrinsic value of the narrative itself. Against
this widespread presumption, I argue that the complexity of complex narra-
tives is rather derived from the ability of these films to challenge established
beliefs, conceptions, assumptions, and prevalent methodological frameworks.
In this manner, complex narratives make it obvious that the complexity of a
system cannot be determined by disengaging it from the observer.
This chapter explores how complex narratives, by shattering a series
of binaries that have long informed Western rationality, can be seen in
relation to an ongoing reconstruction of the notion of complexity currently
undertaken in a series of scientific disciplines. In particular, I explore how
cinematic complexity in this fashion induces a mode of thinking beyond the
cinematic dichotomies of classical and modern(ist), linear and non-linear,
cognitive film science and Deleuzian film-philosophy, cognition and affect,
etc. In doing so, I combine a film-philosophical approach with the complex-
ity theory of Edgar Morin to understand the interplay and interrelation of
components that in the tradition of classical narratology are thought to be

separate. Therefore, the aim is not to downplay the importance of linearity
or cognition for the cinematic experience, but to insist on understanding
these in their relation to the more emotional, affective, and non-linear
dimensions of cinema.
Cinema and its two Temporal Modes
Mary Ann Doane (2002) has argued that early cinema contributed to the
production of a distinct mode of temporality, since it fuses the linearity of the
medium (the 16-24 frames per second) with a non-linear ability to manipulate
temporal order through montage. For these reasons, cinema played a vital
role in the ‘sea change in thinking about contingency, indexicality, temporal-
ity, and chance [that] deeply marked the epistemologies at the turn of the
last century’ (4). Cinema, in other words, ‘participated in a more general
cultural imperative, the structuring of time and contingency in capitalist
modernity’ (4). The two modes of temporality that Doane believes to be
comprised in early cinema match the temporal modes that have often been
associated with classical and modern(ist) cinema respectively (and hence also
with the Deleuzian opposition of movement/action-image and time-image).
On the one hand, the emergence of cinema is inextricably linked to the
industrialization, rationalization, and standardization of time in modernity.
Doane provides a series of examples for systems designed in the spirit of
modernity to manage temporality. These include the systematization of
time in the forms of the introduction of punchcards to regulate work time
in factories. Furthermore, Doane mentions the attempt to orchestrate and
render the physical movements of the workers more efficient (known as
‘Taylorization’), the introduction of railroad timetables that no longer allow
for regional time differences, and the 1844 conference in Washington D.C.,
which established the exact length of a day, divided the world into 24 time
zones, and agreed upon Greenwich as the zero meridian (5). All these serve as
examples of how modernity constructed a new perception of temporality –
symbolized by an explosion in the popularity of watches – that was driven by
a desire to externalize time and render it measurable, consultable, and ‘ready
at hand’. In this manner, the linear temporal mode is a product of capitalist
modernity, according to which time was ‘increasingly reified, standardized,
stabilized, and rationalized’ (5). In the words of Doane, the ultimate goal of
the linear temporal modality is to ‘eliminate unproductive time from the
system’ (6). In cinema the system of continuity editing associated with the
Cinema in the Interstices 27
storytelling of classical Hollywood serves a similar function; insofar as it,

too, eliminates, as discretely as possible, those chunks of time that do not
contribute to the overall (narrative) progression of the film.
However, cinema also incorporates a contrasting mode of perceiving
temporality connected to the ability of time to evade rationality, linearity,
and systematization. Doane points to the work of Alfred North Whitehead, Ed-
mund Husserl, Henri Bergson, and Martin Heidegger, along with writers such
as James Joyce and Virginia Woolf as prominent and well-known examples of
how time was increasingly being conceptualized in terms of chance, indeter-
minacy, and as a viable and subjective flow. In particular, Doane focuses on the
work of Bergson, who at the turn of the 20th century was discontented with
the manner in which time had increasingly become ‘uniform, homogenous,
irreversible, and divisible into verifiable units’ (6). His philosophy of time,
which circles around the concept of durée (duration or ‘lived’ time) can be
seen as an ‘adamant reassertion of temporal continuity’ (7). This concept
is designed to capture the psychological dimension of temporality as being
qualitative, heterogeneous, dynamic, and thus free from the determinism of
classical mechanistic science. Bergson’s conception of time, as András Bálint
Kovács (2000) observes, has had a huge impact on modern(ist) cinema: ‘A major
tendency in modern cinema is to blur the boundary between fact and fancy,
dream and reality. Modern cinema conceives of, and realizes (“virtualizes”),
time in a totally Bergsonian sense: that is, in absolutely subjective terms’ (161).
Doane’s study of early cinema is distinguished from traditional accounts
of ‘primitive’ cinema, since it gives equal weight to both the linear and non-
linear dimensions of cinema.1 Early cinema, following Doane, incorporated
a temporal anxiety that was later to be domesticated by the predominantly
linear system of narrative (Hollywood) cinema. In the age of digital cinema,
scholars have been forced to renegotiate the temporality of the cinematic
apparatus and its claim for indexicality (cf. Rodowick 2007). However,
cinema in its digital form can still transmit the original anxieties ascribed
to the medium, which Doane explicitly connects to the camera’s non-human
and mechanical mode of capturing whatever is placed in front of it, from
which an excess of unpredictable chance events were bound to emerge.2
According to Doane (2002), narrative (Hollywood) cinema plays with
the anxieties of the unpredictable and the contingent, too. Narrative cin-
ema is, however, most often directed towards a recuperation of the linear
logic, which underlines its desire to simulate natural perception. Classical
cinema ‘acknowledges the force of contingency and mobilizes chance, but
ultimately it overrides both’ (138). Linearity is, therefore, not primarily to be
connected to the relentless forward movement of the cinematic medium,
but resides as much in the continuity system of editing. ‘In the classical
cinema,’ as Doane argues, ‘the cut aborts the problem of an excess of the
random, of chance in time’ (137). If continuity editing can rightly be said
to ensure a recuperation of linearity, as Doane suggests, what is the role of
editing in contemporary complex cinema?
In his recent work on ‘post-continuity’ cinema, Steven Shaviro (2010,
2012) has argued that contemporary cinema no longer obeys the dichoto-
mies that followed from the opposition between classical and modern(ist)
cinema. The term ‘post-continuity’ is an offshoot of Bordwell’s (2002b)
concept of ‘intensified continuity’. Mostly considering Hollywood action
films from the late 1990s, Bordwell argues that although certain changes
can be seen ‘in representing space, time, and narrative relations (such as
causal connections and parallels), today’s films generally adhere to the
principles of classical filmmaking’ (16). It is worth noting that ‘continuity’
for Bordwell and Shaviro not only refers to the system of continuity editing,
but also more broadly designates cinematic linearity and thus ‘implies the
homogeneity of space and time, and the coherent organization of narrative’
(Shaviro 2012, par. 17). This is significant for understanding the underlying
claim of Bordwell (2002b), when he states that intensified continuity, ‘far
from rejecting traditional continuity in the name of fragmentation and
incoherence,’ is rather concerned with an ‘intensification of established
techniques’ (16). As such, the Hollywood action films that Bordwell refers to
intensify the cinematic principles of linearity rather than dispose of them.
While not necessarily disagreeing with Bordwell, Shaviro refers to
the work of directors such as Tony Scott and Michael Bay to argue that
intensified continuity is gradually being replaced by ‘post-continuity’ in
contemporary (action) cinema. In contrast to the cinema of continuity,
‘post-continuity’ cinema – and here I would like to add the complex narra-
tive – contains a dispersed rather than an organized or unified narrative.
Other features are a non-Euclidian instead of a Euclidean space-time; that
the role of continuity is incidental rather than essential; that the films occur
in a ‘space of flows’ rather than a space conceived as a rigid container; that
time is no longer primarily linear; that linear causality has been replaced by
structural multicausality; that these films denaturalize perception instead
of simulating natural perception.3 Although certain of these characteristics
reveal the kinship between post-continuity and the artistic experiments
of the time-image, the cinema of post-continuity nevertheless differs from
modern(ist) cinema because it does not accentuate its transgression of
cinematic linearity. Modern(ist) cinema developed its trademark stylistic
devices – such as jump cuts and directional mismatches – in order to violate
the continuity rules and to rupture the sensory-motor linkages that accom-
modated the cinematic image to human perception (Deleuze 2005a, 2005b).
One of the reasons for doing so was to transmit scepticism towards the linear
onto-epistemology, whose foundation – the belief in human reason, scientific
objectivity, and progression – was crackling after the Second World War.
In the cinema of post-continuity the rupture so central to the Deleuzian
time-image is no longer granted a decisive role. In fact, ‘post-continuity’
cinema puts the whole opposition of movement-image and time-image into
question since ‘neither the use of continuity rules nor their violation is at the
center of the audience’s experience any longer’ (Shaviro 2012, par. 20). Instead,
‘continuity has ceased to be important – or at least has ceased to be as impor-
tant as it used to be’ (par. 19, emphasis in original). In this book, I expand on
Shaviro’s thesis to argue that classical narratological concepts have started to
reveal their limitations in the dissolution of the linear-non-linear dichotomy
in post-continuity cinema. In the following, it will be argued that since it
has been increasingly confronted with cinematic complexity the cognitive
concept of the fabula has started to expose its limitations.
The Cognitive Fabula and Linear Storytelling
In Narration in the Fiction Film (1985a), Bordwell analyses the processes

involved in narrative comprehension. Drawing equally on Russian formal-
ism and cognitive psychology, Bordwell articulates a theory of narrative
comprehension based on the manner a given film cues its viewers to make
particular inferences and construct the story in a predefined manner. Alan
Nadel (2005) has argued that the central problem of narrative cinema rests
in the task of – by the course of a limited duration of time – transforming
a two-dimensional space into a temporally limitless, three-dimensional
world. For Nadel this means that
the task of narrative cinema, in other words, is to naturalize a counterin-

tuitive experience by creating the illusion that the viewer has acquired a
privileged window on reality, a window through which one is supposed
to see not objects and actions, but see instead a story (427).
Bordwell uses the terminology of Russian formalism to describe how specta-

tors from the collection of cinematic cues, i.e. the syuzhet, construct the
fabula. However, as Nadel observes, the repeated and systematic use of vari-
ous cinematic devices (characters, framings, camera movements, deadlines,
flashbacks, etc.) does not only cue the spectator to construct a specific
fabula, but also ‘to accept a specific mode of cueing as “natural”’ (430).
In relation to this, we may ask what the conceptual role of the fabula is. In
the cognitive formalism of Bordwell, the fabula is a subjective perception of
the narrative (a cognitive representation) as well as an intersubjective and
coherent entity produced by means of logical inferences and computational
symbol manipulation. The fabula can only be inferred from the syuzhet,
and it thus becomes an entirely abstract entity that nevertheless dominates
our perception of the syuzhet. The syuzhet is constantly being (re)inter-
preted in a manner that accommodates the construction of a coherent and
causal-linear fabula. Not only is the fabula in this conception predisposed to
linearity, but it also participates in an ongoing ‘linearization’ of the syuzhet.
Furthermore, since the causal-linear conception of temporality pervades the
widespread classical narrative conventions, the latter have long contributed
to an automatization of cinematic linearity. In relation to this, Bordwell
(1985a) has argued that the essential principles of classical narration have
grown so dominant that they can now be considered ‘normal’. ‘As a nar-
rational mode,’ Bordwell writes, ‘classicism clearly corresponds to the idea of
an “ordinary film” in most cinema-consuming countries of the world’ (166).
A central concern of cognitive formalism lies in explaining why classical
Hollywood style has proven to be the most persistent one. For precognitive
film studies these questions would be answered with reference to psychoa-
nalysis, philosophy, ideology, history, cultural, or gender issues. Cognitive
formalism, however, turns away from such explanations in favour of more
‘naturalistic’ ones. Bordwell (1997) argues that certain stylistic factors are
‘cross-cultural, trading on the biological or psychological or social factors
shared among filmmakers and their audiences’ (269). For Bordwell this is
of course not to say that cultural and other factors do not play a significant
role in the formation of cinematic conventions, but merely that biological,
psychological, and social factors have traditionally been overlooked in
favour of ideological and psychoanalytical ‘subject theories’, which he
together with Noël Carroll has mockingly termed ‘SLAB-theories’. 4 Well
aware of the difficulties in determining absolute cross-cultural universals,
Bordwell (1996b) proposes a middle way he calls ‘contingent universals’.
These are
contingent because they did not, for any metaphysical reasons, have to
be the way they are; and they are universals insofar as we can find them
to be widely present in human societies. They consist of practices and
propensities which arise in and through human activities (91).
In order to explain the prevalence of classical cinematic norms, Bordwell

(1985a) emphasizes the manner in which these grant the spectator a spatio-
temporal point of reference. He writes:
These three factors [use of informative technique, spatio-temporal

coherence, and stylistically stable devices] go some way to explaining
why the classical Hollywood style passes relatively unnoticed. Each film
will recombine familiar devices within fairly predictable patterns and
according to the demands of the syuzhet. The spectator will almost never
be at a loss to grasp a stylistic feature because he or she is oriented in
time and space and because stylistic figures will be interpretable in the
light of a paradigm (164).
What is of interest here is the particular manner in which the classical film
cues the spectator to construct spatio-temporal coherence in agreement
with a linear onto-epistemology. In their clear separation of ‘nature’ and
‘culture’, ‘naturalistic’ explanations obscure the feedback loops according
to which the ‘naturalness’ of Hollywood continuity and the broader manner
in which humans organize their experiences according to a causal-linear
onto-epistemology constantly reaffirm each other. As Nadel (2005) explains,
[t]he historical conditions that created the conventions of Hollywood-

style film and the historical conditions under which audiences were
instructed in their codes disappear beneath the audience’s transparent
acquiescence to them. In this way, the codes of cinematic representation
become performative. By reinforcing the norms of cinematic reality, they
teach the audience how to read a film as if the process were natural, not
just second nature (429).5
Cognitive formalism is founded on a similar linear onto-epistemology that

makes the Hollywood style appear not only ‘normal’ but also ‘natural’.
Before elaborating on this claim, it is necessary first of all to examine the
onto-epistemological roots of classical cognitive science. The claim here
is not that all film-theoretical work, which has been designated under the
umbrella term of ‘cognitive film science’, adheres to these assumptions.
‘To restrict cognitive film theory to theory rooted in cognitive science,’
as Plantinga (2002) has argued, ‘would clearly be far too narrow, and also
plainly inaccurate’ (21).
Nevertheless, a series of the core assumptions found in cognitive science
still influence the film-theoretical cognitive branches; this is the case even
in works that understand themselves as operating from within an ecological

or embodied – rather than classical – cognitive framework (cf. Chapter V).
Furthermore, even if ‘cognitivists have mostly left the computer analogy
behind,’ their understanding of narrative comprehension and perception
is nonetheless using classical cognitive ‘models of rationality and practical
problem-solving’ (21). In this case, Bordwell’s narrative approach owes a
lot to the core assumptions of the cognitive sciences, since his writings on
schemata, inferences, hypotheses testing, and assumptions that underline
narrative comprehension assume ‘a spectator engaging in goal-directed,
primarily non-conscious procedures to make sense of film narratives’ (21).
In Howard Gardner’s enlightening overview of the history and the future
challenges of cognitive science, The Mind’s New Science (1985), the author
singles out five features prominent of the field. Whereas the first two belong
to the core assumptions of the field, the latter three are methodological
or strategic choices. These are: 1) an insistence on mental representations;
2) the computer analogy; 3) de-emphasis on affect, context, culture, and
history; 4) an interdisciplinary approach; and finally 5) a rootedness in
classical philosophical problems. In the following, these serve as a starting
point for a brief examination of the main characteristics of the classical
cognitive approach.6
Cognitive science is, as Gardner remarks, ‘predicated on the belief that
it is legitimate [...] to posit a level of analysis which can be called “the level
of representation”’ (38). Representations are found between the incoming
data inputs (e.g. audiovisual sense impressions) and the output (e.g. the
audiovisual perception). It is, in other words, the level of the cognitive infor-
mation processing, which is to be understood in terms of ‘symbols, schemas,
images, ideas, and other forms of mental representation’ (39). It is crucial to
understand that ‘representation’ here is used in a highly technical manner,
whose philosophical roots date back to Plato’s concept of ‘Ideas’ (cf. Wheeler
2005, 6). Thus, a researcher may talk about plans, images, representations,
intentions, and beliefs as a manner of describing how people interact with
the world, without necessarily being devoted to the philosophical convic-
tion that such representations truly exist inside peoples’ minds. In the
technical and philosophical sense, the representational theory of mind is
‘the view according to which mental states are, for the most part, conceived
as inner representational states’ (Wheeler 2005, 6). According to this view,
there exist entities, structures, ideas (i.e. representations) that stand in for,
or act as an inner copy of, the external world.
In 1985, the time of the publication of Gardner’s The Mind’s New Sci-
ence, the connection between cognitive science and a commitment to the
representational understanding were basically inseparable insofar as ‘all

cognitive scientists accept the truism that mental processes are ultimately
represented in the central nervous system’ (40). However, as Cummins
(1991, 1996) has noted, the question of the circumstances under which it
is even appropriate to engage in representational explanations remained
underexplored at that time (cf. Wheeler 2005, 6-7). In relation to this, it
is worth noting that despite the growing popularity of more embodied
stances to cognition and neuroscience (to which we shall return in Chap-
ter V) the majority of empirically working scientists remain devoted to a
representational understanding of mind. How, then, do they explain the
processes whereby physical data inputs are transformed into nonmaterial
representations?
In answering this question, the technical invention of the computer
suddenly came to be seen as a living proof that a lump of the physical world
can build and process representations in systematically and semantically
coherent ways (Fodor 1983; Wheeler 2005, 7). The reasoning behind the
computer analogy is that if a man-made machine can be said to reason, have
goals, revise its behaviour, transform information, and the like, it would
seem justified to argue that humans can be characterized in the same way
(Gardner 1985, 40). Thus, the framework in which the mind can be seen
as the software implemented into the neural ‘hardware’ that operates by
manipulating symbolic data processing, is the computational answer to the
problem of representations (Marr 2010; Fodor 1983; cf. Cummins 1991, 13).
This analogy has proven to be extremely fruitful, however, paradoxically,
mostly in providing rich data for why human cognition does not resemble
computers. Gardner (1985) identifies this as the ‘computational paradox’,
which suggests that the portrait of human cognition emerging from the
experiments of cognitive science has turned out to be far removed from
‘the orderly, precise, step-by-step image that dominated the thinking of the
founders of the field’ (386). Nevertheless, the invocation of the computer
metaphor has led to a focus on cognition as logical problem solving (cf. New-
ell & Simon 1972), or, alternatively, on universal structural rules (cf. the
‘universal grammar’ of Noam Chomsky [1965]). Yet, cognitive studies have
paradoxically triumphed in proving that human thinking is ‘messy, intui-
tive, subject to subjective representations – not as pure and immaculate
calculation’ (Gardner 1985, 386). Cognitive scientists have, consequently,
been criticized for their methodological approaches, which rest on bracket-
ing out the role of the affective, cultural, historical, and contextual to focus
exclusively on cognition. It is in this context that the embodied alternative
to cognition can be seen as an assembly of different approaches united by
the search for new answers, since they do what Gardner called for back
in 1985; namely, ‘put their noses to the grindstone and incorporate [the
affective, cultural, historical, and contextual] dimensions fully into their
models of thought and behavior’ (42).
In his survey of classical cognitive science, Gardner explains that while
cognitivists are not hostile towards affective, cultural, contextual, or
historical studies per se, they do ‘attempt to factor out these elements to
the maximum extent possible’ (41). Considering the mind in computational
terms allows the analyst to filter out these ‘disturbing’ elements, so that
cognition can be studied entirely in terms of its data processing. From a cul-
tural historical perspective, however, it is noteworthy that it is the computer
analogy – as an entirely cultural and historical phenomenon – that justifies
such an isolation of cognitive processes from their environment. Critics, as
Gardner remarks, have therefore not surprisingly destined the computer to
be nothing but the latest example of a long series of technologies, which have
been regarded to model human cognition. These critics see no indication
why the computer analogy should prove more adequate than predecessors
such as clockwork mechanics, the switchboard, the hydraulic pump, or the
hologram (40).
The cognitive sciences refer to several fields that are connected to one
another in terms of research interests. The classical fields of cognitive
science are philosophy, psychology, linguistics, artif icial intelligence,
anthropology, and neuroscience among which the interdisciplinary ties
are stronger between some fields and weaker between others (cf. Fig. 1;
Gardner 1985, 37).
As mentioned earlier, the computer provided an example of a purely
physical machine whose syntax-following properties made it possible to
solve any well-specified problem. Alongside the earlier works on logics
and formal systems, this amounted to the emergence of ‘a new level of
analysis, independent of physics yet mechanistic in spirit [...] a science of
structure and function divorced from material substance’ (Pylyshyn 1986,
68; cf. Clark 2000). As Wheeler has argued, the tools of classical AI are
potent when the mission is to explain logic-based reasoning or problem
solving in highly structured ‘search spaces’. Yet, the problematic fact is that
this invokes a kind of processing, which machines are known to perform
exceptionally well, while humans in comparison are much more error-
prone.7 Contrarily, humans – unlike machines – perform rather well when
it comes to generalizing novel cases on the basis of past experiences, and in
reasoning successfully, when provided with corruptive or incomplete data
(Wheeler 2005, 9). One of the earliest opponents to the computer model of
Figure 1. Cognitive Science the First Decades (Gardner 1935, 37)
mind was Rodney Brooks, who took an actionist and embodied – rather
than computational – approach to robotics. Brooks (1999) argued that
‘[r]epresentation is the wrong unit of abstraction in building the bulkiest
parts of intelligent systems’, and famously proclaimed that it is better ‘to
use the world as its own model’ (79).
It can be argued that the separation of mind and body (and hardwired
brain) entailed in the computer metaphor for a long time allowed the cogni-
tive sciences to proceed relatively uninflected by the advances made within
the neurosciences. This view was upheld by the assumption that cognitive
phenomena can be accounted for locally, in other words, that cognition is
what happens between sensory input and behavioural output (cf. Chapter V
on the ‘classical sandwich’). In objection to this view, the neuroscientist
Antonio Damasio (1995) has critically traced the philosophical roots of the
computer metaphor back to René Descartes. Damasio has asserted that the
Cartesian dictum cogito ergo sum, when taken literally,
illustrates precisely the opposite of what I believe to be true about the

origins of mind and about the relation between mind and body. It suggests
that thinking, and awareness of thinking, are the real substrates of being.
And since we know that Descartes imagined thinking as an activity quite
separate from the body, it does celebrate the separation of mind, the
‘thinking thing’ (res cogitans), from the nonthinking body, that which
has extension and mechanical parts (res extensa) (248).
In extension, others have argued that neuronal networks – and even non-
neuronal networks such as the immune system (cf. Varela, Coutinho, Dupire,
& Paz 1988) – display cognitive properties, and that symbolic computation,
therefore, should be regarded as merely a narrow, highly specialized form of
cognition, rather than the privileged model of human cognition (cf. Varela
et al. 1992, 103). More generally, a growing body of research within embodied
cognition has started a search for new philosophical sources as means for
a reorientation of cognitive science in the last decades.8
Following from this, cognitivism as an empirical science has often been
criticized for not being aware of its own philosophical presumptions.
Directed towards its film-theoretical branch, which has proclaimed the
end of ‘Theory’, Gregory Flaxman (2000) has voiced his concerns as follows:
Cognitivism may not produce allegories as obvious as those of its theoreti-

cal predecessors, but that should not be taken to mean that it has eluded
totalization – only that it is deeply and deceptively unaware of its own
habitus. What cognitivism calls science and, better yet, common sense
are the accumulation of conventions whose schematization we have yet
to significantly interrogate (49, n. 37).
That being said, the philosophical habitus of cognitive science is not entirely
underexplored. Undertakings of the philosophical roots have been executed
with emphasis on the cognitive revitalization of Descartes, who according
to Gardner (1985) ‘is perhaps the prototypical philosophical antecedent of
cognitive science’ (50). Granted, there still remains a lot of work to be done
concerning the cognitive habitus, yet the Cartesian influence on cognitive
science has been examined and described in a long series of works.9
That classical cognitive science is Cartesian, does not automatically imply
that it also ascribes to the Cartesian substance dualism – a position that,
according to Wheeler (2005), is in fact dismissed by practically all cognitive

scientists (21-55). The Cartesian ties are instead rooted and implemented
by the very methodologies of the field insofar as it ‘is buried away in the
commitments, concepts, and explanatory principles that constitute the
deep assumptions of the field’ (14). This, furthermore, implies that the
Cartesian roots are ‘typically invisible to the external observer and the
majority of working cognitive scientists’ (14). In order to render these visible,
Wheeler carefully outlines eight principles for what he labels the ‘Cartesian
psychology’ of cognitive science. These partly overlap, partly extend and
enrich Gardner’s five features of cognitive science, and as such provide a
rich understanding of the habitus of the field. Furthermore, I believe that
these do not only apply to the ‘orthodox cognitive science’, i.e. the computa-
tional and connectionist approaches with which Wheeler is concerned, but
also – though with certain significant exceptions – to its film theoretical
counterpart. The eight principles that constitute the ‘Cartesian-ness’ of the
cognitive framework are the following (cf. 21-89):
1. The subject-object dichotomy;
2. The principle of representationalism;
3. Human action is the outcome of general-purpose reasoning processes
that i) is context general and ii) works by manipulating and transforming
representations;
4. Human perception is inferential in nature;
5. Perception is disassociated from action;
6. The environment is i) a furnisher of problems for the agent to solve, ii)
a source of information inputs, iii) a stage on which sequences of pre-
planned actions (outputs of the faculty of reason) are simply executed;
7. Intelligent action remains conceptually and theoretically independent
of the agent’s physical embodiment;
8. Psychological explanations are temporally austere since there is no
need to appeal to richly temporal processes.
At the same time as Wheeler opposes the embodied approach to the above
Cartesian principles of ‘orthodox’ cognition, he also provides four paral-
lel claims that support a more positive understanding of the embodied
approach: 1) a primacy of online intelligence; 2) online intelligence is
generated through complex causal interactions in an extended brain-
body-environment system; 3) cognitive science should increase its level of
biological sensitivity; 4) cognitive science should adopt a dynamical systems
perspective.10 These are important principles for the reconceptualizing of
the fabula proposed in Chapter V. Of central importance is that from the
embodied perspective the fabula can no longer be seen as a representation

of the narrative that can be detached from its spectator. I instead maintain
that the fabula should be perceived as an embodied mental tool for structur-
ing the narrative environment. This approach is taken as a measure to 1)
counter the classical principles for the reduction of complexity (cf. Morin
2007), and 2) avoid re-enacting what Tim Ingold (2000) has called the ‘double
disengagement of the observer from the observed’ (15), which I shall argue
are core constituents of the cognitive conception of the fabula.
In the cognitive-formalist conception of the fabula, the story of the film is
first detached from the spectator that constructs it, and then the spectator’s
construction of the fabula is detached from the analyst’s dissection of the
film to complete the ‘double disengagement’. First of all, Bordwell’s cognitive
theory of narration implies that the fabula is constructed by a method of
‘problem solution’ that involves inferences, hypothesis testing, logical de-
duction, causal reconstructions, recognition, memory, etc. According to this
view, perception and narrative comprehension are governed by an active
engagement of the viewers with the cinematic information. Significantly,
perception and cognition are activities according to which ‘the sensory
input is filtered, transformed, filled in, and compared with other inputs to
build, inferentially, a consistent, stable world’ (Bordwell 1989, 22). In this
manner, spectators form their coherent and stable comprehension of the
fabula from the visual and narrative elements of the syuzhet by applying
various culturally, historically, and film-specifically acquired interpretative
schemata such as a familiarity with traditional narrative plot structures,
historically shaped stylistic conventions, genres, tropes, a knowledge of
actors and directors, etc. This ensures culturally diverse interpretations of
a given film and, therefore, the degree to which the fabula is inconsistent
between viewers depends on those cognitive processes that are culturally
variable, while universal cognitive processes (e.g. Chomsky’s universal
grammar) and hardwired neurological processes ensure a degree of consist-
ency (cf. Bordwell 1989, 22).
Although the differences pertaining to each spectator are thus not absent
from Bordwell’s model, his theory of narrative comprehension coheres with
classical cognitivism insofar as it filters out ‘disturbing’ elements in order
to study cognition in terms of data processing. Therefore, Bordwell isolates
the spectator’s cognitive processes from his or her affective and emotional
responses to the film in question. In order to justify this procedure, Bordwell
(1985a) assumes that a ‘spectator’s comprehension of the films’ narrative
is theoretically separable from his or her emotional responses’ (30). This
assumption is not only, as Bordwell (2011) himself points out, ‘consistent
with 1970s’s and 1980s’s cognitive science’ (sec. 2), it also adheres to the
second and third principles for the rejection of complexity by classical
science as traced out by Morin in his article ‘Restricted Complexity, General
Complexity’ (2007; cf. Chapter I).
In Chapter VI the extent to which these classical principles govern
contemporary film-theoretical attempts to understand narrative com-
plexity is uncovered. For now it suffices to recognize that the cognitive
comprehension of narration succeeds by means of a reductive isolation
of cognitive difficulties from one another. In particular, the isolation of
cognitive reasoning processes from the more ‘filthy’ emotional and affective
processes prepares the first detachment, which separates spectators from
the fabula that they have constructed. Underlying the cognitive conception
of the fabula is the idea that the story exists as a somewhat intrinsic quality
of the narrative – even if it is not entirely accessible and not all spectators
will construct it in the exact same manner. As Ingold (2000) observes in
relation to cognitive anthropology, the
claim of perceptual relativism – that people from different cultural

backgrounds perceive reality in different ways since they process the
same data of experience in terms of alternative frameworks of belief or
representational schemata – does not undermine but actually reinforces
the claim of natural science to deliver an authoritative account of how
nature really works. (15)
In a similar line of reasoning, the cognitive theory of narrative comprehen-

sion assumes a series of (partly) culturally variable cogitators, who ‘model’
or construct an independently existing reality (here: the fabula) that is not
immediately given, but must be inferred by means of reasoning. From this
perspective, narrative complexity becomes merely an expression of ‘the
difficulty of giving a definition or explanation’ and as such ‘complexity
relates only to appearances that are superficial or illusory’ (Morin 2007,
6). This amounts to the first disengagement, namely that between the
organizing spectator and the narrative that must be ordered. In line with
this disengagement, cognitive theories assume it to be the mission of the
analyst to ‘search, behind those appearances, the hidden order that is the
authentic reality of the universe’ (Morin 2007, 6).
This leads us to the second disengagement that concerns the detachment
of the analyst, who observes the fabula from a point of view that comes after
its telling, and the spectator, who in constructing the fabula is prone to
commit to ‘stereotypes, faulty inferences, and erroneous conclusions’ that
‘play a central role in narrative comprehension’ (Bordwell 2002a, 90). This

temporal detachment of the analyst introduces Morin’s first principle for
the reduction of complexity, since it makes the fabula appear as if it were
deterministically constructed (this is a problem that we shall return to in
Chapter VI). Insofar as the analyst from a detached temporal point of view
knows how the story ended, s/he is likely to interpret all narrative events in
terms of their necessity – these events had to be like this otherwise the fabula
would be different. The temporally detached perspective partly results from
the backward reasoning applied by the analysts, whose disengaged modes
of experience allow them to ascend to a realm of universal reason, which
is safely distanced from the erroneous conclusions that characterize the
spectator’s actual experience of the film.
It is crucial to note that the problem addressed here is not to be reduced
to the analysts’ application of reasoning, but concerns instead their detach-
ment from the temporality of the actual film experience that they aim to
comprehend. This methodological procedure, for instance, implies that the
chronological reorganization of the syuzhet has little impact on the level of
the fabula. However, if the fabula is conceptualized as a tool used in the di-
rect encounter between film and spectator, the analyst can no longer reason
from the a-temporal detached viewpoint of the coherent and causal-linear
fabula. Together with the commitment to the classical scientific principles
for the reduction of complexity, the cognitive double disengagement of the
fabula from the spectator and the spectator from the analyst contributes to
the linear onto-epistemology of cognitive film studies that underlines the
restricted conception of narration found in cognitive approaches to cinema.
An expansion of the cognitivist notion of narration can be found in the
work of Jerome Bruner, who perceives narration as a structuring principle
that has not just to do with literature, drama, and cinema. What Bruner (2004)
suggests is that particular modes ‘of telling and the ways of conceptualizing
that go with them become so habitual that they finally become recipes for
structuring experience itself’ (708). Narratives, as the quality of structuring
experience according to familiar patterns, thus become recipes ‘for laying
down routes into memory, for not only guiding the life narrative up to the
present but directing it into the future’ (708). A life, according to Bruner, is
not just a series of blunt facts, but is something which is constantly being
interpreted and reinterpreted. In a similar fashion, cinematic narratives are
open for constantly being recontextualized and reinterpreted.
This means that the relation between cinematic conventions and specta-
tors operates according to a principle of ‘learning to forget’. According to the
narratologist Seymour Chatman (1978), ‘[a]udiences come to recognize and
interpret conventions by “naturalizing” them’ (49). In the process of ‘natural-

izing’ cinematic conventions, spectators must acquire certain reading skills
(e.g. a close-up signals that the particular object or person is of special
interest) that by means of sheer repetition will be incorporated into the
automatic procedures of the cognitive-perceptual system, and thus come
to feel innate or natural to us (in the same manner as riding a bicycle or
playing the piano). Therefore, ‘[t]o naturalize a narrative convention means
not only to understand it, but to “forget” its conventional character’ (49).
This double procedure of learning and forgetting is also reflected in the
manner cinema alters our relation to – or comprehension of – ‘reality’.
As Nadel (2005) explains, ‘the process of rendering [classical cinematic]
conventions invisible acclimates us to specific notions of reality’ (428).
Cinematic conventions do, therefore, not just rehearse an influence on the
manner in which cinema is experienced, but also participate in creating the
‘recipes’ whereby we structure everyday experience. Rather than dismissing
the cognitive-formalist conception of narration, Bruner’s view enables us
to see this conception as a particular narrative itself: the story of the linear
temporality. However, this does not mean that the linear dimension is
not central for understanding the cinematic experience as an encounter
between film and spectator. On the contrary, cognitive film studies have
enriched our understanding of the linear dimension of cinema. What is
important now is to insist that this be combined with approaches capable
of opening up for the more non-linear dimensions of cinema.
Film-Philosophy and the Challenge of Contemporary Cinema
In this book I draw upon film-philosophical insights to challenge the Bord-

wellian mode of thinking about the cinematic experience as constituted by
a spectator (as a self-identical individual), who derives information from the
film audiovisually, which is then cognitively constructed and represented
in the mind in the shape of a coherent and chronological story, or fabula. In
his film-philosophy, Deleuze perceives of cinema in terms of an encounter
between the spectator and the film, from which concepts can be derived that
are able to reconstruct our conception of the cinematic experience. This is
since
theory too is something which is made, no less than its object. For many
people philosophy is something which is not ‘made’, but is preexisting,
ready-made, in a prefabricated sky. However, philosophical theory is
itself a practice, just as much as its object. It is no more abstract than

its object. It is a practice of concepts, and it must be judged in the light
of the other practices with which it interferes. A theory of cinema is
not ‘about’ cinema, but about the concepts cinema gives rise to and
which are themselves related to other concepts corresponding to other
practices, the practice of concepts in general having no privilege over
others, anymore than one object has over others. (Deleuze 2005b, 262)
The perhaps most important film-philosophical contribution consists in

this embrace of alternative ways of thinking about the cinematic medium
by allowing the films themselves to refine our approach and re-evaluate our
dominant methodologies, concepts, beliefs, and presuppositions. Although
Deleuze (2005b) at the end of the second cinema book speculates about a
coming (digital) revolution of cinema (255), he is mostly concerned with
problems related to the different manners in which time is thought of in
the ‘classical-modern divide’ (cf. Kovács 2007, 33-48).11, 12 In contemporary
cinema this opposition is no longer instrumental; rather filmmakers appear
to celebrate the unrestrictive creative potential that has emerged from
its dissolution. Spike Jonze and Charlie Kaufman’s Adaptation (2002)
is humorously depicting how a filmmaker – Charlie Kaufman’s alter ego
(Nicholas Cage) – is forced to choose between the linear requirements of
Hollywood and his own personal artistic ambitions owing to cinematic
modernism. The answer to this dilemma, it could be argued, is the complex
narrative form of the film Adaptation itself.13
Adaptation and other kindred films are discontented with the manner
in which films are fitted into binary oppositions, such as artistic/commer-
cial, linear/non-linear, or narrative schematic/self-reflexive, and decide to
explore the interstices of these instead. Thus, the challenge of contemporary
complex cinema is no longer well accounted for by the overarching Deleuz-
ian cinematic categories of the movement-image and time-image.
In the preface to Gilles Deleuze’s Time Machine (2003), David N. Rodowick
asserts that he could have written another book, ‘one that was thoroughly
critical of Deleuze’ (xiii). Part of what Rodowick finds problematic is the
‘curious anachronism’ of Deleuzian film-philosophy. The anachronism
arises while film-philosophy takes modern(ist) cinema as its vantage point
– ‘a period [...] that has already been exhausted’ (211, n. 2) – its true force
lies in its ability to shed light on the contemporary cinema and digital
culture. Rehearsing a string of similar arguments, Kovács (2000) has as-
serted that ‘from modern cinema, we must put Deleuze’s philosophy to
the task of understanding the future of audiovisual culture’ (169). While
film-philosophy reaches out to grasp the digital cinema yet-to-come, it

is nevertheless methodologically informed by the divide of classical and
modern(ist) cinema.
The Deleuzian separation of movement-image and the time-image has
left the impression that – here exemplified by Raymond Bellour (1998) – ‘the
great opposition between classical and modern cinema corresponds to the
gap between the two titles: The Movement-Image and The Time-Image’ (58).
Yet, the vital difference is that the Deleuzian image categories do not imply
rigid demarcations – a given film consists of an interplay of several different
types of images – in the same way as the distinction between classical and
modern(ist) cinema does. In this context Richard Smith (2001) has warned
that
it is deceptively simple to read the Cinema volumes along the axes of

classical cinema (movement-image) and modern cinema (time-image),
which coincides with classical philosophy (time as effect of movement)
and modern philosophy (movement as aberrance of time). (par. 26)
Smith supports his argument by referring to a series of directors and films

that – like complex narratives – resist any easy categorization into the
Deleuzian image categories.14 Furthermore, it can be added that the time-
image can both be seen as an ephemeral moment within a given film and
as a broader mode of describing post-war cinema.
Deleuze often struggles to find satisfactory examples of cinematic time-
images in their pure form, i.e. as entirely liberated from the sensory-motor
restraints of the linear movement-image. In fact, as Jacques Rancière (2006)
has argued, examples of such images are often derived from the narrative
situation, which inevitably carries a grain of linearity. This is, for instance,
the case when Deleuze refers to Alfred Hitchcock’s Rear Window (1954)
and Vertigo (1958) to explain the rupture of the sensory-motor system
characteristic of the time-image. However,
the ‘paralysis’ of each of these characters [Jeff in Rear Window and

Scottie in Vertigo] is actually only an aspect of the plot, a feature of the
narrative situation. It is hard to see in what ways the character’s motor
or psychomotor problems hinder linear arrangement of the images and
the action from moving forward. (Rancière 2006, 155)
According to Rancière (2006), Deleuze is unable to ‘find a visible incarnation

of a purely ideal rupture’ (116). Similarly, John Mullarkey (2009) argues that
‘the most powerful embodiment of the time-image through-out [Cinema 2]

is not an image at all but the lack of one: the irrational cut’ (96; cf. Rancière
2006, 105-124). In fact, the irrational cut only becomes an image if we accept
Deleuze’s anti-Platonic, Bergson-inspired notion of the image, which ‘abolish
the opposition between the physical world of movement and the psychologi-
cal world of the image’ (Rancière 2006, 109). Quoting Bergson, Deleuze (2005a)
writes, ‘[e]very image acts on others and reacts to others, on “all their facets
at once” and “by all their elements”’ (60; cf. Bergson 1991, 36-37). For Deleuze
existence is thus defined by movement in and between various forms of im-
ages, or as recapped by Pisters (1998), ‘we live in images and images live in us.
Images can affect us and make us think’ (sec. 2). The consequence, however,
is that images are not representational but relational, and thus the idea of
a ‘pure’ time-image is a philosophical-ontological abstraction rather than
an experiential fact. Ultimately, the movement-image and the time-image,
albeit distinct forms of images, acquire meaning from their mutual relation.
Rodowick explains that the time-image cannot be a simple placeholder for
modern(ist) cinema – although it can be a vital concept for understanding this
form of cinema – since it only exists in the encounter between cinema and
spectator. Rodowick (2003), therefore, considers it to be a heuristic abstraction:
The time-image is an image of memory. Or rather, for Deleuze it is

pure memory-image in Bergson’s sense [...] Like pure perception, ‘pure’
memory [...] is a heuristic abstraction. Consequently, in many of Deleuze’s
examples the border between the movement-image and the time-image
is fluid or indistinct. (88-89, emphasis in original)
This, nevertheless, does by no means discredit the ability of the concept

of the time-image to designate the non-linear dimension of the image – a
dimension that is almost completely absent in the Bordwellian theory of
narrative comprehension.
Another risk in perceiving movement-image and time-image as placehold-
ers of classical and modern(ist) cinema is that it inscribes a linear history of
the cinema into the cinema books. Perhaps in order to avoid such a criticism,
Deleuze initiates the preface of the French edition of Cinema 1 (2005a) by
emphasising that ‘this study is not a history of the cinema’ (xix). Several
critics, nevertheless, have criticized Deleuze for conceiving the history of
cinema in linear terms.15 Mullarkey (2009), for instance, argues that ‘there is
a teleology towards the time-image that is unavoidable given the evolutionist
nature of Deleuze’s story’ (101). This has given rise to a criticism of Deleuz-
ian film-philosophy as ‘essentialist’. Bordwell (1997) claims that Deleuze is
guilty of what he, following E.H. Gombrich, calls a ‘Hegelianism without

metaphysics’ (44). Bordwell elaborates as follows: ‘Treating film history as
the exfoliation of the a priori categories of an aesthetic system becomes a
scaled down version of Hegel’s idea that artistic change, like other cultural
developments, embodies the unfolding of the spirit’ (44). Bordwell warns
that the ‘unfolding-essence argument risks turning the result of historically
contingent factors into a necessary product somehow incipient from the
very start’ (32). According to Bordwell, Deleuze adopts a neo-Hegelian com-
monplace when he claims that ‘[i]t is never at the beginning that something
new, a new art is able to reveal its essence; what it was from the very outset
it can reveal only after a detour in its own evolution’ (Deleuze 2005b, 41;
cf. Bordwell 1997, 117).
Bordwell especially seems to have a case on occasions when Deleuze
categorically holds the ‘classical film form to be outmoded, passé, invalid,
discredited’ (Kovács 2007, 41), while paying tribute to the manner in which
modern(ist) cinema explores spatio-temporality in non-Euclidean and non-
linear terms. However, I find Deleuze’s argument convincing that the potential
for non-linearity was not invented by modern(ist) cinema, but formed part of
the inherent powers of cinema from its very birth (although I am not convinced
that this potential was not also realized before the post-war cinema).16 Still,
Bordwell’s criticism seems to be justified to the extent that modern(ist) cinema,
according to Deleuze, was a reaction to an assembly of cultural, social, artistic,
economic, philosophical conditions (cf. Deleuze 2005a), which gave rise to ‘the
evolution of cinema’s inherent power of articulating time’ (Kovács 2007, 41).
For these reasons, it is not unusual for Deleuzian film scholars to regard
a mode of anti-narrative filmmaking associated with modern(ist) cinema to
be the purest form of filmmaking. In relation to this, Bordwell’s criticism of
the ‘essentialism’ of film-philosophy is, perhaps, best applied to this kind of
Deleuzian modern(ist) valorization. Claire Colebrook (2002), for instance,
writes in her introduction to Deleuze that ‘[n]ot all films play with the force
of images, but the power or potential to free images from a fixed point of view
is what makes cinema cinema’ (37, emphasis in original). The drawback of
such a statement is that it restricts the complexity of cinema to one of its
components – its non-linear dimension freed from any fixed point of view.
Regrettably, what is easily transmitted in such an assumption is that first
and foremost the non-linear dimension associated with the time-image
and the modern(ist) cinema holds genuine artistic, aesthetic, and liberat-
ing value. Such a conception leads not only to a depreciation of classical
Hollywood but of the narrative dimension of cinema altogether. In the
following chapter, however, it shall be argued that Hitchcock’s film Stage
Fright (1950) is an example of how cinematic transgression and the rupture

of automatic sensory-motor judgements can equally well be achieved within
a classical, linear epistemology. This point is easily obscured by the Deleuz-
ian alignment of narration and the sensory-motor system.17
In relation to this, it is telling that Deleuzian film-philosophy has been ap-
plied to a wide range of topics such as horror films, feminism, national cinema,
third-world cinema, and cinematic affection. Taking this into consideration,
the amount of works concerned with the role of narration in Deleuzian film-
philosophy has been startlingly modest.18 The reasons for this are obvious,
since it seems impossible to reconcile Deleuzian film-philosophy with classical
narratology. Yet, in relation to the task of this study to reconceptualize the
fabula, film-philosophy as well as cognitive film science play formative roles.
For this to make sense, the film-philosophical interest in narration needs
to be revitalized. Once narration is no longer restricted to the linear onto-
epistemology of classical narratology, it becomes evident that narration is
actually a cornerstone in Deleuze’s film-philosophy. As Kovács (2000) observes:
The fact that the image cannot be divorced from time lies at the heart
of Deleuze’s understanding of the cinema, as is often pointed out, but
my point is that it also lies at the heart of his methodological project. If
time is included in the image by definition, Deleuze concludes that the
cinema is always in some sense narrative, that it cannot avoid telling a
story – though the kind of story it tells will vary radically [...] Indeed, it is
the very mutation of storytelling that informs Deleuze’s categories. (154)
I believe that Kovács’ revised appreciation of narration in the cinema books

enables a more productive and creative application of film-philosophy in
relation to contemporary complex narratives. Such would not only par-
ticipate in the attempts to formulate a positive understanding of recent
narrative innovations, but also provide a conceptual richness able to revise
our appreciation of how directors working within the classical paradigm
explore the inherent ambiguity of the cinematic medium. According to
a claim elaborated in the next chapter, ‘[c]onventional, linear, cause and
effect interpretative methods often fail to discern the multiple options for
meaning inherent in [classical] works because they lack the vocabulary
to express these conditions’ (Gillespie 2006, 123). In order to capture the
cinematic level that goes beyond conscious cognitive processes, Deleuzian
scholars have turned to the concept of affect.
It has been argued that cinematic affect ‘short-circuits’ our perceptual,
sensory-motor habits of selecting images that interest us only for potential
action (Colebrook 2002, 40; Powell 2007, 3). In this sense, the Deleuzian
affection-image expresses virtual possibilities waiting to be actualized
in particular conditions (cf. Deleuze 2005a, 104; Powell 2007, 3). Affect is,
therefore, to be understood as an urge to act, which cannot find its proper
release. It is in this sense that Brian Massumi (2002) defines affect as a
non-signifying, non-conscious intensity disconnected from the subjective,
signifying, functional-meaning axis to which the more familiar categories
of emotion belong. What is left is an urge for action, since ‘[a]ffect throws
into disarray the system of recognition and naming. At once, the image
gives something to feel and takes away my capacity to say “I feel.”’ (del
Rio 2011, 19-21). Affect can, therefore, be related to the plethora of virtual
– never actualized – pathways of a cinematic narration (cf. Chapter VI).
Consider, for instance, the following passage from Massumi’s Parables of
the Virtual (2002):
Intensity [a word Massumi uses interchangeably with affect] is incipi-

ence, incipient action and expression. Intensity is not only incipience. It
is also the beginning of a selection: the incipience of mutually exclusive
pathways of action and expression, all but one of which will be inhib-
ited, prevented from actualizing themselves completely. The crowd of
pretenders to actualization tend toward completion in a new selective
context.19 (30)
However, it is vital to understand affect as distinct from, yet nevertheless

deeply and intimately connected to, emotion and cognition. Ruth Leys (2011)
has argued that when affect is merely understood as a placeholder of that
‘which eludes form, cognition, and meaning’ (450), which it has been argued
is most often the case within the recent ‘turn to affect’ in cultural studies,
we are left with nothing but a renewed expression of the linear-non-linear
dichotomy. This insofar as to the
system of intensity belong all the attributes so prized by today’s self-

professed Deleuzean affect theorists – the attributes of the nonsemantic,
the non-linear, the autonomous, the vital, the singular, the new, the
anomalous, the indeterminate, the unpredictable, and the disruption of
fixed or ‘conventional’ meanings. (449)
Consequently, Leys contends that scholars working with affect tend to

succumb to a ‘false dichotomy between mind and matter’ (457), which is
accompanied by a ‘false opposition between the mind and the body’ (458).
She demonstrates how affect theorists, such as Brian Massumi and Eric
Shouse, display a ‘commitment to the idea that there is a disjunction or gap
between the subject’s affective processes and his or her cognition’ (450).
According to Leys, the main problem with this view is expressed in a
somewhat paradoxical tendency to ‘idealize the mind by defining it as a
purely disembodied consciousness’ (456). Shouse (2005) has, for instance,
argued that ‘the power of many forms of media lies [...] in their ability to
create affective resonances independent of content or meaning’ (par. 14,
emphasis added). For Shouse music is among the clearest examples of how
the ‘intensity of the impingement of sensations on the body can “mean”
more to people than meaning itself’ (par. 13). Shouse finds support for his
claim in the work of Jeremy Gilbert (2004), who has observed that
music has physical effects, which can be identified, described and dis-
cussed but which are not the same thing as it having meanings, and any
attempt to understand how music works in culture must [...] be able to
say something about those effects without trying to collapse them into
meanings. (Gilbert in Shouse 2005, par. 13, emphases in original)
As Leys observes, Shouse opposes the bodily affects to a rather limited – i.e.
a rationalist, computational – understanding of meaning. What is wrong
or confused about this, according to Leys (2011), is ‘the sharpness of the
dichotomy, which operates at once with a highly intellectualist or rationalist
concept of meaning and an unexamined assumption that everything that
is not “meaning” in this limited sense belongs to the body’ (458).
Leys traces the problem back to the double agency of the word ‘repre-
sentation’ within affect theory. First, the term representation identifies a
certain mode of thinking, i.e. the ‘computational model of mind’, which as
we have seen assumes the relation between the organism and the world to
be constituted by a ‘sharp separation between the cognizing, representing
mind and its objects’ (458, n. 43). However, as the recent developments
within the cognitive sciences reveal ‘[t]here is nothing inherently noncogni-
tive or nonintentionalist about [..] an embodied theory’ (458, n. 43). In
Chapter V we return to the possible advantages of adopting an embodied
approach to cognition, however, for now it is important to stress that the
problem concerning affect theory lies in the second sense of the word
‘representation’. Following Leys, the word is
used by the new affect theorists to refer to signification or meaning or

belief, and so on, as if what is at stake in eschewing a representationalist
theory of mind-world relations is not just a matter of rejecting a false

picture of how mind and body interact but involves rejecting the role
of signification, or cognition, or belief altogether. On this second usage,
the claim becomes that, since we do not represent the world to ourselves
according to the wrong, disembodied model of the mind, our relations
to the world are, in large measure, visceral, embodied, and affective and
hence not a matter of meaning or belief at all... (458. n. 43)
It is decisive to avoid this false dichotomy by understanding affect as distin-

guishable, yet not entirely autonomous or independent, from the spectator’s
emotional and cognitive responses to the cinematic event. A special interest
in the following chapters will, therefore, be devoted to how different films
exploit feedback loops in the cognition-emotion-affect circuitry.
For the reasons mentioned above, it is crucial to remember that, as
Anna Powell (2009) observes, the viewer’s affective encounter with im-
ages (even affection-images) is ‘inevitably shaped by plot mechanics and
characterization, which themselves build up the affective landscape of
the film’s narrative context’ (par. 7). Complex narratives do not privilege
affect over emotion and cognition, but rather engage the viewers in a
reconfiguration of their relation. This could, for instance, be achieved by a
fragmented non-linear narrative structure which stimulates the spectator’s
desire for meaning, coherence, order, and chronology, which in turn can
only find satisfaction via an active exploration of the virtual pathways of
the films. According to Powell, Deleuze sees the exact same potential in the
affection-images to ‘open ourselves up to the film’s potential to stimulate
thought beyond what the images show in terms of their obvious content or
what the film is “about” in common-sense terms’ (par. 7).
Affect can be perceived as ‘autonomous to the degree to which it escapes
confinement in the particular body whose vitality, or potential for interac-
tion, it is’ (Massumi 2002, 35). However, this should not lead us to exclude
the role that cognition plays in affect. In relation to contemporary cinema,
Jamie Skye Bianco (2004) asserts that the complex narrative – or, what
she terms the ‘techno-cinematic event’ – ‘registers affectively’ (402, n. 19).
While not disputing this claim (it could be argued that not only ‘techno-
cinematic’ events, but cinema more generally registers affectively, but also
cognitively and emotionally), an exclusive focus on affect merely reverses,
rather than overcomes, the linear-non-linear dichotomy – favouring the
latter on behalf of the former. The danger of this approach is to downplay
the intricate relations between the non-linear, ‘pre- and postcontextual,
pre- and postpersonal’ (Massumi 2002, 217) dimension of affect and the
more fixed, linguistic, semiotic, analytical, rational, and linear dimension

of cognition.
Focusing on the opposition of the movement-image and time-image,
enabled Deleuze to study a non-linear dimension of cinema that had been
ignored, excluded, or diminished in most of the preceding studies on cin-
ema. Deleuze has consequently been a great source of inspirations for film
theoreticians who have attempted to overthrow the dominance of linear
thinking in film studies. Rodowick (2001), for instance, draws heavily on
Deleuze to understand what he terms the ‘figural’:
In short, the figural cannot adequately be described by the logic of iden-

tity characteristic of most extant aesthetics theories and philosophies of
the sign and of language. To comprehend the figural, it is necessary to
transform completely how the term ‘discourse’ is understood by tracing
out what Modern philosophy has systematically excluded or exiled:
incommensurable spaces, non-linear dynamics, temporal complexity and
heterogeneity, logic unruled by the principle of noncontradiction. (49)
Therefore, uncritical rehearsals of f ilm-philosophical arguments to

explain the innovations of contemporary complex narratives risk re-
ducing these to avatars of either classical cinema or modernism, or to
the f ilms’ underlying linear or non-linear components. In relation to
this, I agree with David Martin-Jones (2006), who describes a series of
contemporary films as ‘hybrid’ images and explains: ‘As Deleuze posited
the time-image as an alternative to the movement-image, these “hybrid”
(movement-/time-image) f ilms enable both a rethinking of Deleuze’s
categories, and provide a range of new contexts within which to apply
his terms’ (4). Martin-Jones argues that contemporary cinema can be
perceived as ‘time-images “caught in the act” of becoming movement-
images’ (5, emphasis omitted). However, uneasy with the unresolved
double temporal ontology of contemporary cinema (as both linear and
non-linear) Martin-Jones makes it his task to analytically dissect a series
of these f ilms with the purpose of determining their proper temporal
logic. In this fashion, ‘each hybrid f ilm will be examined to see how
far, or to what degree, it can be seen to exist as a movement- or a time-
image’ (4). I believe this methodological choice to be counterproductive
since complex narratives no longer obey such a strict division (cf. Shaviro
2010, 2012). Consequently, my approach can best be explained as doing
the exact opposite – namely to explore how contemporary cinema by not
resolving the temporal ontological anxiety allows us to reconceptualize
the cinematic experience by insisting on exploring cinema between the

movement-image and time-image.
Gaspar Noé’s Irréversible (2002) is an example of why the temporality
of complex narratives cannot be reduced to either their supposed linear
or non-linear elements. The film consists of a series of linear narrative
segments that are presented in a reverse-chronological order, thereby
encouraging the viewers to ‘restructure’ them into a causal-linear storyline.
From the perspective of the analytically constructed fabula, the narra-
tive situation of the characters moves from happy to sad, yet viewers are
unlikely to share the ‘joy’ of the characters, because the experience of their
personal tragedies – that the viewers have witnessed, but inevitably awaits
the characters – still resonate within them. Interestingly, this structure
serves as a constant reminder of the irreversible nature of temporality. Thus,
it would be tempting to argue that the film adheres to an underlying linear
logic. The problem with this argument is that it does not sufficiently take
into account the complex temporal experience that the film induces in its
spectator, whose causal-linear rearrangement of the narrative ensures that
the peaceful existence of the couple at the end of the film cannot be seen
without being accompanied (virtually, in the spectator’s imagination) with
the horrors that are not to come, but which already have been. What is at
stake here is something other than the movement-image or time-image
although the complex narrative recalls both. In this fashion, films like
Irréversible help us rethink the complex interrelations of the movement-
image and time-image. Thus, it is incisive to remember that Deleuze (2005b)
contrasts his two overarching image-categories in order to understand their
individual qualities. As he writes: ‘[w]e can choose between emphasizing
the continuity of cinema as a whole, or emphasizing the difference between
the classical and the modern’ (39). Where Deleuze chooses the latter option,
the ‘will to complexity’ found in contemporary cinema, on the other hand,
invites us to explore these dimensions according to their interconnections.
In relation to this, Morin (2007) proposes that complexity calls for an
onto-epistemological transition based upon a study of mutual relations.
He argues:
In opposition to reduction, complexity requires that one tries to com-

prehend the relations between the whole and the parts. The knowledge
of the parts is not enough, the knowledge of the whole as a whole is not
enough, if one ignores its parts; one is thus brought to make a come and
go in loop to gather the knowledge of the whole and its parts. Thus, the
principle of reduction is substituted by a principle that conceives the
relation of whole-part mutual implication. The principle of disjunction,

of separation (between objects, between disciplines, between notions,
between subject and object of knowledge), should be substituted by a
principle that maintains the distinction, but that tries to establish the
relation. (10)
What follows from this is that complex problems cannot be sufficiently

tackled from the limited perspective of a single discipline. Indeed, the
reliance upon segregation of knowledge into smaller units or academic fields
is an idea that defies complexity. Morin, therefore, calls for an approach
across scientific disciplines:
It is necessary to amplify [Vico’s] idea of scienza nuova by introducing the

interaction between the simple and the complex, by conceiving a science
that does not suppress disciplines but connects them, and consequently
makes them fertile, a science which can at the same time distinguish
and connect and where the transdisciplinarity is inseparable from com-
plexity. I repeat it, as much as the compartmentalization of disciplines
disintegrates the natural fabric of complexity, as much a transdisciplinary
vision is capable of restoring it. (23)
Film studies have always valued transdisciplinarity in the spirit of Morin

as art studies, psychology, sociology, philosophy, and narrative studies
are connected to understand the complex processes that constitute the
cinematic experience. I aim to profit further from the transdisciplinary
nature of film studies by turning to embodied cognition, cognitive formal-
ism, affective neuroscience, constructivism, Deleuzian film-philosophy,
and complexity theory in an attempt to think beyond the dichotomies
that constitute contemporary film studies (cf. Chapter V and Chapter VI).
Focusing on the interplay of affect, emotion, and cognition does, therefore,
not only pose an alternative to the cognitive-formalist assumption that the
viewer’s comprehension of the narrative can be isolated from his or her
affective response to it (cf. Bordwell 1985a, 2011). It also avoids the reversal
of this argument within the affect theory of Massumi, Shouse, Bianco,
and others. Insisting on the interplay of linear and non-linear elements
counters the linear onto-epistemology of cognitive film science, in a move
that simultaneously takes the study of complex narratives beyond the
Deleuzian opposition of movement-image (or, more precisely, action-image)
and time-image.
III. Narrative Ambiguity in the Classical
Cinema
Just as the Hollywood mode of production continues, the classical style remains
the dominant model for feature filmmaking.
‒ Bordwell & Staiger 1985, 370
Certainly, people continue to make [classical narrative] films: the greatest

commercial successes always take that route, but the soul of the cinema no
longer does […] We hardly believe any longer that a global situation can give rise
to an action which is capable of modifying it – no more than we believe that an
action can force a situation to disclose itself, even partially. The most ‘healthy’
illusions fall. The first things to be compromised everywhere are the linkages of
the situation-action, action-reaction, excitation-response, in short the sensory-
motor links, which produced the action image.
‒ Deleuze 2005a, 210-211
Introduction
The ‘classical Hollywood film’ has traditionally been associated with a lin-
ear, cause-effect mode of sense-making. This is particular the case because
the classical film, due to its focus on narrative progression and resolution,
can be said to adhere to a linear onto-epistemology (cf. Chapter II). This
onto-epistemology has been perceived as a justification for scholars to
utilize a methodology that itself adheres to classical scientific cause-effect
principles that produce a monolithic linear understanding of the classical
cinematic regime. This chapter questions such restricted comprehension
of the classical paradigm, and argues that it stems from a lack of sensitivity
for the inherent ambiguities that reside in the cause-effect dramaturgy of
classical cinema.
The linear conception of Hollywood has been elaborated at length in
Bordwell, Thompson, & Staiger’s The Classical Hollywood Cinema (1985).
In this volume, Bordwell (1985b) sums up the main characteristics of the
classical film as follows: ‘Here in brief is the premise of Hollywood story
construction: causality, consequence, psychological motivations, the drive
towards overcoming obstacles and achieving goals. Character-centered –
i.e., personal or psychological – causality is the armature of the classical
story’ (16). In Deleuze’s (2005a) depiction of the action-image as the model,

which ‘produced the universal triumph of the American cinema’ (145), the
linearity of the Hollywood film is equally emphasized. Consequently, in
spite of usual reservations, Deleuzian scholars have shown a surprising
acceptance of the cognitive-formalist description of the classical film. In
Gilles Deleuze’s Time Machine (2003), Rodowick, for instance, holds that ‘[t]
he relation between time and reading in the movement-image is quite well
accounted for in the formalist model’ (230, n. 8).
Indeed, the classical f ilm does appear to be one of the most linear,
prescriptive methods of discourse, since it centres the action on its main
protagonist’s attempts to solve a single and dominant problem (cf. Gillespie
2006, 123). While this conception has been affirmed by most academic
studies on the classical film, the aim here is not to contribute to the common
(mis)representation of Hollywood as a rigid system, strictly bound by a series
of linear rules and conventions. Although the linear description of classical
cinema is not entirely unjustified, this chapter discusses Alfred Hitchcock’s
Stage Fright (1950) as a borderline example that helps to elucidate how
our interpretative methods contribute to the monolithic linear perception
of this type of cinema.1 While classical cinema may not challenge linear,
interpretative methods in the manner modern(ist) cinema or complex
narratives do, it cannot be concluded from this that the spectator’s mental
experience is entirely dominated by causal reasoning.
In fact, filmmakers (directors, screenwriters, actors, cinematographers,
etc.) working within the parameters of the classical Hollywood film must
‘construct stories and performances that avoid the predictable and stimulate
a range of possible interpretive responses’ (Gillespie 2006, 124). The problem
pertaining to overt linear interpretative methods concerns the danger
of diminishing the non-linear aspects of the classical paradigm, i.e. the
ambiguities, the multiple analytical options, the breaks with the progres-
sive forward flow of the narrative, and the transgressions of established
conventions. Since linear interpretative methods cannot accommodate
the ‘complexity of human behavior and the diversity of viewer expecta-
tions’ (Gillespie 2006, 127), our analytical cause-effect concepts produce a
disproportionately linear image of classical Hollywood.
Hitchcock’s Stage Fright at first appears to confirm most of the linear
traits of the classical film. Yet, due to its (in)famous ‘lying flashback’, the
film actually challenges unreflective and automatic linear reasoning. In
doing so, it enables an examination of the processes whereby the classical
film traditionally facilitates a ‘naturalization’ of specific classical conven-
tions that over time come to appear as the standard and natural mode of
Narr ative Ambiguit y in the Cl assical Cinema 55
organizing experience (cf. Chapter II). Insofar as the deceptive ‘flashback’

of Stage Fright can be seen as a reflection upon the inherent ambiguity
of the cinematic medium (capable of reproducing ‘reality’ in a deceptive
manner), it demonstrates that the classical film is not as clear-cut as it is
often presumed to be. Furthermore, in obstructing the manner in which
cinematic conventions become ‘natural’, the ‘flashback’ potentially renders
visible the degree to which our conceptual frameworks reaffirm the sup-
posed ‘naturalness’ of the linear mode of experiencing and interpreting
our surroundings. Given the strong association between the linear mode
of experience (as described in cognitive [f ilm] science) and the causal
manner in which (classical) cinema encourages us to organize its events,
it is possible to designate a particular manner of structuring cinematic
experience, which henceforth will be referred to as linear cinematic percep-
tion . Through this reasoning, Stage Fright can be seen as an example of
a film that challenges the automatisms of linear cinematic perception .2
Perceiving the film in this fashion resonates with the growing body of
research that challenges the predominant linear conception of classical
cinema. In contrast to cognitive-formalist studies, I maintain that the
classical film does not subordinate everything to the linear progression
of the narrative. Instead, the ‘linear components alternate with non-linear
moments of spectacle’ (Keating 2006, 5). A revisionist study of classical
cinema thus complements ‘[c]onventional, linear, cause and effect interpre-
tative methods [that] often fail to discern the multiple options for meaning
inherent in [classical] works because they lack the vocabulary to express
these conditions’ (Gillespie 2006, 123).
Nowhere is the domination of linearity on our conceptual tools as
evident as in cognitive-formalist film narratology. According to Edward
Branigan (1992), ‘narrative is a way of organizing spatial and temporal
data into a cause-effect chain of events with a beginning, middle and
end’ (3). Bordwell (1985a) also emphasizes the linear relation of events
as essential for narrative comprehension. Since the central goal of the
spectator is the construction of a fabula, ‘the spectators seek to grasp
the filmic continuum as a set of events occurring in defined settings and
unified by principles of temporality and causation’ (34). More precisely,
the spectator comprehends the narrative in terms of the inevitable
linear rearrangement of the narrative continuum. This conception con-
nects with the underlying assumption that the widespread popularity
of the Hollywood format pertains to its easy facilitation of a unif ied
narrative continuum. In this fashion, the cognitive theory of narrative
comprehension and the classical Hollywood format reaffirm each other.
Bordwell states that ‘[a]s a template for organizing causality and time,
the canonic story accords well with a Constructivist theory of narrative
comprehension’ (35).
According to Bordwell’s constructivist theory, ‘perceiving and thinking
are active goal-oriented processes’ (31), and cognitive film science conse-
quently understands narrative comprehension according to a revisionist
cycle of perceptual-cognitive activity (31). Here the pleasure of narrative
comprehension is largely dependent on the spectator’s realization of narra-
tive closure. In order to achieve this closure, the spectator must be active,
since the ‘artwork is necessarily incomplete needing to be unified and
fleshed out by the active participation of the perceiver’ (32). When Bordwell
writes that the ‘spectator thinks’ (32, emphasis in original), he really means
that the spectator employs linear reasoning.
This is exactly the kind of thinking that Deleuze has criticized the
movement-image for inducing. For him, the task of cinema is not, as
Bordwell assumes, to tell stories that can be organized along the lines of
habitual perception. For him, thought in cinema is not primarily concerned
with linear rationalizations, but should instead be seen as realizing the
capacity of the medium for inducing a shock to thought. As Deleuze
(2005b) writes, ‘[i]t is only when movement becomes automatic that the
artistic essence of the image is realized: producing a shock to thought,
communicating vibrations to the cortex, touching the nervous and cerebral
system directly’ (151).
Thomas Elsaesser & Malte Hagener (2010) have observed that Deleuze
connects cinema’s ability to produce automatic movement to modern
philosophy, since it ‘philosophizes’ about movement and time with its own
means (158-159). Filmmakers thus become philosophers, distinct in their
ability to think of technology, body, and brain within a single lifeworld (159).
Although their views on what cinematic thinking more generally entails
differ, Bordwell and Deleuze appear to agree that the classical film primarily
induces a linear mode of thinking.
For Deleuze, the movement-image – and in particular the action-image
of classical Hollywood up until the 1950s – is emblematic of the linear
mode of thinking, because it restricts the image to the constraints of
habitual sensory-motor perception. Deleuze (1995) explains that ‘the
cinema of action depicts sensory-motor situations: there are characters,
in a certain situation, who act, perhaps very violently, according to
how they perceive the situation. Actions are linked to perceptions and
perceptions develop into actions’ (51). The classical cinema, following
Deleuze, is deeply connected to the sensory-motor system and thus comes
with a commitment to the linear onto-epistemology. The action-image

subordinates its constituents to a causal-linear logic and to rational
linkages according to which clearly defined causes prompt actions that
generate new situations, etc. In this fashion, the action-image produces
a narrative trajectory that steadily moves towards the establishment of
the good and healthy moral sense. Therefore, Gregory Flaxman (2000)
has argued that with the ‘sensory-motor schema we entrusted ourselves
to the system of Truth’ (5). Consequently, Deleuzian film scholars often
refer to the classical film to perform a critique of its inherent linearity,
which is deemed philosophically less valuable. As Raymond Bellour (2010)
acutely sums it up while reflecting upon how to distinguish ‘good’ from
‘bad’ cinema:
How does one evaluate a good film, the good films in the cinema? One
might answer, as Deleuze does in his books, that it is by the capacity
that these films give him to think of them philosophically, to produce
concepts from them. (par. 42)
Implied is the idea that the linear format of the classical film generally
lacks the ability to induce ‘true’ philosophical thinking from which new
concepts can be produced.
Hence, many Deleuzians would probably tend to agree with Nitzan Ben
Shaul (2012), who has contended that ‘most films encourage a closed state
of mind, biasing our cognitive processes towards a reductive and selective
attention to incoming data’ (1). For Deleuze, true original thinking in
cinema cannot primarily be the reconstruction of a coherent fabula, since
(linear) narrative explanation is retroactive (cf. Polkinghorne 1988, 21).
As a result of this, he is particularly interested in a kind of cinema that
takes the spectator beyond the sensory-motor restrictions of classical
narration (i.e. the time-image). Contrarily, as we have seen, Bordwell
(1985a) def ines narrative comprehension according to the spectator’s
ability to reconstruct a causal-linear continuum (34). As a result, linear
reasoning constitutes the primary mental activity of the viewer, who
‘must take as a central cognitive goal the construction of a more or less
intelligible story’ (33).
However, since the linear continuum is already implied by the f ilm
itself, the mental activity of the spectator is seriously restricted in the
cognitive model. In relation to this, Ben Shaul (2012) has argued that the
canonical story format actively discourages spectators from comprehend-
ing the narratives in terms of ‘optional thinking’. Ben Shaul uses this term
to describe the cognitive ability to ‘generate, perceive, or compare and

assess alternative hypotheses that offer explanations for lifelike events’ (2).
The opposite of ‘optional thinking’ is ‘closed-mindedness’ which among
other factors ‘implies that in thinking about others we may often stick to
prior impressions or preconceived notions rather than flexibly altering our
opinions whenever relevant new information turns up’ (Kruglanski 2004,
2). According to Ben Shaul, classical cinema encourages closed-mindedness
because it restrains the activity of the spectator, insofar as ‘the viewers’
potential sense of freedom of choice at narrative turning points [...] does
not usually trigger a fruitful or enjoyable consideration of options’ (26).
What is problematic is that Ben Shaul assumes the lack of spectatorial
freedom to be an inherent feature of the films in question (in the same
fashion that cognitivism perceives narrative complexity to be an inherent
feature of the plot). The underlying assumption at work here is that the story
material can be organized in terms of causality and chronology without a
qualitative change of the overall narrative experience occurring from this,
i.e. that the narrative continuum is inherently linear. However, if cinematic
complexity is not an inherent feature of the films themselves, as I have
argued, the same should be said about cinematic linearity. The linearity of
classical cinema must at least be partly produced in the manner we have
traditionally comprehended these films.
Stage Fright is an example of a classical film that embodies a linear
narrative structure, but which in doing so enables us to reflect upon the
automatisms that govern habitual narrative comprehension. The f ilm
demonstrates that in order to enhance our appreciation of classical cinema,
as Gillespie (2006) has demonstrated, the analyst must not reaffirm its
linearity, but instead devote the attention to the underexamined manners
in which ‘the classic Hollywood cinema narrative format underscores
uncertainty’ (126). This requires a reconfiguration of linear conceptual
frameworks ‘with the metaphors derived from new ways of perceiving and
systems of argument that no longer rely upon linear connections to give
them validity’ (129).
Narrative Transgression in Stage Fright
In an interview with François Truffaut initially published in 1967,

Hitchcock reflects on the controversies surrounding his 1950 film Stage
Fright. Central for the debates was the film’s so-called ‘lying flashback’
in which a character’s lie is presented by use of the classical conventions
for the flashback. Reportedly this caused parts of the audience to take
the ‘flashback’ as a truthful visualization of earlier fabula events. The
reception and comprehension of the film’s deceptive technique seems to
have undergone an interesting development from the time of its initial
release, where reports tell of an outburst among audiences and critics
alike, who allegedly felt deeply disappointed with the ‘unfair’ way they
were tricked by the ‘lying flashback’ (Thompson 1988, 141). Even François
Truffaut (1984), who is a known admirer of the work of Hitchcock, ap-
pears to side with the French critics, who were ‘particular critical of [the
“lying flashback”]’ (189). In the same interview, Hitchcock (1984) defends
the ‘flashback’ by reminding us that lies and deceptions are agreed upon
as vital ingredients of cinema. He then goes on to ask why it should be
assumed to be innately against the rules of cinema to present a lie in the
form of a flashback:
Strangely enough, in movies, people never object if a man is shown telling

a lie. And it’s also acceptable, when a character tells a story about the
past, for the flashback to show as if it were taking place in the present,
so why is it we can’t tell a lie through a flashback? (189)
It is in particular its readiness to ‘denaturalize’ cinematic conventions that

makes it possible to regard Stage Fright as a precursor to contemporary
complex cinema (cf. Bordwell 2006, 72-73).
Today experiments with narrative transgressions, such as ‘lying
flashbacks’, are more commonplace. As cinematic conventions engage
in dynamic relations with viewers and their expectations, the latter are
changing as well. Consider, for instance, the positive reception of The Usual
Suspects (Singer 1995) – a film that much later used the device of ‘lying
flashbacks’.3 The different receptions of the two films suggest that such
narrative experiments are more welcomed by a ‘post-classical’ audience
than a ‘classical’ one.
For Stage Fright’s narrative twist to work, it is decisive that viewers
comprehend the film in a manner resembling that described by Bordwell
(1985a). In narrative comprehension, most of our cognitive processes go
unattained. However, the ‘lying flashback’ of Stage Fright is liable to make
us aware of the ordinary processes that govern narrative comprehensions.
Once it becomes clear that Stage Fright is deceptive rather than com-
municative, the film is thus potentially capable of laying bare the otherwise
‘hidden’ conventions of the classical flashback.
Stage Fright opens with a safety curtain going up to reveal the busy
street life of London from a bird’s-eye view. From the opening shot the film
cuts to a car in which Jonathan Cooper (Richard Todd) and Eve Gill (Jane
Wyman) – the two main characters of the film – are driving. Jonathan is
explaining Eve the trouble he has gotten himself into while trying to help
cover for his mistress Charlotte Inwood (Marlene Dietrich) after she has
murdered her husband. Jonathan tells Eve that he believes Charlotte, a
high-society actress, framed him so that the police now takes him as their
prime suspect. Eve, who has long had romantic feelings for Jonathan, agrees
to help her friend by hiding him in her father’s boathouse. Eve, herself an
aspiring actress, welcomes the drama and excitement that has suddenly
entered into her dull existence, and she decides to take matters into her
own hands and to prove Jonathan’s innocence.
At this point, the film introduces a series of role plays that accompany
its theatrical themes of lies, illusions, and deceptions. Eve, for instance, at-
tempts to get closer to Charlotte by pretending to be the temporary replace-
ment for her maid. In her struggles to help Jonathan, Eve encounters the
police officer Ordinary Smith (Michael Wilding) with whom she gradually
falls in love. Together they blackmail Charlotte for information by letting
her believe that they are in possession of a bloodstained dress she wore
on the day of the murder. Their plan works and Charlotte confesses, but
much to their surprise, she insists that she is merely the accomplice, and
that Jonathan is the actual murderer. Despite this, Eve remains devoted to
Jonathan’s version of the events and helps him escape from the police. She
realizes Jonathan’s true character too late, but manages to trick him and
thus avoids becoming his next victim. In the end, Jonathan is decapitated
by the safety curtain and the film thus finishes with a theatrical metaphor
that matches its opening scene.
The plot briefly sketched out above illustrates the prevalence of clas-
sical storytelling strategies throughout the film. The narrative is driven
by Eve and her desire to prove Jonathan’s innocence. As such, the plot
is pushed forward by a goal-oriented main character, who demonstrates
the individual’s ability to change circumstances for the better. The plot is
structured causal-logically so each scene logically extends into the next,
etc. Furthermore, the film contains a double plot structure according to
which Eve’s attempts to unravel the murder puzzle is the main narrative
thread, while the second plotline concerns her romantic affiliation with
Jonathan/Ordinary.
Although the classical film does not usually invite spectators to reflect
upon its narrative construction (Bordwell et al. 1985), the attentive viewer
well acquainted with classical conventions may be alerted, when Eve falls
in love with Ordinary. As the classical film often ends with a successful
outcome for its heterosexual romance (Bordwell et al. 1985), Eve at one point
must be forced to choose between Jonathan and Ordinary. This change of
affiliation could be held as significant if the audience recognizes that in the
cinema of Hitchcock, as Deleuze (2005a) has observed, ‘actions, affections,
perceptions, all is interpretation from the beginning to the end’ (204).
One of Hitchcock’s major contributions to cinema is exactly his keenness
to play with the expectations of his audience. For Deleuze, Hitchcock was
the first to perceive the constitution of cinema as a function that not only
involves director and film, but also the spectator (206). In incorporating
the spectator into the corpus of cinema, Hitchcock’s cinematography has
contributed to a raised awareness of the spectator’s active part in producing
the meaning of the film. 4 In Stage Fright, Hitchcock makes his viewers
aware of their active participation in constructing the story in a manner
that simultaneously breaks with the expectation of having their efforts
rewarded with a gradual increase in knowledge (cf. Bordwell 1985a, 159). It
is not that Stage Fright does not provide a coherent fabula, but the film’s
twist is revealed almost at the end of the story and viewers will thus lack
the feeling of having anticipated the story outcome.
Traditionally, the only permissible manipulation of story order in the
classical film is the flashback, which together with the system of continu-
ity editing stresses the continuous nature of temporality in the classical
paradigm (cf. Bordwell 1985a, 42-49). By allowing viewers to reflect on
their own participation in the construction of the narrative, Stage Fright
challenges the invisible, linear flow of the classical film. In the classical
film, ‘[t]he viewer concentrates on constructing the fabula, not on asking
why the narration is representing the fabula in a particular way – a question
more typical of art-cinema narration’ (Bordwell 2006, 162).
A recurring element in the cinema of Hitchcock is that things are rarely
what they appear to be at first glance. This is also the case in Stage Fright,
which is underlined by the film’s constant play on the themes of lies and
deceptions, and by its abundance of theatrical metaphors. Once scrutinized
the film opens up a perpetual, yet subtle, play with the classical conventions.
To appreciate this fully, the viewers must become aware of the automatic
operations that temporally organize the narrative continuum into a linear
storyline. Without such awareness the linearity may appear as inherent
to the film and thus independent of its spectator. In this view, the non-
linearity of Stage Fright results from its exposure of classical conventions
as conventions. That is, the linear cinematic experience is ruptured once
spectators are made aware that the linear mode of organization so pervasive
in narrative cinema is not natural but conventional.
One of the fascinating things about contemporary complex cinema is
the manner in which viewer expectations have increasingly been incorpo-
rated into the structure of the narratives. The result, as will be argued in
subsequent chapters, is that the linear nature of the cinematic experience
is being increasingly ‘denaturalized’. The frequency with which spectators
are incorporated into the narrative structures of contemporary cinema
has contributed to an increased scholarly awareness of the resemblance
between contemporary cinema and (narrative) games (e.g. Bianco 2004;
Elsaesser 2009; Kallay 2013; Simons 2007, 2008). The ‘lying flashback’ can
be seen as an early case of a cinematic device employed to ‘play games’
with the audience.
Most of the writings on Stage Fright are concerned, in one way or
another, with its so-called ‘lying flashback’. The present treatment will be
no exception. Yet, while most treatments of Stage Fright are concerned
with questions of narrator/narration (e.g. Thompson 1988; Casetti & Bohne
1986; Currie 1995; Richter 2005), the present differs by focusing on how
this device can be seen as both a destabilization and an affirmation of
linear interpretative methods. How can we understand the interpretative
strategies utilized by spectators who felt cheated by the ‘lying flashback’?
Does Hitchcock fail to induce this scene with enough ambiguity or are our
interpretative methods too time-dependent and not sensitive enough to
note such ambiguities? Our comprehension of the ‘lying flashback’ reveals
a lot about the processes whereby narrative information is induced with
meaning.
Actually the ‘lying flashback’ is strictly speaking not a flashback although
it is carried out in agreement with the conventions for this classical device.
In the (in)famous scene, Eve encourages Jonathan to tell her about the events
that caused him to become the prime suspect of the murder of Charlotte’s
husband. As Jonathan starts narrating (‘I was in the kitchen. It was about
5:00’), the scene takes place in his car, but with the next sentence (‘the door-
bell rang and I went down to see who it was’) the image slowly dissolves and
provides us with a visual accompaniment of the events Jonathan narrates
and ultimately the film completely takes over his narrative. What factors
justify taking Jonathan’s story as an objective version of fabula events?
According to the conventions of the flashback, the narrative information
can exceed the restricted viewpoints of the subject having the flashback. As
Bordwell (1985a) explains, ‘the range of knowledge in the flashback portion
is often not identical with that of the character doing the remembering’
(162, emphasis in original). In this fashion, the film appears to fill in the
blank spots of Jonathan’s story. From this it can be inferred that the film
contributes to Jonathan’s lie. This seems to be Bordwell’s interpretation of
the ‘lying flashback’, when he argues that since the film does not mark the
falseness of the ‘flashback’, it can be perceived as an indirect validation of
the information conveyed in it. According to this view, the narration, and
not just Jonathan, is unreliable:
It is not just the character’s yarn that is unreliable. The film’s narration
shows itself to be duplicitous by neglecting to suggest any inadequacies in
Johnnie’s account and by appearing to be highly communicative – not just
reporting what the liar said but showing it as if it were indeed objectively
true. (61, emphasis added)
The problem with this statement, as Sarah Kozloff (1988) reminds us, is that
‘the image’s reliability is not essential; it is just a convention, and conven-
tions are made to be broken’ (115). It would, therefore, be more precise to
maintain that Hitchcock does not show the events as ‘objective truths’, but
in a manner that consciously plays with the viewers’ automatic application
of classical narrative schemata.
Bordwell’s treatment of Stage Fright is representative of a more
general lack of sensitivity towards the inherent ambiguity of the scene.
One reviewer has, for instance, argued that Stage Fright ‘falters because
it betrays the viewing audience’ and holds that ‘one of the fundamental
rules in moviemaking is that flashbacks should not deceive the audience’
(Brady 1998). Although not critical of the ‘flashback’ in the same manner,
Kristin Thompson (1988) holds the film to be a ‘duplicitous text’. For all, then,
Hitchcock does more than deliberately tricking his viewers into making a
series of false assumptions, he ‘betrays’ or ‘cheats’ the audience.
However, it is important to remain sensitive to the range of opportunities,
differences, and options that mark each cinematic experience as unique.
Viewers may interpret the ‘lying flashback’ differently, and some may even
operate with several mutually exclusive hypotheses for how to interpret this
narrative device. Thus, while acknowledging the deceptive nature of the
‘lying flashback’, it is simultaneously important to insist on its ambiguity.
With this in mind, it can be argued that Hitchcock has carefully constructed
the ‘flashback’ to allow (at least) the two following hypotheses:
Hypothesis 1: Charlotte killed her husband and made it look like Jonathan
did it alone. Eve will prove Jonathan’s innocence and win his heart at the
same time.
Hypothesis 2: Jonathan did commit the murder, but failed to win Charlotte’s
heart. He now plans to take advantage of Eve’s affection for him to get away
with it. This will somehow put Eve in danger.
In relation to this, it is interesting that Thompson (1988) assumes that the

‘lying flashback’ of Stage Fright automatically hinders spectators from
generating hypothesis 2. According to her, the text’s duplicity and delaying
devices ‘distort and reshape the true question that would have been posed
in the beginning – something like: Is Jonathan guilty?’ (144, emphasis in
original) Thompson appears to suppose a too homogenous notion of the
audience when she assumes that ‘without the reassuring “evidence” of
the apparent flashback, the spectator would not necessarily accept his
account and might still suspect that he had been involved in the murder’
(Thompson 1988, 144).
Thompson forcefully straightens out the carefully constructed ambi-
guities of the film when she blankly states, ‘the possibility of Johnny’s
guilt is raised only when that guilt is confirmed’ (144). She supports this
thesis by drawing attention to viewers, who upon the original theatrical
release reportedly expressed extreme disappointment with the ‘lying
flashback’ (141), yet several factors render the supposed unambiguity of
hypothesis 1 problematic. First, the ‘lying flashback’ is initially presented as
Jonathan’s story, insofar as the flashback starts with him telling Eve about
the events. Thus, it is Jonathan’s story before the classical conventions of the
flashback permit another assumption – namely that the use of flashback
conventions automatically assures that this is a truthful rendering of fabula
events.
Even if the ‘flashback’ is interpreted in this manner, the initial impression
that the ‘lying flashback’ is Jonathan’s story does not vanish completely. In-
stead, it can be assumed to have a lasting impact, even if this only amounts
to a slight doubt of the nobility of his character. Such lingering doubts are
nurtured by the film’s unflattering portrayal of Jonathan, who is largely
depicted as an anti-hero unwilling to take action himself. While Eve must do
all the hard work to prove that Charlotte committed the murder, Jonathan
is cowardly hiding in the boathouse. In this sense, Jonathan is everything
the typical classical hero is not: passive, dependent, and craven. It only adds
to the spectator’s possible suspicions that Jonathan burns his only piece of
evidence: the bloodstained dress. Yet, even this act is full of ambiguity and
open to alternative interpretations, as Jonathan could also have done this
in a naive hope to protect Charlotte, who we have gradually come to know
as a manipulative diva.
Once we become sensitive to the manner in which the f ilm plays

with our expectations, it becomes fascinating to explore how it opens
up various venues for alternative interpretations. For instance, Eve’s
change in affection from Jonathan to Ordinary Smith can suddenly be
comprehended as a sign of Jonathan’s dishonesty. Such interpretation,
however, requires of the audience not only to apply classical narrative
schemata, but also to reflect upon how the f ilm diverts from certain
classical conventions only to affirm others. Thus, the spectator is not only
required to have learned the classical conventions, but must also be able
to denaturalize them, i.e. to become (re)aware of them as conventions.5
If spectators are able to do so, hypothesis 1 may remain the most likely,
but it will no longer be possible to rule out other, competing hypotheses.
Consequently, Stage Fright will be experienced as an ambiguous
dramaturgical construct able to keep (at least) two hypotheses open
until the moment where Ordinary Smith provides the vital information
that Jonathan has murdered before.
In this reading, Stage Fright amplif ies an encouragement of op-
tional thinking in a similar fashion as Ben-Shaul (2012) has observed in
contemporary cinema. For him the popular Hollywood cinema is usually
‘closed-minded’:
Most movies use those aspects that make them popular in a way that
encourages the reduction and even blocking of the viewers’ optional
thinking processes. Through cognitive affects stemming from the ways
in which movies deploy narrative suspense, surprise, or the arousal of em-
pathy for protagonists, narrative uncertainty is felt as distressing, thereby
heightening the need to avoid it by seeking resolution and closure. (13)
However, to what extent is it the ‘films’ themselves or our interaction with

them that can rightly be deemed closed-minded? Thompson’s interpreta-
tion of the ‘lying flashback’ as leaving only one hypothesis open fails to
discern the inherent ambiguities of (classical) cinema. Thus, Thompson
does not merely assume a ‘closed-minded’ spectator, her reading of the
film is ‘closed-minded’.
As the relation between spectator and film is always bidirectional, our
conceptual and interpretative tools could equally be held responsible for
inducing a closed state of mind. Therefore, it is not sufficient to criticize
Hollywood films for inducing linear thinking (as Deleuze and Ben Shaul do).
At the very least such assessments should be combined with a self-critical
examination of our own cause-effect reasoning, our desire to eliminate
ambiguities, and our analytical narrowing down of available interpreta-

tions. In relation to this, Gillespie (2006) has contended that
as long as moviegoers follow a cause and effect method of tracing hier-

archical connections as the basis for explaining the significance of the
film, important elements will inevitably be ignored. This process of linear
reasoning identifies key features and narrows options for interpreting
them until arriving at an unambiguous response. (125)
Thompson’s treatment of Stage Fright perfectly exemplifies how ‘closed-

mindedness’ need not be an inherent feature of the films examined, but
could as much be perceived as the byproduct of the linear interpretative
methods utilized to comprehend them. In her treatment of Stage Fright,
Thompson assumes a seriously limited range of cognitive responses for
the viewers, whose task it becomes to assemble a linear trajectory already
implied by the narrative.
In Stage Fright, our cognitive-affective bond to Jonathan is abruptly
destabilized, forcing us to reinterpret the narrative continuum. Ben Shaul
assumes the lack of closure – or the openness of the text – as inciting a
‘distressing’ and ‘unpleasant’ state of mind for the viewers. However, the
lack of closure could also induce a productive and pleasurable opportunity
for the audience to explore the narrative continuum more freely, i.e. beyond
its linear restrictions. Since Ben Shaul considers ‘optional thinking’ to be
a mere function of the formal qualities of the film, his theory fails to ac-
commodate the non-linear aspects of the cinematic experience and itself
remains ‘closed-minded’.
The non-linear element that I argue to be present in Stage Fright is
ignored by interpretative models that assume the formal aspects of a film
to be determining for the spectator’s mental processes. Obviously, the
relation between the formal qualities of a film and the thought processes
of the viewers is not arbitrary. However, it can also not be reduced to
simplified cause-effect patterns. Therefore, non-linearity arises once the
classical movement ‘steadily toward a growing awareness of absolute truth’
(Bordwell 1985a, 159) is ruptured, abandoned, or subordinated to other
temporal layers of the narrative (cf. Deleuze 2005a, 2005b). Furthermore,
non-linearity serves a more general role as decisive for opening the in-
terpretative paths of the viewers, and for allowing them to explore the
narrative beyond the linear trajectory implied by the film. Since striving for
closure, the filtering out of irrelevant data, the application of stereotypes,
and a bias towards premature acceptance of given information, are all
aspects that encourage ‘closed-mindedness’, it is possible to argue that

the ‘lying flashback’ of Stage Fright is designed to discourage such linear
cognitive-affective biases. Non-linearity always coexists with linearity,
even if only to induce insecurity about the predetermined linear path of
a narrative.
In Stage Fright, non-linearity works to counter automatic linear
reasoning. This is supported thematically insofar as ‘virtually every scene
[...] relates to deception, to theatrical performance, or to both’ (Thomp-
son 1988, 151).6 Stage Fright tellingly opens with a theatrical metaphor
that introduces the themes of role play, fabrication, deception, and illusion
that permeate the film. The first image presented to the spectator is that
of a safety curtain onto which the Warner Brothers logo is transposed.
Shortly thereafter the curtain opens, and we enter the narrative world of
the characters.
The metaphor of the theatre also permeates the narrative world, es-
pecially because both main female characters – Eve and Charlotte – are
actresses. Added to this is the constant play of characters switching from
playing roles to directing other characters’ role play, or being the (unknow-
ing) audience of such role play.7 Hitchcock carefully creates a universe where
things are not what they claim or appear to be. Eve’s first line, ‘Looks like
we’re getting away with it,’ suggests that she is an accomplice of an as-yet
unknown crime. Although this hypothesis must quickly be replaced, it can
be seen as an early warning that viewers should prepare to discard their
initial assumptions (cf. Thompson 1988, 145).
Stage Fright deals with the topic of truths and falsities in a manner
that is characteristic of the classical paradigm, because these can be clearly
separated at the end. The aim of Stage Fright is not to question the pos-
sibility of absolute knowledge, such as, it can be argued, Rashômon (Kuro-
sawa 1950) – famous for narrating a single event from several, contradictive
viewpoints – does. Although Stage Fright shatters the ‘little-by-little’
movement towards narrative clarity, it does allow spectators to construct a
coherent and unambiguous account of its fabula at the end. The importance
of narrative clarity is underlined by the richness in detail of Jonathan’s
confession to Eve at the end of the film:
Eve, I hated to tell you that phoney story in your car that time, but there
was no other way. Charlotte did go on to my flat after I’d killed her
husband. Her dress was stained a bit, so I brought her a clean one. Then,
when she went to the theatre, I made a big stain on it, to make you believe
me. I’m telling you the truth.
This scene clearly demonstrates that the function of the ‘lying flashback’
is not to question the concept of narrative truth. Instead, the character
Eve, who is actively engaged in figuring out the true nature of events, but
simultaneously sticks to Jonathan’s version for far too long, might provide us
with an interesting take on the possible purpose of the ‘flashback’. This is the
case since Eve resembles the naive ‘closed-minded’ spectator, who is unable
to revise her conception of the events, because she is cognitive-affectively
biased by Jonathan’s deceptive skills. Rather than being interested in ques-
tioning the possibility of acquiring true, ‘objective’ knowledge about the
external world, Hitchcock seems interested in demonstrating how easily
we can be manipulated, deceived or distorted on our path to such truths.
Eve thus in many ways represents the Thompsonian spectator, who is
unable/unwilling to readjust the initial conviction that Jonathan’s story
is true. Eve decides to figure out what really happened based on the false
assumption that Jonathan is innocent. In this fashion, she enacts the im-
portant dramaturgical function of putting forth and exploring different
theories and hypotheses. Thereby, Eve’s interrogating efforts mirror those
of the spectator trying to make sense of the information by constructing a
‘true’ story/fabula. Eve, assumedly like many spectators, fails to recognize
how she is manipulated into taking a false account for granted. Hitchcock
thereby appears to suggest that not much differentiates how Eve becomes
cognitive-affectively biased by Jonathan, and the manner in which specta-
tors blindly accept the conventions of the moving image.
In this view, the ‘lying flashback’ urges the deceived spectator retrospec-
tively to pose the question: How did the narrative cheat me? This question
cannot be answered only by reference to the information provided in the
narrative. Thus, if viewers pose such a question they may become aware of
how the film utilized their automatic acceptance of classical conventions
to deceive them. The ‘lying flashback’ thus serves to make us aware of the
automatism of perception and cognition in relation to cinematic compre-
hension within the classical paradigm, i.e. of linear cinematic perception
. Therefore, the fundamental ambiguity introduced by the unreliable nar-
ration of Stage Fright does not just concern our ability to apply certain
interpretative schemata, but also our ability to replace or ‘forget’ these
(Chatman 1978). In this sense, the f ilm challenges our narrative drive
towards linear coherence, which constitutes our comprehension of cinema
in addition to much of our everyday sense-making (cf. Bruner 1991, 2004).
From this perspective, the unreliability of Stage Fright can be seen
as inducing a reflection upon – rather than simply an application of – the
quotidian beliefs and incorporated schemata that induce ‘closed-minded’
readings, and cause us to jump to premature conclusions, interpretations,

and evaluations (cf. Ferenz 2009, 281). Unreliable narratives, such as Stage
Fright, are not just a window onto the narrative world, but also a mir-
ror reminding us of the constructive nature of perception. Cognitivism
explains the processes whereby cinema transgresses certain conventions
and denaturalizes automatic perceptions by the concept of ‘defamiliariza-
tion’. According to cognitive formalism, the ‘lying flashback’ defamiliarizes
the naturalness with which purely contingent cinematic conventions have
been internalized. This account provides an explanatory framework for the
culturally dynamic nature of cinematic conventions. Yet, it assumes that the
monolithic linear conception of cognition can operate as a constant ‘back-
ground’ against which the culturally variable conventions can be evaluated.
Defamiliarization and Beyond
In order to explain the historical poetics of cinema, and the processes where-
by it transgresses its own norms and the viewers’ expectations, cognitive
formalism employs Viktor Shklovsky’s (1965) concept ostranenie (‘making
strange’), which translates as ‘defamiliarization’.8 In short, ostranenie is an
artistic technique by which spectators are forced to experience familiar
things in an unfamiliar or strange way, which enhances everyday percep-
tion. Frank Kessler (2010) has pointed out that the concept of defamiliariza-
tion thus necessarily presupposes familiarization, insofar as things in the
course of time grow familiar to us, and thus, become ‘invisible’ to us (62).
Laurent Jullier (2010) distinguishes between two forms of cinematic
defamiliarization. Cinema can show us a world, not immediately detect-
able due to the limitations of quotidian perception. However, cinema can
also defamiliarize by means of an alteration of its techniques, devices, or
conventions whereby a film presents a cinematic world. As Jullier points
out, the first, which he terms the ‘platonico-phenomenological’ option,
has been favoured by authors such as Siegfried Kracauer, Sergei Eisenstein,
André Bazin, Christian Metz, and Gilles Deleuze in their search for cinema’s
capability to reveal an ‘unnegotiated presence’, ‘virginal purity’ or ‘pure
state’ (124; cf. Szaloky 2005, 44). From this perspective, Jullier situates the
concept of defamiliarization at the heart of a quarrel between ‘culturalists-
constructivists’ (who believe perceptive habits to be structured by social
habitus or language) and ‘universalist-ecologists’ (who believe perceptive
habits to be products of much longer evolution to which the modern era is
of relative little importance) (139).
Similarly, László Tarnay (2010) holds the concept of defamiliarization

to illustrate the conflicting views of ‘ecologically committed theorist like
Joseph D. Anderson’ (who argues for a biological-evolutionary basis of film
viewing) and ‘philosophers of the new media’ (who highlight the cultural
determinants of vision and the role of prosthetic devices in shaping hu-
man perception) (142). Comparable to Jullier, Laszlo concludes that the
‘ecological and cultural conceptions of vision do not necessarily exclude
each other’ (142). Although I would argue that neither Bordwellian cognitive
formalism nor Deleuzian film-philosophy hold viewpoints as radical as
those sketched out by Jullier and Tarnay, the similarities and differences
in their respective approaches to the concept of defamiliarization reveal a
lot about their understanding of cinema more generally.
Defamiliarization is the central analytical device for Thompson’s (1998)
understanding of the aesthetic functions of the ‘lying flashback’ of Stage
Fright. She argues that the film motivates its ‘lying flashback’ by way
of ‘a play with the hermeneutic line and a great deal of conventional
“poetic” imagery related to lying and theatrical deception’ (45). She ex-
amines how the film ‘uses a series of classic techniques [...] to justify the
unconventional device of the lying flashback’ (158). In this process, the
‘narration’s refusal to use conventional types of motivation to cover over
the deception’ means that it ‘acknowledges its own workings and [...]
defamiliarizes them’ (161).
Cognitivism thus understands defamiliarization as the process of
‘making strange’ those conventions that through repetition have become
automatic or naturalized. Consequently, for the ‘lying flashback’ to have
the effect of estrangement the classical conventions of the flashback must
have been internalized by the spectator to some degree. Defamiliarization
can, therefore, be related to the processes whereby audiences naturalize
cinematic conventions, which involve an element of forgetting (cf. Chatman
1978, 49). Narrative conventions in this conception have the potential to be
experienced as innate and ‘natural’. In relation to Stage Fright, the ques-
tion of whether viewers accept Jonathan’s story as true fabula information
or not depends on the individual spectator’s ability to both internalize
and denaturalize (classical) cinematic conventions. With this in mind, it
will be argued that cognitivism restricts defamiliarization to culturally
variable conventions, while it maintains that the linear, cause-effect mode
of organizing experience is natural and innate. Thereby, cognitive formalists
‘forget’ the conventional nature of the linear mode of experience, which
we have termed ‘linear cinematic perception’. When Bordwell argues that
certain conventions are ‘cross-cultural’, the universality of causal-linear
rationalizations is often implied. The aim of the following is to take the

concept of defamiliarization beyond its cognitive limitations to include
the linear cinematic perception and cognition that informs the cognitive
formalism of Bordwell, Thompson, et al.
‘Neoformalism’, as Kristin Thompson (1988) declares, ‘does offer a series of
broad assumptions about how artworks are constructed and how they oper-
ate in cueing audience response’ (6).9 Where the previous chapter explored
the philosophical assumptions of cognitive science (cf. Chapter II), this
part will examine the role these play in the cognitive-formalist concept of
defamiliarization. First, Thompson insists on a clear distinction between
the realm of art and that of everyday perception. Artworks can defamiliarize
perception exactly because they do not belong to the realm of everyday
perception, which is governed by practical purposes:
For neoformalists, then, art is a realm separate from all other types of
artifacts because it presents a unique set of perceptual requirements. Art
is set apart from the everyday world, in which we use our perception for
practical ends. We perceive the world so as to filter from it those elements
that are relevant to our immediate actions. […] Our brains have become
well adapted to concentrating on only those aspects of our environment
that affect us practically; other items are kept peripheral. Film and other
artworks, on the contrary, plunge us into a nonpractical, playful type
of interaction. They renew our perceptions and other mental processes
because they hold no immediate practical implications for us (8).
Thompson’s contention that perception operates as a filtering mechanism

resonates with Deleuze’s Bergson-inspired film-philosophy. Bergson (1991)
famously proposed that perception adds nothing new to the image, rather
‘it results from the discarding of what has no interest for our needs, or more
generally, for our functions’ (38). Therefore, he saw perception as a ‘neces-
sary poverty’ (38) and in this sense consciousness arises from a suppression
of what is of no interest – and a conservation of what does interest – our
bodily functions (cf. Lawlor & Leonard 2013).
James Curtis (1976) has remarked upon the intimate relation between
Bergson and the Russian formalists, and considers ‘Shklovsky the foremost,
and certainly the most energetic, proponent of the Bergsonian paradigm
in Russia’ (115). In particular, Curtis points out Shklovsky’s Bergsonian idea
that the goal of art is to give a feeling of the thing as seeing, not recognition.
From this it follows that art is a device for ‘estrangement’ (ostranenie) that
increases the difficulty of duration and perception (115).
Therefore, the basic idea of defamiliarization is not foreign to Deleuzian

film-philosophy. When Deleuze criticizes the movement-image for imitat-
ing habitual perception, he actually criticizes it for not defamiliarizing
natural perception. Deleuze (2005b) argues:
On the one hand, the image constantly sinks to the state of cliché:
because it is introduced into sensory-motor linkages, because it itself
organizes or induces these linkages, because we never perceive eve-
rything that is in the image, because it is made for that purpose [...].
Civilization of the image? In fact, it is a civilization of the cliché where
all the powers have an interest in hiding images from us, not necessarily
hiding the same thing from us, but in hiding something in the image.
On the other hand, at the same time, the image constantly attempts to
break through the cliché, to get out of the cliché. There is no knowing
how far a real image may lead: the importance of becoming visionary
or seer. (20)
Like Thompson, Deleuze understands perception in Bergsonian terms as a

kind of filtering mechanism of that which does not interest our quotidian
purposes. In this manner, everyday perception becomes habitual, familiar,
and ‘automatized’ (cf. Bergson 1991; Deleuze 2005a, 2005b). Therefore,
Thompson sounds almost Bergsonian, and hence Deleuzian, when she
quotes Shklovsky:
The purpose of art is to impart the sensation of things as they are

perceived and not as they are known. The technique of art is to make
objects ‘unfamiliar,’ to make forms difficult, to increase the difficulty
and length of perception because the process of perception is an aes-
thetic end in itself and must be prolonged. (Shklovsky in Thompson
1988, 11)
A similar desire to go beyond the limits of everyday human perception

is at the heart of Deleuze’s philosophy and is a key to understanding the
time-image. Both have the ability to
open us up to the inhuman and the superhuman (durations which are

superior or inferior to our own), to go beyond the human condition: This is
the meaning of philosophy in so far as our condition condemns us to live
among badly analyzed composites, and to be badly analyzed composites
ourselves. (Deleuze 1991, 28)
The polarized positions of Deleuzian film-philosophy and cognitive formal-

ism rarely seem so intimately akin, as they do in relation to the concept of
defamiliarization.
Nevertheless, the important difference is that Thompson (1988) insists on a
strict line separating art from ‘real life’. This is the case when she argues, ‘our
nonpractical perception allows us to see everything in the artwork differently
from the way we would see it in reality, because it seems strange in its new
context’ (10). According to this logic, defamiliarization only exists because
art occupies a different realm to everyday perception. This conflicts with
Bergson’s concept of multiplicity according to which the immediate data
of consciousness are a multiplicity, and therefore cannot be ‘represented’
by a unified consciousness (cf. Lawlor & Leonard 2013, sec. 2). From this,
Deleuze (2000) advances the idea that we must ‘understand cinema not as lan-
guage, but as signaletic material’ (361). From this perspective, ‘reality’ cannot
be separated from the ‘brain’ that perceives it. Furthermore, since cognitive
formalism believes our comprehension of art to be guided by representations,
it takes little interest in the way cinematic material affects us directly.
Thompson understands perception broadly as a simplif ied formula
according to which the primary function of emotions is cognitive. Thus,
permeating the cognitive-formalist version of defamiliarization is that a
higher level of awareness of everyday ‘automatized’ cognition and perception
should be obtained before defamiliarization can claim a meaningful effect.
Consequently, higher-level cognitive processes are essential to the cognitive-
formalist conception of defamiliarization. Thompson (1988) explains:
For the neoformalist critic, conscious processes are usually the most
important ones, since it is here that the artwork can challenge most
strongly our habitual ways of perceiving and thinking and can make us
aware of our habitual ways of coping with the world. (27)
Therefore, Thompson utilises the concept of defamiliarization to understand

Stage Fright at a purely formal level, in a manner that pays little interest
to the viewer’s cognitive, emotional, or affective experience.
However, defamiliarization has the power to bring to consciousness
otherwise non-conscious or unattended processes. Because perception and
cognition in relation to cinema are applied for non-practical ends, Bordwell
(1985a) argues that in cinema
what is nonconscious in everyday mental life becomes consciously at-

tended to. Our schemata get shaped, stretched, and transgressed: a delay
in hypothesis-confirmation can be prolonged for its own sake. And like

all psychological activities, aesthetic activity has long range effects. Art
may reinforce, or modify, or even assault our normal perceptual-cognitive
repertoire. (32)
In short, cognitive formalism believes that art challenges our cognitive

and perceptual apparatus by transgressing certain established norms. In
particular, art has the ability to challenge psychological capabilities such
as recognition, comprehension, inference making, and the application of
schemata. In this fashion, the cognitive tasks that artworks encourage us
to perform can potentially extend into everyday perception.
Since cognitive formalists understand narrative comprehension in terms
of a ‘problem-solution’ model of cognition (cf. Bordwell 1985a), a central
question concerns the true nature of fabula events. For Thompson (1988),
the defamiliarization of Stage Fright works mainly because it eventually
settles on an unambiguous version of the fabula. This enables the audience
to reconstruct their mental engagement with the film and to localize their
erroneous inferences retrospectively. She believes the process whereby
viewers evaluate and reflect upon their cognitive construction of the story
to be comparable to mental exercise:
The nature of practical perception means that our faculties become

dulled by the repetitive and habitual activities inherent in much of daily
life. Thus art, by renewing our perceptions and thoughts, may be said to
act as a sort of mental exercise, parallel to the way sport is an exercise
for the body. Indeed, individuals’ use of artworks is often comparable to
their use of non-exercise games – chess, for example – and to the aesthetic
contemplation of nature for its own sake. Art fits into the class of things
that people do for re-creation – to ‘re-create’ a sense of freshness or play
eroded by habitual tasks and the strains of practical existence. Often
the renewed or expanded perception we gain from artworks can carry
over to and affect our perception of everyday objects and ideas. As with
physical exercise, the experience of artworks can, over a period of time,
have considerable impact on our lives in general (9).
Nowhere is the core difference of cognitive formalism and Deleuzian

film-philosophy as pronounced as with the above-cited chess metaphor.
Whereas Deleuze conceptualizes the cinematic experience in terms of a
direct encounter, cognitive formalism perceives the spectator in terms of
a disengaged ‘information-processing system’.
Computational cognitivism is here expressed with reference to an

abstract logical game that requires its player to perform highly complex
cognitive tasks according to a ‘problem/solution’ – or ‘question/answer’
(cf. Carroll 2008, 138-139) model. Following this view, spectators continu-
ously test different hypotheses based on a cluster of knowledge about the
rules of cinema until they find the key for solving or answering the questions
proposed by the film. In doing so, spectators will note the rules of the
game and take notice when deviations from the rules occur. Once exposed
to a particular deviation from the rule a sufficient amount of times, the
spectator will integrate this as a new rule.
The advantages of the computer metaphor in accounting for our
hypotheses testing and schemata applications are evident. However,
Sinnerbrink (2011) has argued that this model encounters problems when
accounting for the pleasures that spectators experience once confronted
with narratives that upsets our expectations. He argues:
Cognitivist theories assume that it is our intellectual satisfaction in

reconstructing meaning and solving narrative puzzles that accounts
for our pleasure in a film. On this view, however, it becomes difficult to
explain our aesthetic delight in being misled or deceived by a work of art.
Why do we enjoy narrative deception? If our engagement with narrative
film were primarily about processing and resolving narrative puzzles,
one would expect to experience displeasure at having one’s desire for
cognitive closure thwarted. (52)
In her analysis of Stage Fright, Thompson does not account for the seeming
paradox that some viewers experience pleasure once deceived, but instead
focuses on the reports about the outbursts upon the film’s original theatrical
release. The reason, I believe, is that cognitive formalism has constructed its
theory of narrative comprehension upon a linear conception of the classical
film and the classical spectator.
As we have seen, one of the guiding assumptions of cognitivism is that
the central cognitive goal of viewers is the construction of a more or less
unified and coherent fabula (cf. Bordwell 1985a, 33). This provides an acute
framework for theorizing how spectators are active in structuring the nar-
rative according to a causal-linear string of events that retrospectively form
the fabula. However, it has proven less apt for examining the openness,
unpredictability, and insecurity about the direction the narrative will take,
which equally shapes the cinematic experience. This is partly because
cognitivism presupposes a restricted understanding of the spectator as
being deeply rooted in the conventions of classical storytelling. For cogni-

tive formalism, classical cinema forms the ‘background’ against which all
other cinematic modes can be understood and weighted. Thompson (1988)
explains that ‘neoformalism calls norms of prior experience backgrounds,
since we see individual films within the larger context of such prior experi-
ence’ (21, emphasis in original).
Thompson accounts for three basic types of ‘background’. First, there
is the ‘real world’, without a knowledge of which we would not be able to
recognize referential meaning (e.g. character traits or analogies to society).
This introduces the basic assumption of the cognitive-formalist use of the
concept of the background, namely that there is a strict division between
object and subject and between the external world and the perceiver. From
this follows the second, which is our knowledge of artworks’ functions,
and the third, which is a differentiation between art and other types of
information (e.g. commercials). Thompson acknowledges the implication of
using the classical cinema as the all-encompassing background. However,
she justifies doing so since it would be impossible to fully reconstruct
the original viewing circumstances of most films and since the original
background of a set of spectators upon release can never be restored (21-22).
In relation to the study of early cinema, Thompson concludes that ‘critics
and historians must analyze these early films against the background of
later, classical filmmaking’ (22, emphasis added).
In this manner, the analysts risk confusing the classical film with that
of films more generally. Consider, for instance, when Thompson writes:
Much of our reaction to stylistic devices may be preconscious in that we

learn cutting, camera movement, and other techniques from classical
films, and we learn them so well that we usually no longer need to think
about them, even after only a few visits to the cinema. (27, emphasis
added)
From the perspective of early cinema, Tom Gunning (1986) has famously
addressed the problems that arise from taking classical narration as the
background for understanding other narrational modes. He writes:
The history of early cinema, like the history of cinema generally, has
been written and theorized under the hegemony of narrative films. Early
filmmakers like Smith, Méliès, and Porter have been studied primarily
from the viewpoint of their contribution to film as a storytelling medium,
particularly the evolution of narrative editing. (63)
However, as Gunning elaborates, ‘every change in film history implies

a change in its address to the spectator, and each period constructs its
spectator in a new way’ (68).
In elaboration of Gunning’s critique, it can be argued that the device
of defamiliarization not only reveals the historical nature of cinematic
conventions, it also carries the potential to reveal something about the
analyst as a perceiver situated in a specific historical and cultural context.
Defamiliarization, as Frank Kessler (2010) has argued,
is always ‘in the eye of the beholder,’ and by this very token it reveals itself
as an inherently historical phenomenon. Any defamiliarizing device is
bound to turn into a habitualized one as time goes by, so to the read-
ers or viewers of later generations, it may indeed appear as an utterly
conventional feature. (78)
That cognitive formalism settles on the classical cinema as background is

certainly no arbitrary choice, because this background is perceived to have
a naturalistic justification. As such, classical conventions comply with the
quotidian manner in which perception and cognition work. Following Kes-
sler, ‘[t]here is a risk, however, that a background/foreground constructed
in that way becomes “automatized”’ (65). That is, even though the cognitive
choice of using Hollywood as background is both an obvious and convincing
one, it ‘might also block an understanding of other logics at work in the
construction of a given film, or group of films’ (65).
Therefore, it is important to recognize the concept of defamiliarization
beyond its cognitive-formalist restriction to linearity. In doing so, defamil-
iarization can prove a valuable concept for examining the process whereby
not only narrative conventions are internalized, but also how classical
cinema can naturalize and denaturalize particular modes of perception. In
order to do so, we must ‘never cease to defamiliarize defamiliarization’ (79),
and thus the concept should be made to include those linear processes that
cognitivism perceives as natural and innate. A first step is to challenge
the cognitive claim that classical cinematic conventions have a privileged
relation to natural perception.
Cognitive formalism proposes that f ilm studies should take a more
positivistic or scientific approach that is founded on neither metaphysics
nor epistemology:
Contrary to what many believe, a study of United Artists’ business

practices or the standardization of continuity editing [...] need carry no
determining philosophical assumptions about subjectivity or culture, no

univocal metaphysical or epistemological or political presumptions – in
short, no commitment to Grand Theory. (Bordwell 1996a, 29, emphases
in original)
As we have seen, cognitive formalism perceives cinematic spectatorship

in terms of a form of mental exercise in which our senses are sharpened.
This emphasizes the cinematic experience as a puzzle where spectators
are encouraged to gather the pieces and put them into a coherent and
chronological order. This modus operandi works best if a logical order can
be deduced from the pieces of the film. Therefore, a manner in which to
test Bordwell’s epistemological commitments is against a type of cinema
that does not comply with the rules of linear storytelling.
Bordwell’s (2006) brief treatment of David Lynch’s Lost Highway (1997)
– a film about a man’s psychological breakdown in which the linear logic
dissolves – may help to shed light on this question. Because Bordwell as-
sumes the linear epistemology of classical cinema to be normal and natural,
his depiction of Lost Highway appears to miss the point of the film:
If complex storytelling demands high redundancy, Lynch has been

derelict in his duty. [Lost Highway’s] phantasmagoric body-switches
occur without explanation in a milieu soaked in dread and threatened
violence. The eerie mix of horror-film atmospherics and radiant naïveté
may urge us to construct each as presenting the fantasies of a possessed
protagonist, but the cues are not nearly as firm as they are in A Beautiful
Mind [Howard 2001]. Instead, the absence of definite reference points al-
lows Lynch to rehearse a few obsessive scenarios of lust and blood without
settling on which are real and which are imagined. (89, emphases added)
Lost Highway does not provide the viewers with a puzzle to be solved; nor
is it interested in the kind of defamiliarization found in Stage Fright. Yet,
Bordwell seems mistakenly to assume that Lost Highway is trying, but
failing, to be a conventional Hollywood film. Instead, it could be argued that
what is explored in the film is the question of what happens to our concept
of logic when linearity dissolves (cf. Hven 2010). Although defamiliariza-
tion is the cognitivists’ best analytical tool for understanding cinematic
transgressions, its ultimate task is the reintegration or reorganization of
cinematic elements that do not apply to the linear norm.
In his criticism of the inferential model of narrative reconstruction
developed by Bordwell (1985) and modified by Buckland (2003), Miklós
Kiss (2010) revisits an often-discussed scene from Jean-Luc Godard’s

Pierrot le fou (1965). In the words of Buckland (2003), the director in this
particular scene ‘dismantles a well-formed sequence and then rearranges
the shots into a new order, to express the characters’ sense of panic and
haste’ (126). Kiss (2010) convincingly argues that ordering the non-linearity
of the syuzhet into a linear chain does not guarantee the facilitation of the
episode’s comprehensibility (170). Not least because Godard’s intention
with the disjunctive narrative of the scene is not the creation of a puzzle
or a mystery to solve.
Instead, Godard manages to enhance the scene with a ‘realistic effect’
that ‘works on a physiological rather than an inferential level’ (171). Even if
we accept the somewhat dubious claim that physical reality is linear and
chronological, Kiss reminds us that this still does not alter the fact that ‘our
perceptual experience of this reality is non-linear, achronological’ (170).
From the inferential perspective of Bordwell and Buckland, defamiliariza-
tion thus turns into a ‘refamiliarization’ of the non-linear qualities of the
film, which must be reverted back into its presupposed original and natural
linear state. For Kiss, however, ‘[t]he heart of the matter is not to solve some
riddle, to understand what is represented, but what matters is the alteration
of sensation’ (171).
Similarly, Daniel Frampton (2006) argues that in order to comply with the
cognitive framework transgressive cinematic elements must be rationalized.
In Bordwell’s cognitive film science, he writes, ‘radical cinema is reduced
to principles, systems, all towards trying to bring artistic cinema into the
rational fold of classical cinema’ (104, emphasis in original). In a similar
manner, defamiliarization for cognitivism is a concept capable of making
us aware of the conventional nature of cinematic conventions such as the
flashback. The problem is that the various modes of cinematic transgression
are always evaluated against the stable background of linear cinematic
perception. Therefore, it can be argued that the function of defamiliariza-
tion within the cognitive framework is paradoxically to refamiliarize the
elements that have been defamiliarized. That is, the concept is used to
evaluate variable cinematic conventions against a stable, culturally, and
historically non-transforming, naturalistic background, which is informed
by the linear onto-epistemological roots of cognitive science. Unlike Stage
Fright, Lost Highway and Pierrot le fou do not invite viewers to ‘refa-
miliarize’ the transgressed cinematic elements to make them conform to
the linear onto-epistemology.
Stage Fright constructs its spectator in a classical manner that simul-
taneously reflects upon the double procedure of learning and forgetting
involved in the comprehension of cinematic narratives. However, rather

than simply confirming the classical viewpoint the film holds the potential
to allow us to reflect upon how cinema cues us to construct a particular
‘reality’. As such, it renders visible how cinematic conventions acclimatize
us to specific notions of reality (cf. Nadel 2005, 428; cf. Chapter II). Therefore,
cinematic conventions do not just rehearse an influence on the manner in
which cinema is experienced, but also participate in creating the ‘recipes’
whereby we structure everyday experience. In reference to the work of
Bruner (1991, 2004), it can be argued that the linear mode of experience is
not more or less natural than the non-linear leaps associated with memory,
dreaming, or the stream of consciousness. However, the invention of the
cinematic apparatus has contributed to enforcing and establishing the
linear mode of perception as the most natural mode of experience. From this
perspective, the cognitive theory of spectatorship is a further affirmation
and upheaval of this particular mode of experience. This should neverthe-
less not cause us to devalue cinematic linearity (as it can be argued Deleuze
does), rather it should allow us to become more aware of how the various
dimensions of the image interact to form a complex cinematic experience.
While cognitive film studies have enriched our understanding of the linear
dimension of cinema, it is important to insist that this be combined with
approaches capable of disclosing the non-linear dimensions of the cinematic
experience.
This brings us back to the cognitive-formalist uneasiness with ambi-
guities. Although the modern(ist) cinema is explored in the following
chapter, I would already here like to refer to Bordwell’s understanding of
this cinematic regime, since it demonstrates some of the risks involved
in posing the classical cinema as the background against which all other
narrational modes can be understood. In short, Bordwell (1985a) argues
that ‘art films’ (and ‘parametric narration’) do not mainly rely on a direct
relationship to reality or psychological processes. Instead, this paradigm
breaks with ‘contingent universals’ to introduce ambiguity. For Bordwell,
narrative ambiguity serves only one purpose: ambiguity. In a deliberately
crude passage he states that ‘the procedural slogan of art-cinema narration
might be: “interpret this film, and interpret it so as to maximize ambigu-
ity”’ (212). According to John Mullarkey (2009), Bordwell’s conception of
ambiguities in art cinema reveals his limited, linear conception of reality:
For a science of film such as Bordwell’s […] ambiguity in film cannot be

realistic because reality really is clear cut. If there is any ambiguity in the
film, it must be because the movie is saying something about itself. (35)
The concept of defamiliarization demonstrates how cognitive-formalism

relies on a systematic exclusion of non-linear elements (ambiguities, non-
chronological temporal flows, the cinematic experience, etc.) that threaten
to challenge the core assumptions of cognitive science (cf. Chapter II).
However, the concept of defamiliarization once opened up onto the
non-linear dimension of cinema, may prove useful for understanding
how contemporary complex cinema takes us beyond the restrictions of
a Cartesian-Newtonian space-time, given that continuity in these films
no longer plays the central role for organizing cinematic experience
(cf. Shaviro 2010, 2012). Due to its ambiguities and willingness to challenge
automatic cognition, Stage Fright challenges the core mechanisms of
linear cinematic perception. Therefore, the film can be seen as an important
precursor of how contemporary cinema increasingly denaturalizes linear
cinematic perception, for instance by demonstrating the conventional
nature of causal-linear reasoning. Cinema is always involved in a dual
process of constituting and denaturalizing particular modes of organizing
experience. It is in relation to this that not only narrative but also cinema
is a contributing factor to how we organize experience.
Therefore, it shall later be argued that our experience of films is not based
on backgrounds or mental representations, but instead form an ‘assemblage’,
or a direct encounter, of ‘world-spectator-film’ (cf. Brown 2013; cf. Chapter V
& Chapter VII). As such, cinema cannot claim to depict or ‘represent’ an
objective reality. Instead, it can provide us with perceptions that open new
pathways for sensing, experiencing, and thinking. Modern(ist) cinema can
be seen as a revolt against the linear and representative foundations of
cinema, and as an opening up for new ways of conceptualizing the relation
between cinema and spectator.
IV. Modern(ist) Cinema: Logic of the
Encounter
Once there is no longer a fabula to interpret, once we have no stable point of

constructing character or causality, ambiguity becomes so pervasive as to be of
no consequence.
‒ Bordwell 1985a, 233
We run in fact into a principle of indeterminability, of indiscernibility; we no

longer know what is imaginary or real, physical or mental, in the situation, not
because they are confusing but because we do not have to know and there is no
longer even a place from which to ask.
‒ Deleuze 2005b, 7
Introduction
One of the main claims of Deleuzian film-philosophy is that ‘[t]he movement-

image of the so-called classical cinema gave way, in the post-war period, to
a direct time-image’ (Deleuze 2005b, xi). The present chapter examines the
time-image of the modern(ist) cinema through a study of Alain Resnais’s
Hiroshima mon amour (Hiroshima My Love 1959), which is emblematic
of this Deleuzian image-regime. In relation to this, Patricia Pisters (2011a)
has observed that the film ‘is a crystal of time, which gives us the key to the
time-image in general’ (102; cf. Deleuze 2005b, 67). This is particular evident
in the film’s exploration of the temporality that occurs ‘in-between’ events,
which opens up a ‘Borgesian labyrinth of time’ full of virtual pathways.1 For
Deleuze, the bifurcation of causal linearity in the time-image is intimately
connected with the rupture of the sensory-motor linkages that constituted
the movement-image. The cinema of Resnais contains prime examples of
how non-linearity becomes the ‘soul’ of the medium in post-war cinema
for Deleuze (2005b), for whom the time-image emerges from cinema’s own
Kantian revolution in which ‘the subordination of time to movement was
reversed’ (xi). Questions of temporality are for Deleuze by nature philo-
sophical questions, because ‘time has always put the notion of truth into
crisis’ (126). In this manner, the opposition between the sensory-motor
movement-image (chronological, causal, action-response, representational,
physical, etc.) and its breakdown in the time-image (sensory-motor ruptures,
Bergsonian durée, encounters, anti-representationalism, mental landscapes,

etc.) is deeply philosophical.
For Bordwell, on the other hand, as Oliver Speck (2010) observes, ‘art cinema
appears to be a genre whose primary characteristic is the creation of ambigu-
ity’ (50). According to Bordwell, the modern(ist) formal ‘ambiguities’ are justi-
fied by a new vraisemblance. Since all narrational paradigms use different
strategies to represent the ‘real’, Bordwell is unwilling to grant any of these
strategies precedence. In this manner, he eschews the onto-epistemological
scepticisms raised in the modern(ist) cinema against the linear paradigm of
thought. Bordwell instead aims to demonstrate the usefulness of his cognitive
theory of narrative comprehension to understand the ‘game of form’ that
modern(ist) films play with the spectator. It will be argued that this theory
may go some way to explain modern(ist) spectatorship, but ultimately fails
to incorporate the non-linear cinematic dimensions of these films.
It could seem that the major difference between cognitive film science
and film-philosophy is that the former focuses on questions of narration
and the latter on temporality. However, temporal questions are always also
questions of narration, even when they take the form of narrative scepticism.
Addressing the influence of Hiroshima mon amour, Kovács (2007) remarks:
Three years had passed since the release of Hiroshima, My Love and it
became common practice, almost compulsory, for a modern filmmaker
of the time to merge past and present and make reality and fantasy indis-
cernible. Hiroshima’s novelty of course was not the flashback technique,
but that the memories evoked in the film were not associated with a
well-defined story line with a beginning and an end (320-321).
As Hiroshima mon amour brings questions of temporality, philosophy,

narration, memory, and history together it not only lends itself especially
well to film-philosophical treatments, it also becomes central for under-
standing the film-historical position of contemporary complex cinema.
This is the case, since modern(ist) cinema changed the linear foundations
of cinematic spectatorship.
Cinema and the Modern
It appears to be commonplace to initiate any treatment of Hiroshima mon

amour with the anecdote of how Alain Resnais was first commissioned
to do a documentary on Hiroshima as it looked in the aftermath of the
Modern(ist) Cinema: Logic of the Encounter 85
atomic bomb in 1959; an idea he allegedly spent months considering before

finally abandoning. According to the anecdote Resnais already had realized
with Nuit et brouillard (Night and Fog 1955) – a documentary about
the horrors rehearsed in the Nazi concentration camps – ‘that one cannot
document a memory’ (Kreidl 1978, 54). The director discussed his struggles
with friend Marguerite Duras – one of the leading figures of the nouveau
roman movement – who eventually agreed to write a screenplay for the
film about Hiroshima. However, the screenplay was not going to be ‘about’
Hiroshima in any traditional sense. As Duras (1961) poetically expresses
her thoughts on writing the script for the film: ‘Impossible to talk about
Hiroshima. All one can do is talk about the impossibility of talking about
Hiroshima’ (9).
One may ask why this anecdote continuously appears when Resnais’s and
Duras’s modern(ist) landmark is discussed. The reason, it seems to me, is
that this anecdote in a single and very concise paradox – ‘representing the
unrepresentable’ – captures the crisis of representation that characterizes
not only Hiroshima mon amour, but the movement of modern(ist) cinema
by and large. Why did Resnais not feel able to tell the ‘story’ about Hiroshima
in the genre of the traditional documentary?2 In an interview that appeared
in the television programme Cinepanorama from 1961 about L’Année
dernière à Marienbad (Last Year at Marienbad 1961), Resnais explains
that he does not think it is the role of the director to present stories that
can be understood as explanations or solutions. He clarifies this position
as follows:
It’s not my role to give explanations. For that matter I don’t think [Last
Year at Marienbad] is a real enigma. By that I mean that each spectator
can find his own solution, and it will in all likelihood be a good one.
But what’s certain is that the solution won’t be the same for everyone,
meaning that my solution is of no more interest than that of any other
viewer. (Resnais 2008, par. 3)
If Resnais’s notion of the relations of film, director, and spectator is contrasted

with how these have generally been conceived in classical cinema, three
interconnected differences can be deduced. Resnais is pointing to 1) a crisis
of representation or of not allowing spectators to construct an intersubjective
fabula (‘the solution won’t be the same for everyone’). From this 2) a renewed
kind of spectatorship (‘each spectator can find his own solution’) is induced,
which in turn relies on 3) a renewed concept of ‘truth’ (‘my solution is of
no more interest than that of any other viewer’). In this manner, Resnais’s
comments acutely sum up how modern(ist) cinema revises questions of

history, memory, temporality, cinematic representation, and epistemology.
After the release of Hiroshima mon amour in 1959 a series of critics and
filmmakers – including Jean-Luc Godard, Jacques Rivette, and Eric Rohmer
– held a round-table discussion at the Cahiers du cinéma. This discussion took
place at a time when Godard was preparing À bout de souffle (Breath-
less 1960), and Michelangelo Antonio L’avventura (1960); both films are
today considered landmarks within modern(ist) cinema. It is interesting
to note the state of simultaneous excitement and uncertainty about the
future direction of cinema among these cineastes, who could not know that
new waves were soon to spread throughout (European) cinema. One thing,
however, was certain: Hiroshima mon amour had nurtured a hope that
cinema was about to enter a modern period. Rohmer addresses this as he
entertains a thought experiment concerning the future impact of the film:
There is no doubt that the cinema also could just as soon leave behind
its classical period to enter a modern period. I think that in a few years,
in ten, twenty or thirty years, we shall know whether Hiroshima [mon
amour] was the most important film since the war, the first modern
film of sound cinema, or whether it was possibly less important than we
thought. (Rohmer in Domarchi et al. 1985, 61)
Whether Hiroshima mon amour from a contemporary perspective can

rightly be regarded as ‘the first modern film of sound cinema’ or not is a
problematic question, since no broad consensus about the boundaries of
modern(ist) art cinema exists (cf. Chapter I).
Nevertheless, the Cahiers critics place Hiroshima mon amour firmly
within a modern(ist) context when they claim the film to be a cinematic
companion to Stravinsky’s serial music, to the ‘descriptive’ style of the
nouveau roman, and to the fragmentary style of Cubist painters like Pablo
Picasso (cf. Domarchi et al. 1985). This is partly due to Resnais’s disjunctive
style of editing that breaks with the ‘transparent’ continuity editing of
classical cinema. The director Jacques Rivette declares that ‘montage, for
Eisenstein as for Resnais, consists in rediscovering unity from a basis of
fragmentation, but without concealing the fragmentation in doing so’ (Riv-
ette in Domarchi et al. 1985, 60). The fact that Duras wrote the screenplay
together with the film’s exploration of the themes of history, memory, and
temporality further connect the film to modernist art.
As Kovács (2007) asserts, understanding modernist cinema ‘historically
means understanding how it differs from its counterpart, nonmodern or
classical narrative (art) cinema’ (37). In the history of film theory the di-
chotomy of classical and modern(ist) has functioned as an effective tool for
letting these two narrative paradigms shed mutual light on each other. Such
an approach has its advantages, but a cinematic paradigm cannot mainly
be understood in terms of its oppositions to another paradigm. Bordwell &
Staiger (1985) have complained that ‘theorists usually discuss alternatives
to the classical cinema in general and largely negative terms’ (379). It can
nevertheless be argued that Bordwell (1985a) applies exactly such a strategy,
when he argues that modern ‘art cinema’ has become a coherent style or
‘genre’ due to its deviation from the ‘standard’ classical narrative schema.
According to Bordwell, the decline of the Hollywood dominance after
the Second World War allowed ‘art cinema’ to develop into a coherent
alternative narrational paradigm. Consequently, ‘art cinema’ is intimately
connected to the national movements that spread all over Europe in the late
1950s and the 1960s (such as the French nouvelle vague), and it is associated
with the works of auteurs such as Resnais, Fellini, Antonioni, and Bergman,
among others (cf. Bordwell 1985a, 230). While Bordwell’s cognitivist theory
of ‘art cinema’ is capable of shedding light on certain aspects of ‘art cinema’,
it can be argued that it comes with an unfortunate tendency to downplay
the more non-linear aspects of the films. Consequently, it risks conflating
this narrational paradigm to the core principles of classical cinema.
This conflation is attained by means of two methodological choices
employed by Bordwell (1985a) in his classical scientific endeavour to de-
termine the ‘underlying principles that enable the viewer to comprehend
the film’ (205). First, Bordwell draws on the cognitive-formalist theory of
the background to propose an understanding of modern(ist) cinema as a
series of deviations from the standard mode of film practice (cf. Chapter III).
Second, Bordwell turns to a close examination of Resnais’s La guerre est
finie (The War is Over 1966) to demonstrate that despite the film’s onto-
epistemological divergences from classical cinema, ‘art-cinema narration’
can be subsumed under the linear framework of cognitive film theory.
In his characterization of modern(ist) cinema, Bordwell (1985a) states that
‘art-cinema narration has become a coherent mode partly by defining itself
as a deviation from classical narrative’ (228). For Bordwell this becomes a jus-
tification for his methodological use of classical narration as a background
against which ‘art-cinema narration’ stands out as a coherent narrational
paradigm. As has been argued earlier, the drawback of this method is that
it risks becoming automatic and thereby blocking for our understanding
of other logics at work in the construction of a selection of films (cf. Kessler
2010, 65; cf. Chapter III). In any case, Bordwell conceives of ‘art-cinema’ in
terms of a series of systematized deviations from classical cinema of which

the most important are: a less redundant syuzhet; less generic motivation;
a lack of deadlines; a focus on character rather than plot; a concern with
reaction rather than action; flexible means of expression rather than fixed
conventions; a tenuous linking of events rather than tight causality; more
permanent causal gaps; distortions motivated by ‘psychological’ time of the
sort discussed by Bergson rather than by ‘Newtonian’ time; unreliable rather
than reliable narration; ‘subjective’ rather than ‘objective’; it is ‘non-classical’
in creating permanent narrative gaps and in calling attention to processes
of fabula construction; a far more frequent undermining of norms than is
common in a classical film (cf. Bordwell 1985a, 205-235).
According to Bordwell, these various deviations connect to form a new
vraisemblance distinguishable from that of classical cinema. Bordwell
(1985a) sums up the classical vraisemblance as follows:
For the classical cinema, rooted in the popular novel, short story and well
made drama of the late nineteenth century, ‘reality’ is assumed to be a
tacit coherence among events, a consistency and clarity of individual
identity. Realistic motivation corroborates the compositional motivation
achieved through cause and effect. (206)
This is contrasted with ‘art-cinema narration’, which Bordwell argues takes
its cue from ‘literary modernism’ [and] questions [the classical] definition
of the real: the world’s laws may not be knowable, personal psychology
may be indeterminate. Here new aesthetic conventions claim to seize
other ‘realities’: the aleatoric world of ‘objective’ reality and the fleeting
states that characterize ‘subjective’ reality. (206)
Thus, one way of understanding the differences between classical cinema

and ‘art cinema’ appears to be through an examination of their respec-
tive vraisemblance. Yet, Bordwell emphasizes that the epistemological and
aesthetic differences of classical cinema and ‘art cinema’ amount to nothing
more than two divergent strategies for fictional representation. Bordwell ar-
gues that modern(ist) cinema has too often, and unjustifiably, been perceived
as the more sophisticated depiction of ‘reality’. Against this Bordwell states:
Of course the realism of art cinema is no more ‘real’ than that of the
classical film; it is simply a different canon of realistic motivation, a new
vraisemblance, justifying particular options and effects. Specific sorts of
realism motivate a loosening of cause and effect, an episodic construction

of the syuzhet, and an enhancement of the film’s symbolic dimension
through an emphasis on the fluctuations of character psychology. (206)
Bordwell goes on to argue that many of the stylistics of ‘art cinema’ are used
strategically to communicate the idea that this kind of cinema is superior
to its classical counterpart. He argues that
within some aesthetic positions, ambiguity is what philosophers call a

‘good-making property,’ because they are denotatively unequivocal, while
art films become good because they ask to be puzzled over. Within the
framework of this book, however, ambiguity is only one aesthetic strategy
among many, all of potentially equal interest. (212)
In this manner, Bordwell attempts to disengage himself as an analyst from

the culturally divergent modes whereby classical and modern(ist) cinema
understand their relation to a deeper objective notion of reality. However, in
attempting to take upon him an ‘objective’ and value-free scientific position
in the name of universal rationalism, Bordwell fails to acknowledge the degree
to which his methodological choice of using classical cinema as a background
is already ‘a choice made by a scholar when deciding how to approach an
object’ (Kessler 2010, 78). Modern(ist) cinema may not propose a more ‘real’ or
‘authentic’ representation of reality, but opens up another way of experiencing
reality. However, to understand how this is achieved, the analyst must explore
aspects of the cinematic experience that are not immediately apparent once
modern(ist) cinema is perceived against the background of Hollywood cinema.
Bordwell’s methodological choices bring him to the conclusion that
‘ambiguity’ is the defining narrative device of ‘art cinema’. As has been
argued earlier, the cognitive model is uneasy with cinematic ambiguities
(cf. Chapter III). In relation to modern(ist) cinema, Bordwell (1985a) inter-
prets the focus on narrative ambiguity as an expression of the ‘relativistic
notion of truth’ (212) that governs this narrational paradigm. When this
notion is taken to its extremes, ‘ambiguity becomes so pervasive as to be
of no consequence’ (233). Ultimately, Bordwell uses the concept of ‘ambigu-
ity’ as a grand unifier supposedly capable of determining the underlying
principle that connects the otherwise disparate narrational practices of ‘art
cinema’. Thanouli (2009) formulates Bordwell’s use of ambiguity as follows:
The concept of ambiguity becomes thus a handy tool that, on the one
hand, relieves us of the obligation to f ind determinate answers and
explanations regarding the function of the formal devices, while, on the

other, it helps us unite under the same umbrella term, the art cinema, a
very wide range of narrational possibilities that would otherwise seem
endless and chaotic. (4)
Furthermore, the concept of ‘ambiguity’ serves to encapsulate the relativist

epistemology that Bordwell perceives to be at the heart of ‘art cinema’.
This brings us to the second methodological choice taken by Bordwell
in his study of art cinema narration. In his treatment of La guerre est
finie, Bordwell studies how the film engages the viewer in a cognitive
‘game of form’. Although these formal games do shed light on a particular
aspect of the modern(ist) film, my interest lies primarily in Bordwell’s
methodological shift. In the f irst part of his treatment, ‘art cinema’ is
defined by a catalogue of deviations from classical cinema. However, in
the second part Bordwell appears to be more interested in determining
the similarities of the two paradigms, since these can be perceived as an
affirmation of the ‘cross-cultural’ and linear principles that are constitutive
of all narrative cinema.
The ‘game of form’ that Bordwell (1985a) argues La guerre est finie is
playing with its spectators requires that these must ‘draw on tacit conven-
tions of comprehension characteristic of the art film [...] in order to construct
the fabula and identify the rules unique to this film’s narrational work’ (213).
Bordwell thus presumes the construction of a coherent narrative to be as
central for the modern(ist) film as it is for the classical film. However, the
problems spectators experience in constructing a coherent fabula cannot be
reduced to mere puzzles that can be solved according to a ‘problem-solution’
model, since these films have been designed to explore the shortcomings of
the very linear sense-making that constitute this ‘problem-solving’ type of
spectatorship. Nevertheless, Bordwell is neither interested in an examina-
tion of what happens when ‘art cinema’ continuously disrupts our cognitive
efforts to render the narrative continuum linear, nor in what arises from
the dissolution of classical narrative principles.
The reason, I believe, relates to the fact that cognitivism, as Bord-
well (1989) has explained, is ‘in general more concerned [than the Freud-
ian framework] with normal and successful action’ (12). This indicates
that cognitive film science primarily focuses on those aspects that can
be directly linked to the linear dimension of cinema. As such, ‘cognitive
theory wants to understand such human mental activities as recognition,
comprehension, inference-making, interpretation, judgement, memory and
imagination’ (13). It is, however, worth noting that cognitive film science
does not share its exclusive focus on ‘successful actions’ with cognitive
neuroscience more generally. Much of the scientific progress in these fields
results from studies of brains that – for various reasons – do not behave
according to the ‘normal’ or standard brain assumed by cognitive film
scientists.3
Furthermore, when Bordwell regards the ‘relativistic notion of truth’
found in ‘art cinema’ to be merely an aesthetic strategy, he is simultaneously
attempting subtly to disarm the philosophical claim behind this strategy.
Unfortunately, Bordwell in this manner avoids a genuine confrontation
with the epistemological concerns of ‘art cinema’. That would be interest-
ing since modern(ist) cinema challenges some of the core principles of
cognitive (film) science – one of which is the philosophical concept of
representation (cf. Chapter II). In the beginning of Narration in the Fiction
Film (1985a), Bordwell states that ‘we can treat narrative as a representation,
considering the story’s world, its portrayal of some reality, or its broader
meanings’ (xi, emphasis in original). Yet, from the perspective of embodied
cognition explored in Chapter V, a treatment of narrative must also include
the sensory-motor contingencies, bodily affects, posture, and movements
that ‘enter into cognition in a non-representational way’ (Gallagher 2015, 97).
Another Bordwellian assumption that has been challenged by the
embodied approaches is the idea that the brain mainly functions in
terms of rational, linear computation. In an interview cited by Bordwell,
Resnais explains that the editing style of La guerre est finie was
meant to underline ‘the mind’s tendency to leap’ (Resnais in Bordwell
1985a, 219). Resnais is thus challenging the computational metaphor,
suggesting instead that the mind is an intricate network of connections
according to which the subject leaps. However, this should not be taken
as a bold statement about the functioning of the brain, but rather as a
metaphor for the processes according to which thinking is constituted.
As such, Resnais’s statement is critical towards the causal-linear mode
of sense-making that has been associated with the classical cinema, but
which could be equally connected to the linear interpretative strategies
discussed in Chapter III. In relation to this, it is interesting that Bordwell
(1985a) quickly disarms Resnais’s appeal to link his editing style to the
non-linear functioning of the brain. He blankly remarks: ‘[o]ne could just
as easily argue that it is more plausible for the mind to plan its moves in
chronological order’ (219). Luckily we do not need to settle on either, since
it can be argued that the brain is perfectly capable of both. Therefore,
both – although in different manners – are central for understanding
the cinematic experience.
The problem addressed here is that Bordwell adheres to a methodology

that is only capable of capturing the linear aspects of modern(ist) cin-
ema. This means that he ignores, rejects, neutralizes, or downplays those
aspects of modern(ist) cinema that do not affirm his monolithic linear
understanding of the cinematic experience. In this manner, Bordwell can
convincingly argue that La guerre est finie ‘plays a game of gaps with
the viewer’ (219), and engages the spectator in finding the ‘key to the film’s
narrational method’ (221). However, in his insistence on the primacy of a
‘problem-solution’ mode of spectatorship in ‘art cinema’, Bordwell fails to
provide a convincing treatment of what he believes to be the key narrational
device: ambiguity. Ultimately, Bordwell’s answer is that ambiguities exist
in ‘art cinema’ for the sake of ambiguity. Eventually, the linear focus taken
by Bordwell means that the classical and modern(ist) paradigms become
conflated. As such, the ‘ambiguity’ Bordwell initially determined as unique
to ‘art cinema’ is all of a sudden no longer fundamentally different from
the classical film, because ‘the ambiguity of the art cinema is of a highly
controlled and limited sort, standing out against a background of nar-
rational coherence not fundamentally different from that of the classical
cinema’ (222).
Drawing upon the challenges stemming from the dissolution of the
linear onto-epistemology in Hiroshima mon amour, it will be argued
that f ilms that do not primarily facilitate meaning through the con-
struction of a strict line of causality, demand a revised understanding of
cinematic narration. The problem with the cognitive comprehension does
not relate to its constructivist aspect according to which ‘[t]he fabula is
[...] a pattern which perceivers of narratives create through assumptions
or inferences’ (49). Rather the idea that is challenged here pertains to the
conception of the fabula as ‘the developing result of picking up narrative
cues [...]’ (49). In the following, it will be demonstrated how this concep-
tion goes against the narrative logic of Hiroshima mon amour, insofar
as it assumes the film to contain pre-existing and observer-independent
‘cues’ for the spectator to ‘pick up’. As Nick Redfern (2005) observes, Bor-
dwell’s model of cognitive spectatorship assumes ‘that a narrative exists,
albeit only partially, prior to the active processes of construction that
the spectator brings to a film’ (sec. 6). In this manner, Bordwell’s model
intends to uncover, interpret, or represent the inherent meanings that a
given film is ‘really’ communicating (cf. Redfern 2004, 42). As we shall
see, Hiroshima mon amour complicates the assumption that cinema
can communicate inherent meanings in any such straightforward or
linear manner.
Hiroshima mon amour – A Cinematic Love Affair
Hiroshima mon amour revolves around a love affair between a French

actress (Emanuelle Riva), who is in Hiroshima to shoot a film about peace,
and a local Japanese architect (Eiji Okada). These two main characters
remain nameless in the script and are simply referred to as Elle and Lui (or
Her and Him). Their love affair is initiated shortly before the French woman
has to return home to France, where she plans to continue her marital life
as if nothing had happened. This affair, we learn, is one of several (‘I’m very
fond of men’). The Japanese man, too, will return to his wife, and continue
his life as it was before. The French woman experiences the affair as both
intense and impossible, and it therefore brings back memories of a long
forgotten, equally intense and impossible love affair she had with a German
soldier during the occupation in Nevers, France. The two impossible love
affairs, which form the two narrative threads of the film, gradually merge
as present-day Hiroshima and wartime Nevers interweave. Towards the
end of the film, Resnais cuts between images of Hiroshima and Nevers in a
way that emphasizes the gradual convergence of the two places in the film.
Simultaneously the couple re-enacts the events of Nevers, as the Japanese
man embodies the German solder.
Consequently, the script – as Duras envisioned it – does not amount to
the story of Hiroshima told from the perspective of eyewitnesses or expert
observers (a strategy associated with the documentary genre), but rather
from the perspective of
the emotions of a woman of thirty-four who, because the brutal war

evidence at Hiroshima evokes her first love affair fourteen years previ-
ously, confesses the secret tragedy of her youth to a Japanese architect.
Love grows between her and the architect, but she refuses to sacrifice
herself to it. She claims that forgetfulness is a stronger force and that she
will forget him just as she has forgotten her first love – a German soldier
shot on liberation day – and as the people of Hiroshima have subdued
their memory of the atomic explosion. (Duras in Kreidl 1978, 54)
Hiroshima mon amour thus explores the temporality of memory in which

the past is subject to both a collective preserving (through historical facts,
museums, memorial days, sites, documentary f ilms, historic footages,
etc.) and an unforgettable personal trauma. In the beginning of the film,
the actress seeks to ‘understand’ Hiroshima by reducing it to mere facts
(‘200,000 dead and 80,000 wounded in 9 seconds. Those are the official
figures’) and by appealing to its preservation in our collective past (‘four

times at the museum in Hiroshima’). However, already in the film’s opening
act the authenticity of such a perspective is rendered problematic at the
very same time as it is announced.
The 13-minute poetic opening act of Hiroshima mon amour consists
of three separate ontological realms. Rather than inviting spectators to
construct a fabula (or story) out of these disentangled realms, Resnais’s
style of editing is simultaneously seeking an effect of opposition and unity.
As Rivette acutely observes:
It’s a double movement – emphasizing the autonomy of the shot and

simultaneously seeking within that shot a strength that will enable it
to enter into a relationship with another or several other shots, and in
this way eventually form a unity [...] this unity is no longer that of classic
continuity. (Rivette in Domarchi et al. 1985, 61)
The first image of the film (after the opening titles) is that of two lovers – the
Japanese man and the French woman – in the act of lovemaking. This image
of unity in its purest form is disrupted as the bodies are alternately covered
with ashes, dust, and sweat. Therefore, instead of being a sheer depiction of
two bodies in perfect harmony, Resnais presents the viewers with an image
that comprises the oppositions of love/desire and death/destruction. The
image in this sense becomes an encounter between oppositions, like the
combination of the words ‘Hiroshima’ and ‘amour/love’ that resonates from
the film’s title. The couple in the act of lovemaking constitutes the first of
three realms of the opening ‘montrage’. 4
The second realm commemorates the type of documentary footages for
which Resnais was initially commissioned. This realm consists of ‘represen-
tations’ of the horror of Hiroshima, whose role is to depict the ‘collective’
past of the atomic trauma. The French woman is seen as a tourist visiting the
different sites – the hospital, the museum, the tour bus, etc. – of Hiroshima
and strolling around the city, while trying to absorb the immense history of
the place. This realm of the actress, who is in Hiroshima to shoot a movie
about peace, is encumbered with representations, reconstructions, and re-
enactments. Hiroshima is portrayed as a museum piece, and the film’s first
image of the city is a miniature model of how it looked after the bombings
on 6 August 1945.
The third realm exists purely as the off-screen voices of the Japanese
man and the French woman. The language in the dialogue is poetic, and is
initiated by the Japanese man’s negation (‘you saw nothing at Hiroshima,
nothing’) followed by the French woman’s affirmation (‘I saw everything’).

This pattern continues throughout the dialogue. Although the voices are
detached from the other realms, they appear – although in no straightfor-
ward manner – to comment on the images they accompany. The images of
the hospital, for example, are accompanied by the voice of the woman, who
claims: ‘I saw everything. I saw the hospital I’m sure of it.’
The actress insists in the representations’ ability to transmit something
genuine about Hiroshima, although she acknowledges their shortcomings
(‘the reconstructions, for lack of everything’ and ‘the films were as authentic
as possible’). In spite of the inability of the artefacts to replace reality, they
are capable of arousing affects that convey the sensation of ‘experiencing’
the horrors of Hiroshima (‘the illusion, quite simply, is so perfect that tour-
ists weep’). These, however, exist as pure and naked affects deprived of any
deeper sense. This is made clear by the Japanese man’s continuous rejection
of the documentary images’ claim of depicting a ‘reality’. Thereby, the film
constructs a conflicting space between the affective engagement and the
reflective disengagement with the images. In so doing, the film accentuates
vision as both a historical and social construct. In Hiroshima mon amour
the ancient problem of representation finds a unique cinematic expression.
Interestingly, Doane (2002) connects the paradoxes of cinema’s archival
desire to the image of a body no longer capable of ‘seeing’ the present:
[I]t is finally in the new representational technologies of vision – photog-

raphy, the cinema – that one witnesses the insistency of the impossible
desire to represent – to archive – the present. And if a perception of
radical finitude is a condition for archival desire, the longing to grasp
the present in representation finds its basis in an image of a body whose
visual powers are defective, lacking, riven by delay – a body which cannot
‘see’ that present which is so crucial to modernity. (102)
In extension of this, the opening act of Hiroshima mon amour per-

fectly illustrates the epistemological shift in spectatorship undertaken by
modern(ist) cinema. The spectator is no longer perceived to be a fixed and
stable subject, but is instead schizophrenic and centrifugal. As such, the
gaze of the spectator ‘is not based in simple perspectivalism but in a sy-
naesthetic virtuality that constantly negotiates between all the senses and
their respective memories’ (Elliott 2010, 4). The opening act of Hiroshima
mon amour beautifully demonstrates how this is achieved by rupturing the
classical cinematic flow that ties the affective to the cognitive responses
of viewers (cf. Deleuze 2005a).
The poetic opening act of Hiroshima mon amour is an acute cinematic

expression of the crisis and scepticism that surrounded the linear repre-
sentational scheme in post-war Europe. In the opening act, Duras’s dictum
concerning ‘the impossibility to talk about Hiroshima’ is given a voice as
her scepticism towards representation is embodied in the Japanese man’s
repetition of the dictum: ‘You saw nothing in Hiroshima’. The disconnec-
tion of affect from cognition exposes the manner in which the collective
memory of Hiroshima is formed through an ensemble of representations
that work on spectators emotionally (e.g. news footages, museums, me-
morials, miniatures, photographs, eyewitnesses, etc.). Yet, in the context
of the opening act, these representations appear standardized, lifeless,
schematic, and habitual. What is rendered problematic is not the inability
of these images to affect its audience, but rather the presumption that
‘Hiroshima’ is represented in these; an idea that is symbolically conveyed by
the miniature form Hiroshima takes at the museum. Furthermore, Resnais’s
disjunctive style of editing in the opening act no longer works on a principle
of assimilation (linking a chain of actions and reactions), but instead on a
non-linear principle of disparity that connects the images in a plethora of
ways. In this sense, the spectator’s main cognitive goal is no longer that of
constructing a coherent and causal-linear logic. Thus, the scene expresses a
non-representational concept of thought that resonates well with Deleuzian
film-philosophy.
In relation to Hiroshima mon amour, Resnais has stated: ‘Any suffering
is beyond measure. We have brought together these two dramas, these two
traumas, to apprehend them better’ (Resnais in Wilson 2009, 54). In the
film, cinema’s ability to communicate and transmit objective information
is denied. Human knowledge is confined to language and other means of
representation and the ‘true’ world is beyond reach and incommunicable.
Related to this is the argument that classical cinema can be perceived as
a preinterpreted world in which everything is organized to facilitate a
particular meaning, whereas modern(ist) cinema is obsessed with the loss of
classical meanings in a world that no longer lends itself to representational
sense. Along these lines, modern(ist) cinema can be perceived as being
‘obsessed with the loss of the world of classic cinema and [it] constantly
tries to express this loss, either by deconstructing the forms of classicism
or by formally emphasising the loss of meaning’ (Kretzschmar 2002, par. 7;
cf. d’Allonnes 1994).
Since Hiroshima mon amour works on a narrative logic that is not
classical (but could be expressed as a logic of the encounter), it makes
little sense to talk about how it cues its spectators. Its opening act is a
perfect example of the manner in which modern(ist) film challenges the

linear onto-epistemology that still governs prevailing conceptions of narra-
tion, history, representation, and memory. It is due to this epistemological
displacement, which separates the classical and modern(ist) film, that
Oliver Fahle (2005) has argued that it would be meaningless to attempt to
understand the work of directors such as Jean-Luc Godard, Wim Wenders,
Alexander Kluge, Michelangelo Antonioni, and Federico Fellini exclusively
from the perspective of classical narratology (11).
As a manner of capturing this dimension of modern(ist) cinema, Deleuze
invokes a concept of ‘thought’ that is not identical with its use within the
linear onto-epistemology of cognitivism. Since classical film facilitates the
possibility of narrative coherence as a causal-linear story, narrative com-
prehension within this aesthetic regime is associated with the ability to as-
similate new information into schemata, which are pre-existing ‘organized
clusters of knowledge’ (Bordwell 1985a, 31) that ensures the integration of
non-linear elements into the linear order of the fabula. Yet, such a cognitive
‘problem-solution’ process is not what constitutes thought for Deleuze, who,
as Paul Patton (1996) rightly observes, ‘has always maintained a rigorous
distinction between knowledge, understood as the recognition of truths
or the solution of problems, and thinking understood as the creation of
concepts or the determination of problems’ (6). For Deleuze, thought instead
occurs in the ‘hesitant gestures which accompany our encounters with the
unknown’ (8-9). However, this logic is not completely detached from the
linear logic, because it feeds on a criticism of it. Deleuze (2004a) writes:
The conditions of a true critic and a true creation are the same: the
destruction of an image of thought which presupposes itself and the
genesis of the act of thinking in thought itself. Something in the world
forces us to think. This something is an object not of recognition but of
a fundamental encounter. (176, emphasis in original)
In a similar vein, the modern(ist) cinema can be said to confront spectators

with a fundamental ‘encounter’ with classical cinema and how it facilitates
thought. To be more precise, modern(ist) cinema comprises a challenge
to everyday quotidian cognition, and thereby forces spectators to explore
unknown territories of thought. However, the conditions of these territories
vary from viewer to viewer, and thus a linear, narrative trajectory cannot
be predicted in advance.
This does not mean that the modern(ist) f ilm disposes completely
of linear causality, although the linear reorganization of the narrative
continuum is no longer a guarantee for ‘understanding’ the story. Follow-

ing from this is a significant difference in the manner that classical and
modern(ist) cinema facilitate meaning. The movement of thought in the
first is defined by a linear trajectory progressing towards a predefined goal,
and consequently narrative comprehension becomes organized around
(the obstacles towards) constructing a fabula. As has been argued, this
does not justify a restrictive use of linear interpretative methods that are
insensitive to the inherent ambiguities of the classical film (cf. Chapter III).
In the modern(ist) f ilm, linear rationalizations are ultimately deemed
inadequate or insufficient for providing a meaningful appreciation of the
films.
Therefore, Deleuze (1995) argues that modern(ist) time-images stimulate
brain circuits based on ‘more creative tracing, less “probable” links’ (61).
The power of cinema, according to Deleuze, lies in its ability to produce a
shock to thought by communicating vibrations to the cortex whereby the
nervous and cerebral system are touched directly.5 Both cognitive film
science and Deleuzian film-philosophy are interested in the workings of
the brain. However, with respect to modern(ist) cinema it becomes evident
that their theories rely on very different conceptions of the human brain.
In short, cognitivists study the psychology of the brain to derive prin-
ciples of unity and to establish ‘cross-cultural’ universals (cf. Bordwell &
Carroll 1996). In general, they look to the study of the brain to establish
a scientific, standardized, and generalized concept of the viewer, who is
assumed to utilize alternative frameworks of belief or representational
schemata to arrive at culturally divergent conceptions of the narrative
reality (cf. Chapter II). This assumption becomes explicit when Joseph
Anderson in his pivotal contribution to cognitive film science, The Reality
of Illusion (1998), argues:
The viewer can be thought of as a standard biological audio/video proces-

sor. The central processing unit, the brain along with its sensory modules,
is standard. The same model with only minor variations is issued to
everyone. The basic operating system is also standard and universal, for
both the brain and its functions were created over 150 million years of
mammalian evolution (12).
Contrarily, Deleuze conceptualizes the brain in terms of an undifferenti-

ated mass of singularity, for which, as Bellour (2010) has argued, ‘there
is no science of the viewer; there is only the spectator’s thoughts and
experiences’ (92).
According to Deleuze, cinema is not based on representations such as

cognitivists assume. Instead, the cinema provides a fusion of film and
spectator that immediately produces brain circuits that open up for
new ways of experiencing the world. Cinematic spectatorship for him
is, therefore, not a matter of a cognitive reorganization of images into a
rational and chronological representation. Rather cinematic spectatorship,
as Richard Rushton (2009) explains, requires that we are willing to ‘lose
control of ourselves, undo ourselves, and forget ourselves while in front
of the cinema screen’ (53). Deleuze perceives the self-identical Cartesian
subject of cognitivism, who remains unaffected by the cinematic material,
while mentally constructing a story by means of temporal integration and
schemata application, to be an abstract illusion.
Instead, the Deleuzian conception of spectatorship is intimately
connected to processes of ‘desubjectivization’ – this goes some way in
explaining why Deleuzian scholars tend to valorize affect over cognition. As
Rushton explains, it is only when the cinema no longer primarily addresses
us as individual subjects – as it does with the movement-image – that we
are ‘able to loosen the shackles of our existing subjectivities and open
ourselves up to other experiences and knowing’ (53). In short, whereas
the Bordwellian spectator is cognitively mastering the images through
rational organizations, the Deleuzian spectator is constituted in the af-
fective encounter between mind and screen. According to Pisters (2012),
‘spectatorship, in terms of Deleuzian film-philosophy, can best be seen in
terms of being “affected” by “signaletic material” that changes and forms
our subjectivities in an ongoing process’ (31). Furthermore, it can be argued
that when the modern(ist) film disconnects spectators’ affective from their
cognitive responses, the linear, habitual mode of existence where ‘human
beings function through sense-stimulus and motor-response, definable
action and dependable reaction’ (Ashton 2008, par. 5) is destabilized. This
occurs through the rupture of the sensory-motor linkages that constitute
movement-images, as related to a particular mode of thought ‘motivated by
intention, selection, need for use value, action, and relation’ (Ashton 2008,
par. 5; cf. Deleuze 2005b, 38).
Resnais is the modern(ist) filmmaker per se to cinematically explore the
non-linear, pre-subjective, and precognitive venues of spectatorship. In the
cinema of Resnais, says Deleuze (2005b),
[t]he world has become memory, brain, superimposition of ages or lobes,

but the brain itself has become consciousness, continuation of ages, crea-
tion or growth of ever new lobes, re-creation of matter as with styrene.
The screen itself is the cerebral membrane where immediate and direct
confrontations take place between the past and the future, the inside and
the outside, at a distance impossible to determine, independent of any
fixpoint [...] The image no longer has space and movement as its primary
characteristics but topology and time. (121)
From this passage it becomes obvious that for Deleuze, the brain is as much
a model of thought as its progenitor (cf. Elliott 2010, 1). As Deleuze and
Guattari (1987) have famously proclaimed,
[t]hought is not arborescent, and the brain is not a rooted or ramified matter
[...]. The discontinuity between cells, the role of the axons, the functioning
of the synapses, the existence of synaptic microfissures, the leap each mes-
sage makes across these fissures, make the brain a multiplicity immersed
in its plane of consistency or neuroglia, a whole uncertain, probabilistic
system (‘the uncertain nervous system’). Many people have a tree growing
in their heads, but the brain itself is much more a grass than a tree. (15)
Thus, Deleuze perceives the brain as a non-linear and complex network

of possible connections, capable of bringing the past into the present, and
the present into the future. From this it follows that in the rupture of the
sensory-motor linkages of the images, modern(ist) cinema opens up new
ways of experiencing temporality as constituted in ‘the immediate and
direct encounter between spectator and screen’.
In Hiroshima mon amour the past occasionally manifests itself in the
present as a sudden force that breaks the succession of past, present, and
future. After the couple have slept together, the French woman watches the
Japanese man as he is sleeping. Suddenly an image disturbs the woman’s
(and consequently the film’s) perception, as the arm of the Japanese man
for a brief moment becomes that of her dead German lover. It is significant
that this memory-image is not represented through a classical flashback,
since it is not immediately integrated into the narrative flow. At this point,
there has been no mentioning of her German love affair, and spectators can
therefore only retrospectively organize this information into the ongoing
fabula. Thus, initially the hand of the German lover is experienced as a
‘naked’ image deprived of its temporal anchorage. Consequently, this im-
age is closer to a flashback in the psychoanalytical sense, where, as Cathy
Caruth (1995) reminds us, a ‘flashback [...] provides a form of recall that
survives at the cost of willed memory or of the very continuity of conscious
thought’ (152; cf. Wilson 2009, 52).
From this perspective, the ‘flashback’ is not a device that serves the
linear narrative flow; instead, it opens up a non-linear realm attached
to the woman’s subjectivity. Here the memory of Nevers is omnipresent,
even if it is not necessarily ‘recollected’ or ‘recalled’. As such, it reveals how
memory can be present and absent at once. As the French woman explains:
‘Nevers, you see, is the town I dream about most in all the world, is the very
thing I dream about most. At the same time it is the thing I think about
least in all the world.’ Memory here should not be understood in terms of
recollections, but rather as what Deleuze – with reference to the philosophy
of Bergson – calls ‘sheets of the past’.
Following Bergson, memories are continuously integrated into percep-
tion, which causes the impression of time as a linear unfolding. Memories,
in other words, fold into perception before they are experienced as explicit
recollections. The act of reading proves an illustrative example of this
constant enfolding of memories and perception, since we do not read a
word letter by letter.6 For Bergson (1991), ‘[p]erception and recollection,
always interpenetrate each other, are always exchanging something of
their substance as by process of endosmosis’ (67). Elsewhere Bergson
concludes: ‘There is no perception which is not full of memories’ (33). In
action-oriented (or, problem-solution oriented) perception such as de-
scribed in cognitive film science, memory aids the sensory-motor schemata
and functions to organize and filter our perception in accordance with
our immediate interests. The classical flashback conceives of memory in
a similar fashion, since a recollection here (e.g. in the form of a subjec-
tive recollection) disrupts the linear flow of the narrative to reveal new
information that must be cognitively reorganized and integrated into the
fabula.7
On the other hand, a sheet of past is a part of memory, which we do
not have access to in the same way as to a recollection. It is nevertheless a
‘region of time’ from which recollections arise. In Hiroshima mon amour
the image of the German soldier initially takes the form of a sheet of past.
However, as the narrative progresses and the woman starts to recall her past,
the memory transforms into a proper recollection, and in cinematic terms
her memories thus become more aligned with the (classical) flashback. In
the Bergsonian conception of time, memories – unlike recollections – do
not exist as successive temporal layers. The Bergsonian model of the cone
(cf. Fig. 2) illustrates how perception and memory always coexist – although
in various degrees of contraction.8
The memories that are close to point S in Bergson’s inverted cone are
intimately connected to perception (P), where recollection-images are
formed. In the least contracted part of the cone, memory is closer con-
nected to ‘pure’ memory.9 In broad terms, classical cinema is interested in
those memories that occupy the most contracted part of the cone. Thus,
the classical film is interested mainly in memories as recollection-images
that can be depicted as flashbacks. The time-image, on the other hand,
explores the least contracted areas of Bergson’s cone, and it can thus be
said to explore another dimension of memory.
According to Deleuze (2005b), this endows the images with a more
complex and multilayered mode of temporality:
Memory is clearly no longer the faculty of having recollections: it is the

membrane which, in the most varied ways (continuity, but also discon-
tinuity, envelopment, etc.), makes sheets of the past and layers of reality
correspond, the first emanating from an inside which is always there,
the second arriving from an outside always to come, the two gnawing at
the present which is now only their encounter. (199-200)
Moreover, not only subjective memories are able to form sheets of the past.
The traumatic past of Hiroshima forms collective, rather than subjective,
sheets of the past. In this manner, memory relates to history in a dynamic
sense insofar as the past event is continuously being reconstituted in the
present. Therefore, the lovers’ historical memories of Hiroshima are incom-
mensurable and it is not until the French woman shares her own personal
trauma with her Japanese lover that a common ground can be established
between them.10 Deleuze (2005b) writes:
There are two characters, but each has his or her own memory which is
foreign to the other. There is no longer anything at all in common. It is
like two incommensurable regions of past, Hiroshima and Nevers. And
while the Japanese refuses the woman entry into his own region [....], the
woman draws the Japanese into hers, willingly and with his consent, up to
a certain point. Is this not a way for each of them to forget his or her own
memory, and make a memory for two, as if memory was now becoming
world, detaching itself from their persons? (113-114)
As John Rajchman (2010) has pointed out, Resnais’s aim is no longer to

recollect the past in a consciousness, individual, or collective, ‘but on the
contrary, to prevent any such closure within private memory or public
commemoration, showing, rather, the sense in which it is still at work in
the present’ (290). In this manner, the dynamic conception of the past in
Figure 2. Bergson’s Cone of Memory (Bergson 1991, 152)
Hiroshima mon amour amounts to something more than a mere deviation

from the classical reliance on recollection-images.
In a similar vein, the portrayal of temporality found in Hiroshima
mon amour is not that of a series of events tied together causal-linearly.
Such logic is abandoned in favour of an exploration of the temporality
that occurs in the interstice of events. In relation to this, Marie-Claire
Ropars (2007) has observed that time in Hiroshima mon amour is one
of ‘in-betweens’:
Weak times, dead times; the important thing is ‘between acts’, what is
drawn out right across time and remains incomplete – like this intermi-
nable parting in Hiroshima, these sixteen hours of time to kill, during
which the man and woman wander through the city – waiting room, tea
room, river’s edge – leaving and refinding each other without us ever
knowing the why or wherefore of these moments. (par. 4).
In the time in-between, perception opens up vast temporal layers of the past,
i.e. to the least contracted part of the Bergsonian cone. It can, therefore, be
argued that temporality is central to modern(ist) cinema in a conspicuous

Bergsonian form.
Bergson’s cone offers a way of conceptualizing the different functions of
memory and temporality at play in the classical and modern(ist) cinema.
In the first, recollections or flashbacks can be used as a narrative strategy
to withhold or introduce information at moments carefully orchestrated
to elicit emotional response. In the latter, narration is not a causal-linear
unfolding of events but rather a non-linear assembly of such events. Con-
sequently, the spectator’s emotional responses become less orchestrated,
schematic, or predictable. Deleuze (2005b) explains that what brings the
classical cinema into question after the war is the very breakup of the
sensory-motor schema, which means that ‘the viewer’s problem becomes
“What is there to see in the image?” (and not “What are we going to see in
the next image?”)’ (261). However, it is not only affectively but also in terms
of thought that the modern(ist) cinema incites its viewer beyond schematic
sensory-motor assimilation.
Against André Bazin, Deleuze argues that the real innovation of Italian
neorealism was not a more authentic or ‘true’ representation of the real,
but instead a new relation to the sensory-motor system and consequently
to thought:
However, we are not sure that the problem [of neorealism] arises on the
level of the real, whether in relation to form or content. Is it not rather at
the level of the ‘mental’, in terms of thought? If all the movement-images
[…] underwent such an upheaval, was this not first of all because a new
element burst on to the scene which was to prevent perception being
extended into action in order to put it in contact with thought and gradu-
ally, was to subordinate the image to the demands of new signs that would
take it beyond movement? (1)
In the movement-image, images are linked according to laws of ‘association,

of continuity, resemblance, contrast, or opposition’ (265). The sensory-motor
system works on a double movement of integration and differentiation to
form ‘a regime of localizable relations, actual linkages, legal causal and
logical connections’ (123). This narrative mode of organizing thought is
opposed to that of the time-image, which is constituted by ‘gaps, “irrational
cuts”, discontinuity, and differentiation’ (123). For Deleuze, narration is
consequently intimately attached to the sensory-motor system.
With the modern(ist) cinema having often been associated with the
formal device of ambiguity – i.e. in denying spectators a possibility of
constructing a coherent fabula – it has also been associated with non-

narrative cinema, and Deleuzian film-philosophy appears to reaffirm this
impression. However, when closely read it becomes clear that Deleuze
(2005b) rather tries to deduce another type of narration based on the
time-image that gives ‘narration a new value because it abstracts it from
all successive action’ (98). According to this view, the basis of narration in
the time-image has changed from its classical form. In fact, narration, like
representation, becomes the ultimate paradox of the time-image.
Resnais demonstrates an uneasiness with interpretations of Hiroshima
mon amour based upon the construction of a coherent fabula, when he
states ‘the past shouldn’t be a flashback [...] You might even imagine that
everything the Emmanuelle Riva character narrated was false; there’s no
proof that the story she recites really happened. On a formal level, I find that
ambiguity interesting’ (Resnais in Jones 2015, par. 7). In any case, whether
or not the French woman is lying is no longer the fundamental issue of the
narrative. For Deleuze (2005b), this is a significant characteristic of the
time-image, since the indeterminability of the imaginary and the real, the
physical and the mental, etc., is no longer a central concern here (7). To
determine the kind of narration that arises in this blurring of Cartesian
dualisms (imaginary-real; physical-mental), it may be helpful to differenti-
ate this kind of narration from the unreliable narration of Stage Fright.
First of all, the term ‘unreliable’ implies that the film contains an objec-
tive and stable narrative truth. This despite the fact that the narration
systematically obstructs, misleads, and deceives spectators on their path
towards a coherent fabula. Contrarily, Deleuze argues that narration in the
time-image becomes ‘falsifying’. However, this should not be understood in
relativistic terms, as Deleuze (2005b) makes it clear that it is ‘not at all a case
of “each has its own truth”’ (127). Governing falsifying narration is instead
an effort to replace the sensory-motor system along with the philosophical
principle of representation as the core constituents of narration. Deleuze
explains that in the time-image ‘description stops presupposing a reality
and narration stops referring to a form of the true at one and the same
time’ (130).
As narration has traditionally been viewed as inseparably bound to
chronology and causality (e.g. in the fabula-syuzhet distinction), Deleuze,
following Alain Robbe-Grillet, regards modern(ist) narration to be ‘dysnar-
rative’. This concept has been explained according to three decisive tactics
(Neupert 1995, 137-138; cf. Vanoye 2005). First of all, the narrative purpose
becomes the revelation of the arbitrary nature of narration. Secondly, it
aims to show that narration is always a reduction of a reality which is far
more complex. Finally, it overthrows the logic of everyday mental processes

involved in traditional fabula-construction. Hiroshima mon amour em-
ploys these three tactics via a mediation on questions of memory and our
relationship to history and the past. Thereby, the film can be aligned to the
Deleuzian concept of ‘becoming’. As Francis Vanoye (2005) has contended,
‘dysnarration replaces a finished product with elements of a product in the
process of becoming’ (Vanoye in Neupert 1995, 138). Characteristic of the
dysnarratives of modern(ist) cinema is thus a reversal of the hierarchical
relation of linear and non-linear elements.
This reversal can also be related to the two cinematic modes of depicting
memory in Hiroshima mon amour. The first is classical linear, which
means that memory takes the form of recollections that are organized
and integrated chronologically to form the story of Nevers. The second is a
non-linear memory that becomes associated with the trauma: too immense
to grasp, too disturbing to recollect, and too intimate to share. Thus, a
memory only remains non-linear if it escapes actualization, i.e. recollection,
schematization, representation, or narrative integration. The film, therefore,
establishes a dichotomy between a narrative and representational discourse
of recollection and an anti-narrative, anti-representational discourse of
non-linear sheets of the past.
As mentioned earlier, the first memory-image of the German soldiers is
neither framed nor integrated into the ongoing narrative. It is only later,
when the woman tells her lover about her love affair, that her memory
transforms into a proper recollection-image that can be integrated into
the story of Nevers. The film narrates the story of Nevers in harmony with
the classical conventions for the flashback, since the woman is granted
the role of narrator as she tells the story of how she secretly met with the
German soldier in abandoned places. As the woman starts narrating (‘It
was in Nevers’), her off-screen voice accompanies the images. In contrast
to the voice-over in the opening sequence, the voice of the woman is this
time firmly located in time and space (in bed with her Japanese lover),
and spectators are not led to doubt about the relation between images and
off-screen voice. The function that this classical part is assigned in the
overall framework of the film is particularly fascinating.
While it may be tempting to perceive (the events of) Nevers as a key
to understanding the psychology of the woman, and maybe even a key
to understanding post-war ‘Hiroshima’, the story of Nevers is presented
as merely one of many possible stories and entrance points. Asked about
his insistence on hearing the woman’s story of Nevers, the Japanese man
provides not a single but four answers.11 On a metalevel the film thereby
reminds its audience that it could have been told differently, and that a
myriad of other sheets of the past could have been actualized instead of
that of Nevers. The contingency of the film’s narrative is further emphasized
as the Japanese man states: ‘I’m only just beginning to know you, and from
the many thousands of things in your life, I chose Nevers’. Due to such
moments, Pierre Kast has contended that with Hiroshima mon amour
‘[w]e are at the very core of a reflection on the narrative form itself’ (Kast
in Domarchi et al. 1985, 64).
Yet, the film’s essential moment of narrative reflection comes at a later
point. After telling the Japanese man her story, the French woman looks
at her mirror image full of self-loathing, and addresses her German lover:
You were not quite dead. I have told our story. I have betrayed you this
evening with this stranger. I have told our story. You see, it could be told.
For fourteen years I have not refound… desire for an impossible love. Since
Nevers. Look how I’ve forgotten you. Look at me.
The woman conceives of narration – having made her memory into a story
to be recounted – as a form of betrayal. Her fear is that recollecting her past
as a narrative to be retold, relived, and re-enacted will eventually lead to
an indifference towards that memory (what in psychoanalysis would lead
to a treatment of the trauma). Reni Celeste (2005) acutely observes that ‘[t]
o have made the lost German lover into a story is to have betrayed him, to
have made him into discourse and turned intimacy into mere narrative’
(sec. 5). Thereby, the film invites a comparison of the myriad collective
representations of Hiroshima with the re-enactment of the woman’s trauma.
In this fashion, the film questions the desirability of a narrative (and thus
linear) forming of our understanding of memory, knowledge, temporality,
and history.
Hayden White (1980) has famously argued that the problem with nar-
ration is exactly that it appears to be a natural, objective, and value-free
discourse. Narration is as ‘universal as language itself’, and a ‘verbal repre-
sentation so seemingly natural to human consciousness that to suggest it
is a problem might well appear pedantic’ (1). Nevertheless, White contends
that any science must be sceptical of not only the structure and patterns
it unravels, but also the discourse with which this is accomplished. White
explains: ‘For whatever else a science may be, it is also a practice which must
be as critical about the way it describes its objects of study as it is about the
way it explains their structures and processes’ (1).
The historical event of Hiroshima is perhaps one of the most narrated in

modern history. In the film, the French actress is reminded of the indifference
towards Hiroshima that gradually developed in the French collective con-
sciousness.12 In commemorating the past, narrative recounting of traumatic
events can provide a possibility to relate to something to which it is otherwise
impossible to relate. However, as Hiroshima mon amour spells out, the
likely price of this is that our cognitive, emotional, and affective responses
to the historical event gradually become schematic and habitual. According
to Deleuze (2005b), this is particularly the case when narrative recounting
is subsumed under the classical narrative logic of the sensory-motor system:
We see, and we more or less experience, a powerful organization of

poverty and oppression. And we are precisely not without sensory-motor
schemata for recognizing such things, for putting up with and approving
of them and for behaving ourselves subsequently, taking into account our
situation, our capabilities and our tastes. We have schemata for turning
away when it is terrible and for assimilating when it is too beautiful. (19)
In the beginning of the film, the woman both experiences and tries to
understand Hiroshima according to the narrative, schematic, and represen-
tational mode of recounting historical events. In this fashion, she perceives
Hiroshima in terms of what Michael S. Roth (1995) calls ‘narrative memory’,
which ‘integrates specific events into existing mental schemes’ (98). Here
‘the specific events are decharged, rendered less potent as they assume
a place in relation to other parts of the past’ (98, emphasis in original;
cf. French 2008, 7). In the same way as it has been argued about cinematic
conventions, narrative memory requires a double procedure that involves
forgetting (cf. Chapter II). The question, then, becomes to avoid the indif-
ference that follows from narrative recounting.
The woman in Hiroshima mon amour is caught in a catatonic memory,
which neither allows her to remember nor to forget. Not unlike the people
of Hiroshima, she is thus caught in a state of involuntary repetition of the
traumatic past. Ultimately, the film identifies the lovers with the geographi-
cal place of their trauma. In the final scene of the film, the woman says
to the man: ‘HI-RO-SHI-MA. That’s your name,’ and he replies: ‘That’s my
name. Yes. Your name is Nevers. Ne-vers in France’.
In this manner, Hiroshima mon amour is critical of the ability of history,
narration, epistemology, and (affective and reflective) spectatorship to
represent reality. Sarah French (2008) perceives the film as an
early example of a text that pre-empted the crisis of history through its
suggestion that certain historical events are fundamentally unrepresent-
able. History is thereby displaced from the discursive realm and memory
is imbued with a privileged status as the primary means by which to
recapture the past, as well as the most ethical and truthful method
through which to re-examine historical trauma. The film largely rejects
historical discourses for a more fluid and less hegemonic depiction of
memory that emphasises subjective and intersubjective experience. (11)
According to French, the drawback of this is that the woman remains

trapped in a ‘compulsive repetition of the past,’ and is thus granted ‘limited
subjectivity and agency with which to recover from her traumatic past’ (11).
Her insistence on an anti-narrative discourse propels her into an ‘obsession
with her traumatic past [that] results in madness, the loss of identity and
the detachment of the self from the social world’ (10).
Hiroshima mon amour is conspicuously modern(ist) in its obsession
with the loss of innocent (classical) meanings, and in this quality, I would
argue that the film simultaneously evidences both the potentials and the
limitations of Deleuzian film-philosophy in relation to complex narratives.
These concern the rupture of the sensory-motor linkages, which introduce
a radical disjunction between situation and action. Herein lies for Deleuze
the potential of time-images to open up for the non-linear dimension of
cinema. However, Rancière (2006) has questioned the Deleuzian idea
that movement-images and time-images are different by virtue of their
cinematic nature. He argues that ‘movement-image and time-image are
by no means two types of images ranged in opposition, but two different
points of views on the image’ (112-113), and consequently ‘Deleuze’s division
between a movement-image and a time-image doesn’t escape the general
circularity of modernist theory’ (108).
In my contention, a strict opposition of movement-image and time-image
comes with the risk of introducing an unproductive reversal of the hierarchy of
the linear and non-linear dimensions of cinema that favours the non-semantic,
autonomous, virtual, singular, unpredictable, and indeterminate becoming of
the cinematic process above a hermeneutic search of conventional and estab-
lished meaning or sense (cf. Leys 2011, 449; cf. Chapter II). The problem with
such a valorization of ‘non-linearity’ is that it leads to an ‘embrace [of] a highly
abstract and disembodied picture of mind or reason in order to repudiate it’
(Leys 2011, 458, n. 43). Furthermore, Deleuzian scholars often adopt a series
of cognitive-formalist assumptions about narrative comprehension insofar
as a film-philosophical narratological framework remains to be developed. In
order to cut across the linear-non-linear divide, Chapter V draws on a plethora
of embodied insights drawn from cognitive (neuroscience), film-philosophy,
and phenomenology to formulate the concept of the embodied fabula.
V. Towards the Embodied Fabula
In fact, an important and pervasive shift is beginning to take place in cognitive

science under the very influence of its own research. This shift requires that we
move away from the idea of the world as independent and extrinsic to the idea of
a world as inseparable from the structure of these processes of self-modification
[...] The key point is that such systems do not operate by representation. Instead of
representing an independent world, they enact a world as a domain of distinctions
that is inseparable from the structure embodied by the cognitive system.
‒ Varela et al. 1992, 139-140
What is Mind? No Matter. What is Body? Never Mind.1
Introduction
In the opening scene of David Fincher’s Fight Club (1999) spectators are
literally taken on a ride through the brain of the film’s protagonist. The
film can thus be taken as emblematic of a tendency within complex narra-
tives to move inside the characters’ brain spaces. Observing this, Patricia
Pisters (2012) labels such films ‘neuro-images’ (a category of films largely
corresponding to complex narratives) since ‘[w]e no longer see through
characters’ eyes, as in the movement-image and time-image; we are most
often instead in their mental space’ (14). Another example is to be found in
Darren Aronofsky’s Requiem for a Dream (2000) in which the effects of
the main protagonists’ drug use are not merely ‘represented’ but cinemati-
cally ‘enacted’ so as to make spectators embody the drug experience. In his
trademark ‘hip-hop montages’, Aronofsky uses short close-ups to intensify
particular actions that demonstrate the effects of the drugs such as the
widening of pupils, the preparation of cocaine, a fast-motion illustration of a
sudden explosion of energy (e.g. dancing, talking, cleaning), or microscopic
images demonstrating changes in brain chemistry.2 Jamie Skye Bianco (2004)
has described the cinematic experience of such montages as follows:
We sense and feel drugged in this explosion of intensive powers and

control that normally bind the apparatus to the organic clock. The game
that Requiem [for a Dream] plays out is the relentless organization of
non-organic rhythms, temporalities, diffractions, and affects decentering
the capacities of the observer. (388, emphases in original)
According to Bianco, the film confronts its audience with ‘an unbalancing
sensation of forces’ (392) that calls for a rethinking of the relatively stable
and affectively detached spectator equally associated with classical cinema
and cognitive film science.
In a similar vein, this chapter maintains that examples such as those
above cannot be adequately captured from the perspective of a cogni-
tive appraisal theory (e.g. Lazarus 1984) based on the assumption that all
‘mental phenomena whether conscious or unconscious, visual or auditory,
pains, tickles, itches, thoughts, indeed all our mental life, are caused by
processes going on in the brain’ (Searle 1984, 18). However, in refuting ap-
proaches to narrative and emotion that regard cognition as the main key for
understanding cinematic spectatorship (e.g. Carroll 2003; Plantinga 2009b;
Plantinga & Smith 1999; Smith 1995), it is vital not to ascribe everything
that does not fit into the rationalist categories of ‘knowledge’ or ‘cogni-
tion’ to the autonomous, anti-cognitive, and affected ‘body’ (cf. Leys 2011,
458). In approaching an embodied conception of the fabula, it becomes
possible to avoid an inverted version of the cognition (mind) versus affect
(body) dichotomy, without reducing the aesthetic and beauty of cinema to
a rationalist-intellectualist instrumentalism.
The embodied reconceptualization of the fabula contests three assump-
tions automatically implied in the cognitive-formalist fabula. The first
assumption is that 1) dominating spectators’ narrative experience is a search
for an unambiguous and chronological storyline. Framing the narrative
continuum according to this search, the cognitive-formalist fabula favours
a particular cognitive, analytical, and detached mode of experiencing the
narrative, which, however, is merely one of several interrelated modes of
narrative experience. The second assumption is 2) that the narrative experi-
ences of the spectator can be deduced from an analysis of how the film ‘cues’
them to perform a series of mental operations (e.g. inferences, evaluations,
and hypotheses) that allow for the construction of a chronological and (more
or less) coherent storyline. In this fashion, the fabula not only renders the
narrative continuum causal-linear, it also implies that this is the default
mode of narrative comprehension. In this fashion, narrative comprehension
becomes an exclusively rational, linear, and analytically detached activity,
which can be secluded from the immediate, bodily, and affective dimensions
of the cinematic experience. From this, the third assumption arises, namely
that 3) our affective, sensuous, visceral, emotional, and haptic experiences
of the film have no impact on our cognitive construction of the narrative.
The concept of the embodied fabula can thus be seen in the context of
an ongoing reinterpretation of the spectator’s bodily, intellectual, mental,
Towards the Embodied Fabul a 113
and cognitive activities. In recent years, scholars have drawn upon such
seemingly diverse frameworks as phenomenology, cognitive and affective
neuroscience, and Deleuzian film-philosophy to propose an appreciation of
the cinematic experience as embodied, non-mediated, and direct. Authors
such as Vivian Sobchack, Laura Marks, Steven Shaviro, Vittorio Gallese
& Michele Guerra, Torben Grodal, and Patricia Pisters have successfully
dismantled one of the most prominent metaphors in the study of cinema,
namely that of the ‘disembodied eye’ (cf. Branco 2012). In The Neuro-
Image (2012), Patricia Pisters connects the notion of embodiment with a
new image-regime: the neuro-image. Insofar as it describes ‘a new type of
cinema belonging to twenty-first-century globalized screen culture’ (6),
the neuro-image is akin to complex narratives. Among Pisters’s main argu-
ments is that contemporary screen culture no longer considers images to
be ‘illusions of reality’ but ‘realities of illusions’. Images in contemporary
cinema, in other words, ‘operate directly on our brains and therefore as real
agents in the world’ (6).3
Despite of the pioneering work of Pisters and others, there remains a lot
to be done in carving out how disembodied assumptions remain buried in
the conceptual and narratological tools we apply to analyse cinematic works
and to understand the cinematic experience. This chapter makes a case for
utilizing the embodied insights to revise the analytical tools that form our
basic understanding of the cinematic experience. In taking up this task, I
propose the concept of the embodied fabula as a film-philosophical and
narratological basis for understanding the embodied and situated nature
of cinematic spectatorship. In order to elucidate how disembodied assump-
tions dominate prevalent modes of conceptualizing narrative comprehen-
sion, it is useful to cast a glance on the criticism that embodied cognition
has launched against the central assumptions of its classical counterpart.
Towards an Embodied and Enactive Approach to Cognition
The argument pursued in this part is that the notion of embodiment – as it

has been used broadly in cognitive sciences, (M.L. Anderson 2003; Shapiro
2011; Wheeler 2005), f ilm studies (Gallese & Guerra 2012; Grodal 2009;
Marks 2000; Sobchack 1992), philosophy (Clark 1998; Gallagher & Zahavi
2012; Noë 2004; Varela et al. 1992), cognitive and affective neuroscience
(Damasio 1995; Gallese, Fadiga, Fogassi, & Rizzolatti 1996; Rizzolatti &
Sinigaglia 2008), linguistic (film) theory (Coëgnarts & Kravanja 2012; Lakoff
& Johnson 1999), and narrative theories (Caracciolo 2011; Fludernik 2010;
Hogan 2012) – demands a radical reinterpretation of the classical scientific

and representational assumptions of dominant structural-linguistic and
cognitive-formalist theories of cinematic narration (e.g. Bordwell 1985a,
2007; Branigan 1992; Chatman 1978).
Within the cognitive sciences embodiment is a reaction against the
classical cognitive approach, which according to Wheeler (2005) has failed to
‘reflect the complex, temporally rich, neurobiological dynamics that occur
in [the] brain’ (283). Furthermore, in the classical approach ‘the nonneural
body is conceived merely as a physical vessel in which the cognitive system
resides’ (283-284). The result has been an overemphasis on the inferential
and rational aspect of cognition, which in reference to the computational
analogy has been systematically isolated from the ‘causally complex and
temporally rich dynamics of the underlying neurobiology’ (284). In relation
to this, neuroscientists have grown increasingly disconcerted with the
linear, mechanical, and causal concept of mind that connected cognitive sci-
ence and neuroscience at earlier stages. As early as 1999, the neuroscientist
Rafael Núñez remarks this about his field:
[W]ithin the last couple of decades, the study of mind has experienced an
interesting and gradual change. There has been a tendency to move from
a rational, abstract, culture-free, centralized, non-biological, ahistorical,
unemotional, asocial, and disembodied view of the mind, towards a view
which sees the mind as situated, decentralized, real-time constrained,
everyday experience oriented, culture-dependent, contextualized, and
closely related to biological principles – in one word, embodied. (59)
In his review of the new cognitive research paradigm Embodied Cogni-

tion (2011), Lawrence Shapiro highlights three general themes of embodied
cognition that contrast with the computational model of mind. These are
1) conceptualization, 2) replacement, and 3) constitution (cf. Shapiro 2011,
4-5). In short, ‘conceptualization’ can be summarized under the header
‘bodies matter’ and involves the idea that the concepts that an organism
develops and relies upon for it to understand its environment depend on the
kind of body that it has. In brief, were the bodies of an organism to differ so
would their understanding of the world. ‘Replacement’, on the other hand,
describes the manner in which the organism’s body in interaction with
the environment replaces the need for the representational explanations
of classical cognitive science. Finally, there is ‘constitution’ – also known
as the ‘constitution claim’ – according to which the body or the world is a
constituent of, and not merely a causal influence on, cognition.
Today most cognitive researchers accept that the body plays a role in
cognition, yet, as Shaun Gallagher (2015) has recently observed, there are
distinct ways to conceptualize this. As such, the notion of embodiment
ranges from
[t]he idea that the body, and not just the brain, processes information
both prior to and subsequent to central manipulations; the idea that
representations can be action-oriented; the idea that the body itself plays
a representational role; or the idea that sensory-motor contingencies,
bodily affects, posture and movement enter into cognition in a non-
representational way; the idea that the body is dynamically coupled
to the environment; the idea that action affordances are body- and
skill-relative, and so on, are all ways of shifting the ground away from
orthodox cognitive science. (97)
Most researchers, however, trace the term ‘embodied’ back to The Embodied
Mind (1992) by Varela, Thompson, and Rosch. Here the term is used to make
two separate, yet related, points:
By using the term embodied we mean to highlight two points: first, that
cognition depends upon the kinds of experience that come from having
a body with various sensorimotor capacities, and second, that these
individual sensorimotor capacities are themselves embedded in a more
encompassing biological, psychological, and cultural context. By using
the term action we mean to emphasize once again that sensory and motor
processes, perception and action, are fundamentally inseparable in lived
cognition. (Varela et al. 1992, 173)
Thus, as the term ‘embodied’ was first introduced, it consisted of two inter-
related claims. The first claim – the moderate claim – is today generally
accepted within the cognitive sciences and states that cognition depends on
corporal processes. The second claim – the more radical and controversial
– states that these processes themselves are constituted by, situated in, and
extend into the specific cultural, psychological, or (media-)technological
context in which they are embedded.
As embodied cognition has gained dominance a wave of what Gal-
lagher (2015) refers to as ‘body snatchers’ promote a version of ‘embodied
cognition that leaves the body out of it’ (97). These researchers invoke the
term ‘embodied’ in a form that undermines the more radical version of em-
bodiment according to which ‘the body as it is coupled to the environment,
plays a constitutive role in cognition’ (99). Against such ‘body snatchers’, the
embodied fabula embraces the original, radical notion of embodiment that
fundamentally associates mindfulness with purposeful embodied cognitive
activities that extend into the tools, technologies, and our sociocultural
context that together define our meaningful engagements with the world
(cf. M.L. Anderson 2003, 104). As such, the concept of the embodied fabula
differs from the ‘moderate’ notion of ‘embodiment’ to be found in cogni-
tive (film) science, which maintains that the body and brain are, partly
at least, determined prior to our existence as social and cultural beings.
According to this approach, ‘we are universal in the making, before we get
a specific language and are formed by the circumstances and times we live
in’ (Bondebjerg 2015, 4).
Embodiment within contemporary cognitive film theory is thus per-
ceived as an invitation to understand culture on the basis of a neurobiologi-
cal determination of the brain-body. This is, for instance, the case in Torben
Grodal’s (1999, 2009) theory of ‘universal film genres’ as corresponding to
various aspects of our evolutionary hardwired reptilian and/or mammalian
brain. Another example can be found in Patrick Colm Hogan’s (2012) ‘affec-
tive narratology’. Here the author argues convincingly that ‘story structures
are fundamentally shaped by and oriented by our emotion system’ but takes
this to mean that ‘in order to formulate a systematic theoretical account
of stories, we should turn first of all to affective neuroscience and related
fields of study’ (1, emphasis added).
Given their emphasis ‘on cognitive and emotional mechanisms that all
spectators presumably share – or on the universal features of our cognitive
and emotional make-up’ (Vaage 2009, 161), cognitive film scholars tend to
assume a fundamental distinction between the ‘subject’ (spectator) and
the ‘object’ (the film) and in extension, between the ‘story’ and the ‘telling’.
From this perspective, the notion of ‘embodiment’ is reduced to the affective,
biological, and evolutionary pregiven machinery of the human brain-body
while the film’s story is determined by a realist reference to inherent ‘facts’
or ‘clues’ (cf. Bordwell 1985a, 1991, 2007). However, although cognitive and
affective neuroscience may prove important for a provisional understand-
ing of emotionality and affect, the ambition cannot be the replacement
of structural-linguistic universal laws with a neurobiological reductionist
discovery of cross-cultural structures that determine our brain-body and thus
our corporal-affective, cognitive, and emotional reactions to cinema. Against
this wave of neuro-reductionism, the embodied fabula is an attempt to move
beyond the ‘unbridgeable ontological chasms between “objects,” which are
“out there,” and subjectivity, which is “in here”’ (Lakoff & Johnson 1999, 93).
Furthermore, the embodied fabula proposes an intimate linkage

of action, cognition, and perception. This contrasts with the classical
cognitive position, which assumes that these belong to clearly separate
domains. Susan Hurley (2002) has encapsulated the computational and
representational presumptions of this position with the metaphor of the
‘classical sandwich’. Hurley explains the framework of this metaphor as
follows:
A view of perception and action as separate input and output systems

complements a view of thought and cognition as ‘central’ and in turn
separate from the ‘peripheral’ input and output systems. The virtual
processing of cognition is seen as central, even if its implementation is
distributed; input to it is provided by perception, and it issues output
that generates action. The subpersonal underpinnings of the mind are
conceived as vertically modular, with cognition interfacing between
perception and action. (20)
Hurley’s acute metaphor encapsulates how the classical stance perceives

cognition as the ‘sandwich filling’ between perception and action. In this
model, cognition is seen as the necessary mediator between the inputs we
receive from the environment and the motor actions we rehearse upon it.
Following from this is an ontological disassociation of perception, cognition,
and action that has legitimized a detachment of cognition as an autonomous
register that can be studied in isolation from bodily changes, emotional and
affective reactions, and visceral responses. This classical assumption is also
present in cognitive film theory, as, for instance, in Bordwell’s theoretical
isolation of the spectator’s cognitive and emotional responses to a film.
Here both ‘bottom-up’ and ‘top-down’ processes are inferential insofar as
‘perceptual “conclusions” about the stimulus are drawn, often inductively,
on the basis of “premises” furnished by the data, by internalized rules, or by
prior knowledge’ (Bordwell 1985a, 31). Consequently, as Richard Allen (2001)
has remarked, according to this framework, ‘all perceptual processes are
cognitive, because perception is a form of inference, albeit unconscious
inference [i.e. computational]’ (177).
Insofar as the cognitive-formalist fabula is the product of the spectator’s
cognitive comprehension of the story material and since it can be theoreti-
cally isolated from other types of responses to the film (e.g. emotion and
affect), it is possible to argue that the fabula occupies the role of the filling
in Bordwell’s classical sandwich of narrative comprehension. Thus, the
fabula can be seen as the mediating entity between film (perception) and
the spectator’s responses (action). In this fashion, cognitive formalism

isolates and accentuates the mental operations usually associated with
high-level cognition and conscious awareness (e.g. inferences, evalua-
tions, hypothesis testing, and schemata application) as principal for the
construction of the fabula.
Consequently, the cognitive analyst often reasons backwards from a
complete and coherent version of the fabula with the aim of analysing
the obstacles and hindrances that the narration played on the viewers in
their construction of the narrative. Cinematic narration is here perceived
according to a unidirectional communication model constructed around
two isolated parts (input and output) and mediated by cognitive processes
such as inference-making, picking up the film’s cues, problem-solving,
a framing and testing of hypotheses, applying schemata, etc. Due to the
constant flow of incoming inputs, the cognitive machinery must constantly
revise its output until a final fabula can be constructed. Bordwell proposes
this model on the assumption that our main cognitive goal can be taken
as (or, perhaps we could say our inner computer is programmed for) the
‘construction of a more or less intelligible story’ (Bordwell 1985a, 33).
The fabula has long proven a useful analytical tool; however, Bordwell – in
drawing upon cognitive science – has gone a step further than seeing the
fabula as merely an analytical tool. He considers the fabula to be a mental
representation, i.e. to be psychologically real. In a blog entry from 2011,
Bordwell reflects upon the criticism that Narration in the Fiction Film has
received since its publication in 1985. Interestingly, Bordwell appears to be
especially critical towards his conception of the fabula as an easy-accessible
and coherent mental construction. Considering the criticism maintaining
that his computational version of the fabula is psychologically implausible,
Bordwell (2011) writes:
Eventually I had to agree. For one thing, we aren’t aware of building up

a fabula in our heads, the way we can be at least partially aware of, say,
solving a crossword puzzle. For another, we can’t access it easily: try
stopping a film on video and reciting the entire chain of events leading
up to the moment of pause. Worse, try at the end of the movie to grasp
mentally the entire fabula you’ve purportedly worked out. Chances are
you can’t do it. Given that our memories are reconstructive rather than
photographic, creating an accurate fabula is extremely difficult [...] What
is built up in our memory as we move through a film is something more
approximate, more idiosyncratic, more distorted by strong moments, and
more subject to error than the fabula that the analyst can draw up. (sec. 2)
It would appear that Bordwell here acknowledges that the rational, com-
putational data processing and the production of a final, coherent, closed,
linear, and causally ordered fabula no longer amounts to a satisfactory
description of the viewer’s mental experience while watching the film. The
Bordwellian fabula may, nevertheless, still prove to be an acute descrip-
tion of the particular analytical and rational engagement with ‘solving’
cinematic narratives, which I have referred to as the analytical fabula. Thus,
the analytical fabula remains a useful concept for describing a particular
manner of organizing narrative perception with the aim of conjuring up a
clearly defined linear trajectory. It is, in other words, a valuable description
of the processes underlying linear cinematic perception (cf. Chapter III). Yet,
the cognitive model is still shaped by ‘the observer of the modern’, which
according to Elliott (2011), today ‘has given way to the experiencer of the
postmodern’ (2). Therefore, the analytical fabula must be complemented
with a concept of the fabula better apt at highlighting how films rely on
‘embodied sensations for their meaning’ (5).
As most of the pioneer cognitive film theory has focused its attention
on matters such as classical narrative techniques (cf. Bordwell et al. 1985),
narrative comprehension (cf. Bordwell 1985a; Branigan 1992), and the basic
rules of spectator psychology (cf. Currie 1995; Persson 2000), it was initially
thought that ‘a weakness of the cognitive approach was its inability to deal
with the elicitation of emotion in film’ (Plantinga 2002, 19). Yet, due to the
importance of emotions in cinema, it did not take long before a catalogue of
cognitive approaches to the examination of the role of emotion and affect in
cinema appeared (cf. Carroll 1996; Tan 2013; Grodal 1997; Smith 1999; G.M.
Smith 2003). Carroll (2006) has, for instance, argued that ‘affect is the glue
that holds the audience’s attention to the screen on a moment-to-moment
basis’ (21). Consequently, despite Bordwell’s original reluctance to include
emotion and affect into his model of narrative comprehension, it would
be an overstatement to criticize cognitive film science more generally for
ignoring the role of these in cinema. 4
What is of interest here is, therefore, the role that emotions have been
granted within cognitive film theory in light of their adherence to a research
paradigm that has traditionally attempted to factor out emotion and affect to
the maximum degree possible (Gardner 1985, 41). The cognitive film-theoretical
answer to this problem, as Plantinga (2002) concisely sums up, has been ‘that
emotions have reasons’ (24). According to this view, ‘our emotional response
to texts (and other phenomena) is dependent in part on how we evaluate and
assimilate information’ (24). This enables film cognitivists to treat emotions
and affect as part of – rather than secluded from – our cognitive processing.
In this fashion, emotions are the result of ongoing cognitive processes that
can be evaluated along with other modes whereby a film ‘cues’ its viewers.
Emotions must thus bear on the textual ‘data processing’ and are consid-
ered primarily in terms of cognitive-textual processing of the narrative. Ed
S. Tan (2013) expresses the cognitive comprehension of emotions in cinema,
when he argues:
As is customary in theories of the cognition of discourse, we are assuming

that the viewer’s comprehension of the film narrative begins with the
formation of the text base, a propositional representation of the discourse.
This text base is the first result of following with understanding the
filmic action, which is relatively close to the directly observable surface
structure of the film. (197)
As Thomas Elsaesser & Malte Hagener (2010) have noted, the separation of
body and mind is guiding most cognitive film scholarship, insofar as
[t]he cognitivist position would hold that all sense perception and
physical sensations are processed by the brain, including in the cinema.
Therefore it is neither ‘the body’ nor ‘the senses’ but the brain that decides
whether something is pleasant or painful, hot or cold, wet or dry. (166)
Mullarkey (2009) has argued that as a result ‘affectivity is reduced to the

brain’s information processing’ (56). A related criticism of the cognitive
account of how cinema elicits emotions has been proposed by Daniel Framp-
ton (2006), who criticizes the cognitive approach for failing to do justice to the
affective, poetic, pure sensuous, and aesthetic dimension of cinema (106-107).
Indeed, a general objection to the cognitive approach concerns its anchorage
in what is assumed to be ‘everyday’, ‘universal’, or ‘natural’ processes specific
neither to art nor cinema. This focus, as Daniel Yacavone (2015) maintains,
necessarily causes an emphasis on cinema’s realism over its more creative
potential (169). Robert Sinnerbrink (2011) has labelled this dimension – often
ignored by cognitive film scientists, possibly because they have a problem
fitting it into the rational vocabulary of computational data processing – the
‘aesthetic dimension’ of the image. This comprises
those features which contribute to, but also remain independent of,
narrative meaning: the images’ sensuous qualities, their visual rhythms
and tempo, their use of colour, texture and form, their dramatic (and
undramatic) moments of singularity in gesture and performance, their
mood-disclosing capacities, their orchestrating of aural and visual pat-

terning, their ability to reveal and conceal nuances of expression in the
human face and body, their capacity to express movement and time in
novel ways, and so on. (52)
Although he recognizes that Bordwell does have a theory of film style,

Sinnerbrink insists that the aim of cognitive formalism is to ‘avoid “impres-
sionistic” interpretation in favour of formalist analysis, generic classification
and historical contextualization’ (52).5 In a manner comparable to that
of Yacavone and Frampton, Sinnerbrink (2011) questions whether ‘this
intellectualist account of narrative does justice to the receptive viewer’s
aesthetic experience of cinema’ (52). In relation to complex narratives,
Sinnerbrink concludes that the cognitivist approach overlooks ‘the varie-
ties of “non-cognitive” affective response, cognitive dissonance and visual
fascination that such f ilms can also powerfully evoke’ (52). In recent years,
much research within embodied cognition has been devoted to replace
the ‘sandwich’ model, emphasizing instead a more direct relation between
stimulus and response (Gallese & Guerra 2012; Hurley 2002). Rather than
being reliant on cognition and the construction of representations, Alva
Noë (2004) posits that ‘we take advantage of the fact that we have more
immediate links to the world because we are in the world from the start,
and that we have the sorts of bodily skills to exploit those linkages’ (24).6
Embodying the Narrative Experience
An alternative account of how cinema elicits emotions and affects is cur-

rently being formulated with reference to the much-debated mirror neuron
system (cf. Rizzolatti & Sinigaglia 2008) and/or to affective neuroscience
(cf. Damasio 1995, 1999, 2003, 2011). Both have been interpreted as challenges
to the earlier discussed principles of cognitive science as identified by
Wheeler (2005); in particular to the principle #7, which states that ‘intel-
ligent action remains conceptually and theoretically independent of the
agent’s physical embodiment’ (27; cf. Chapter II).
This principle has been challenged by the discovery of mirror neurons
by a group of researchers at the University of Parma.7 These unique neurons
received their name because they ‘fire’ not only when an animal performs
an action, but also when it observes the same movement performed by
another animal (cf. Gallese, Fadiga, Fogassi, & Rizzolatti 1996; Rizzolatti
& Craighero 2004). Thus, at a neuronal level at least, no difference is made
between observation (perception) and sensory-motor activation (action).

Consequently, mirror neurons are perceived to demonstrate how intimately
cognitions and sensory-motor activities are coupled in the brain, insofar as
the same brain areas are activated upon execution of a motor action and per-
ception (cf. Pfeifer & Bongard 2007, 171). The discovery of mirror neurons has
had immense implications throughout the cognitive sciences, and Vilayanur
Ramachandran (2000), one of the leading neuroscientists, has predicted that
‘mirror neurons will do for psychology what DNA did for biology’ (par. 5).
Rizzolatti & Sinigaglia (2008) have explicitly stated that mirror neurons
render visible the inadequateness of the ‘classical sandwich’ in which ‘the
problems inherent to movement would be reduced to the mechanics of its
execution – according to the classical pattern: perception → cognition →
movement’ (x). Against the presumptions of classical cognitive theories
mirror neurons ‘show how recognition of the actions of others, and even of
their intentions, depends first of all on our motor repertoire’ (xii). The result
is that the link between body and cognition (or mind) becomes a lot more
intricate and direct, and emotions and affect are not merely the products of
higher processes of cognition nor the hardware ‘noise’ they were conceived
to be in classical cognitive science.
While it appears that mirror neurons may work best in real life, face-to-
face situations (cf. Iacoboni 2008), there is evidence that suggests mirror
neuron activity in humans while being exposed to moving images (cf. Mur-
ray et al. 2008; Iacoboni & McHaney 2009). Although it is still unsure what
precisely the future research on mirror neurons will reveal, the discovery
of mirror neurons has been an encouragement for those who explain
mental activity beyond the abstract level of representations. Similarly,
mirror neurons make it possible to conceptualize a mode of spectatorship,
which is not modelled on cognitive and emotional representations, but on
a much more direct and immediate bond between film and spectator. As
such, mirror neuron activity caused by observation calls for new modes of
thinking about the cinematic experience in a more enactive and embodied
manner than allowed for by representational and computational cognitive
studies of spectatorship.8 Vittorio Gallese (one of the researchers behind the
discovery of the mirror neurons) & Michele Guerra (2012) have proposed
a theory of ‘embodied simulations’ to understand how we engage with
cinematic stories and characters. The authors reject the ‘mind-reading’
and ‘folk-psychological’ explanations that underline cognitive theories on
cinema from Bordwell’s on narrative comprehension to Murray Smith’s on
character engagement. As an alternative, Guerra and Gallese propose an
immediate, non-mediated, and direct bond between spectator and film,
which appears surprisingly commensurable with the Deleuzian conception

of ‘the brain is the screen’.9
John Protevi (2010) has maintained that the Deleuzian tripartite ontologi-
cal differentiation of the virtual, intensive, and actual could prove a valuable
philosophical framework for embodied scholars. Following the work of
Brian Massumi (2002), Manuel DeLanda (2002), and Jeffrey Bell (2006),
who have demonstrated that Deleuze’s philosophy offers a wide-ranging
ontology that can be applied to current research projects that like embodied
cognition use complex system theory, Protevi emphasizes the ability of
Deleuzian philosophy to ‘think material systems in terms of their powers of
immanent self-organization and creative transformation’ (421). Deleuze in-
terestingly maintains that ‘individuated entities’ (e.g. a person, a hurricane,
or a perception) are produced by an actualization of a virtual field. In short,
such a view replaces representations (or platonic Ideas) with a tripartite
‘ontological difference’ in which ‘(1) intensive morphogenetic processes
follow the structures inherent in (2) differential virtual multiplicities to
produce (3) localized and individuated actual substances with extensive
properties and differentiated qualities’ (422).
The Deleuzian virtual is non-Platonic, insofar as it is different from, yet
not completely separated from, the actual. As such, the actual ontological
register is the level of properties of formed substances, while the virtual is
the ‘level of the structures of the intensive processes productive of such actual
substances’ (424, emphases added). The communication between the virtual
and the intensive processes is bidirectional, because ‘the interaction of in-
tensive processes changes the virtual conditions for future processes’ (422),
and the virtual can therefore not be opposed to the real. In Bergsonism (1991),
Deleuze explains that the virtual can be distinguished from the possible,
since the possible is opposed to the real, whereas the virtual is opposed
to the actual. Thus, ‘the possible has no reality (although it may have an
actuality); conversely the virtual is not actual but as such possesses a reality’
(96, emphasis in original). Furthermore, to become actualized, ‘the virtual
cannot proceed by elimination or limitation, but must create its own lines
of actualization in positive acts’ (97, emphasis in original).
Returning to Protevi, the point is that the replacement of the opposi-
tion between the possible and real with that of the virtual and actual,
can also be utilized in extension of the criticism that embodied cogni-
tion has directed towards the ‘classical sandwich’ model of perception.
Whereas the ‘classical sandwich’ leads to a unidirectional and causal-linear
information-processing model according to which perception is inferential
and representational, the embodied approach differs by defining cognition
in terms of constant feedback loops or ‘continuous reciprocal causation’

(cf. Clark 1998).10 According to Protevi (2010), Deleuze’s philosophy grants
perception with a new ontological framework, which no longer depends
on the ‘classical sandwich’:
Sensory input (changes in body correlated with changes in the world)

continually feeds into the system along the way, either reinforcing
the settling into a pattern or shocking the brain out of a pattern into a
chaotic zone in which other patterns strive to determine the behavior
of the organism. The neurological correlate of a decision is precisely the
brain’s falling into one pattern or another, a falling that is modeled as the
settling into a basin of attraction that will constrain neural firing in a
pattern. There is no linear causal chain of input, processing, and output.
Instead there is continually looping as sensory information feeds into an
ongoing dynamic system, altering or reinforcing pattern formation; in
model terms, the trajectory of the system weaves its way in and out of a
continually changing attractor landscape whose layout depends on the
recent and remote past of the nervous system. (426)
Hence, perception is no longer regarded to be constituted by linear mechan-

ics, but is instead understood as a complex or non-linear system. Conse-
quently, the linear laws guiding classical cognitivism have been replaced
with a more dynamic conception of the brain-body.
The feedback loops of embodied cognition combined with the Deleuzian
tripartite ontology make it possible to challenge the ‘classical sandwich’
model of perception that shapes the Bordwellian conception of spectator-
ship. Furthermore, it can be argued that complex narratives – although
containing classical, causal-linear traits – experiment with non-linear
starting points that resonate with complex system theory and the embodied
cognitive conception of the brain.11 Thereby, these films likewise challenge
the classical ‘sandwich’ of cognitive film science, where the output (the
fabula) is inferentially constructed from the input (the syuzhet) in terms of a
computational data processing that has been ‘programmed’ to comprehend
the story material in causal-logical and coherent terms. In the ‘hip-hop’
montages of Requiem for a Dream or in the reverse-chronological
narrative structure of Memento, spectators cannot simply master the
images cognitively (i.e. comprehend them by ‘re-chronologizations’ or
‘de-complexifications’). As shall be argued in the next chapters, our at-
tempts nevertheless to do so – i.e. to provide the images with causal-linear
meanings – play a vital part to the narrative immersion that characterizes
these films. The affective and cognitive dimensions are thereby being
reconfigured in a manner that challenges the ‘classical sandwich’ according
to which our emotional, affective, and bodily responses are divorced from
the cognitive-representational mode of comprehending the narrative.
According to Damasio, the discovery of mirror neurons is merely one indi-
cation of why cognitive neuroscience must revise its philosophical roots and
replace the computational model of mind with a more embodied concept of
cognition. In Descartes’ Error (1995), Looking for Spinoza: Joy, Sorrow and the
Feeling Brain (2003), and more recently Self Comes to Mind (2011), Damasio’s
central thesis is that the body and brain (and, it could be added, conse-
quently the body and brain of the cinema spectator) cannot be separated.
Not surprisingly, his research has drawn the attention of both embodied film
cognitivists (cf. Grodal 2009) and film-philosophical scholars (cf. Bellour
2010; Pisters 2012). His invocation of Spinoza as a relevant philosophical
framework for understanding the neuroscientific findings concerning the
body’s relation to the mind (and as Pisters [2012] rightly notes, ‘specifically in
the direction of body to mind’ [19, emphasis in original]) reveals the possible
linkage to – plus a scientific pendant to – an appreciation of cinematic
spectatorship as embodied. This is encapsulated in the Deleuzian phrase:
the brain is the screen (cf. Deleuze 2000). This formulation suggests that
the brain of the spectator cannot properly be thought of as independent
from the environment that it simultaneously forms and by which it is being
formed. In fact, Damasio’s research, as Elliott (2010) observes, is remarkably
complementary to the philosophy of Deleuze:
In Damasio’s conception of the neuro-somatic organism, we see reflec-

tions of the Deleuzio-Guattarian heuristic framework [...]: the connec-
tions between the flesh and the mind are indeed revealed to be more grass
than tree and the perceptual processes are discussed within the physical
framework that they are inevitably rooted in. As Damasio is at pains to
point out throughout his texts, this is a recent direction in neuroscience,
a discipline that has traditionally privileged thought over feeling and that
has, more recently, viewed the mind as the more important software that
runs on the hardware of the body. 12 (6)
However, we should be careful not to conclude that the discovery of mir-

ror neurons and Damasio’s research point univocally in the direction of
Deleuzian film-philosophy insofar as it has also been used in support of a
moderate interpretation of embodiment (cf. Bondebjerg 2015; Grodal 2009).
It is important to note that this research primarily points to an intricate
bond between the cognitive and the affective, and thus holds the potential
of bridging the polarized positions of cognitivism and film-philosophy. In
fact, Damasio’s answer to the Cartesian error (of posing a mind untouched
by the body) is adding a third dimension between the body proper and the
cognitive mind. This would be the emotional and feeling somatic self, which
couples the activities of the mind with the functioning of the body. Thus,
Damasio retains the idea and importance of a cognitive brain (although
cognition can no longer be taken as the primary and autonomic machinery
of human thought), while simultaneously confirming the Spinozian/Deleuz-
ian claim that ‘to think is to learn what a non-thinking body is capable of,
its capacity, its postures’ (Deleuze 2005b 182, emphases in original).
Implied is a far more complex relation between the affective, emotional,
sensuous, audial, and cognitive aspects of cinema than proposed not only in
the ‘sandwich’ models of classical cognitivism, but also in film-philosophy,
since Deleuze – being a philosopher rather than film theorist – is mainly
interested in the morphogenetic processes in which the virtual and actual
interchange. In short, Deleuze neglects classificatory questions that can be
posed from the perspective of the actual register (cf. Protevi 2010), which is
why it is important always to keep in mind that he is primarily interested
in cinema as a philosopher and not as a film theorist. Consequently, it is
worth noting that Deleuzian film-philosophers and cognitive film theorists
may converge in terms of their research interests, but are ultimately not
intrigued by the same problems. As a philosopher Deleuze is, in other words,
not concerned with the cognitive-representational symbol manipulation
that occupies cognitive theorists. At the same time cognitive theorists have
proven to be less concerned with understanding the experimental and
aesthetic aspects of cinema. In proposing the embodied fabula, the aim is
to offer a narratological and film-philosophical tool sensible to both higher
cognitive processes and the ‘aesthetic’ dimension of cinema.
In David Lynch’s Lost Highway (1997), the narrative folds around itself
like a Möbius strip – a strip of paper twisted 180° and then looped so the op-
posing ends are connected (cf. Pickover 2007). This figure has been used to
explain how the film’s two main characters – Fred Madison (Bill Pullman)
and Pete Dayton (Balthazar Getty) – suddenly double each other and switch
places. In fact, the film’s co-author Barry Gifford (1997) explicitly mentions
the Möbius strip as a source of inspiration for the narrative of the film:
We realized we didn’t want to make something that was linear, and that’s
why the Moebius strip [came to function as a metaphor for the film’s
structure]. A Moebius strip is a long strip of paper curved initially into
a circle, but with one end flipped over. The strip now has only one side
that flips both inside and outside the shape. It made it easier to explain
things to ourselves and keeping it straightforward. The story folds back
underneath itself and continues. (par. 3)
The Möbius strip is a helpful device for conceptualizing the play of inside and
outside (e.g. subjective/objective and bodily/cognitive) in Lost Highway,
which comes full circle as Fred at the end of the film is revealed to be the
sender of the mysterious message he receives in the beginning of the film.13
Yet, the explanatory powers of the Möbius strip can also be extended
and act as model of the feedback loops of complex narratives more gener-
ally. Here the cognitive and the affective, the subjective and the objective,
and the linear and the non-linear constantly change roles and interact.
Elsaesser (2009) also refers to the continuum of the Möbius strip to explain
contemporary ‘mind-game’ films, where
the narrative engenders its own loops or Möbius strips, where there may
well be a beginning, a middle, and an end, but they certainly are not pre-
sented in that order, and thus the spectator’s own meaning-making activity
involves constant retroactive revision, new reality-checks, displacements,
and reorganization not only of temporal sequence, but of mental space,
and the presumption of a possible switch in cause and effect. (21)
The Möbius strip further contributes to highlight contemporary cinema’s

indebtedness to classical and modern(ist) cinema, while simultaneously
demonstrating the difficulties of trying to reduce the narratives to the
avatars of either of these.
As we have seen, Deleuze sees in modern(ist) cinema an expression of the
Bergsonian conception that memory, opposed to how it has traditionally
been viewed, is not something stored in our brains as a mental representa-
tion of the past (cf. Chapter IV). Rather memory is a powerful, expressive,
and inventive force, which must be constantly re-enacted (even when it
takes the form of an involuntary Proustian memory) and thus nothing that
the brain can be said to contain. We call up memories, ideas, and fantasies
as a means of navigating the world. Deleuze (2005b) finds an expression of
this by quoting Federico Fellini, who once stated that ‘[w]e are constructed
in memory; we are simultaneously childhood, adolescence, old age and
maturity’ (96, emphasis in original). Deleuze gives this a concise expression,
when he writes: ‘Memory is not in us, it is we who move in a Being-memory,
a world-memory’ (96).
In extension of this, I suggest that contemporary complex cinema makes

evident that narrative is not in us, but it is we who move in a Being-narrative, a
world-narrative (cf. Hven 2014). This creative rewriting of Deleuze’s take on
the role of memory in the time-image provides us with a conceptualization
of how complex narratives challenge us to replace the ‘classical sandwich’
conception, where spectators create a cognitive-symbolic representation of
the narrative (i.e. fabula). It, furthermore, allow us to formulate a principle
of narration as a form of embodiment, enactment, and experience that
occurs in the dynamic process of film viewing. Since this concept includes
the existence of the analytical fabula and acknowledges the influence of
linear cinematic perception , it reconfigures – rather than abandons –
the higher, cognitive and analytical processes that guide our interaction
with the narrative environment. Although reminiscent of the cognitive-
formalist fabula, the analytical fabula differs in being the description of
one particular analytical, goal-oriented, logical, and detached attitude to
the cinematic narrative. Consequently, it is not an adequate representation
of the ‘actual’ story of the film (at best, a particular analytic and ‘offline’
depiction of it). Both the analytical and embodied fabula should thus be
understood as
operational representations by corporeal agents engaged in culturally

mediated, practical activities of perceiving, working, playing, deciding,
evaluating, and judging in a world that simultaneously responds to those
operational representations and exceeds them. (Connolly 2002, 91)
Such a conception allows for an incorporation of both the cognitive and

affective aspects of spectatorship, but also for the study of their interac-
tions. Furthermore, it challenges the cognitive computational analogy
and its preference for organizing narrative in linear terms exclusively. The
embodied fabula allows us to think about complex narratives as being
structured as Möbius strips capable of immersing the audience in their own
cognitive and affective processes of narrative comprehension. As a result,
narrative comprehension is not simply dependent on textual processing
but is also (mis)guided by cognitive distortions and emotional or cognitive
dissonances.
The argument that cinema and narrative in this fashion involve us
directly, furthermore, proposes a richer conception of narrative pleasure,
which is able to complement cognitive theories that examine how ‘[a] film’s
narrative structure is largely a structure designed to cue emotional, visceral,
and cognitive experience’ (Plantinga 2001, sec. 5). Plantinga traces out five
principles of the spectator’s pleasure of which the first four pertain to what
he terms ‘intratextual pleasures’ organized by the film’s structure, such as
orientation and discovery, visceral experience, and character empathy. To
these Plantinga includes the ‘intertextual pleasures of the text’, which he
dubs ‘reflexive criticism and appreciation’. He explains:
Any technique that draws attention to itself, and away from the story, is
thought to transgress that fundamental rule. Yet although this is a com-
mon rule of thumb, it is also one that is commonly ignored, as reflexive
works become increasingly popular on both film and television. We also
enjoy the intertextual pleasures of the text. (sec. 6)
Complex narratives frequently employ a non-linear or achronological

sequencing of the events. Susan L. Feagin (1999) explains that sequencing
‘is the order of the presentation of events of a narrative or story. This order,
the order of the discourse, may or may not be the same as the order of the
events in the story’ (173). Pulp Fiction (Tarantino 1994) provides Feagin
with an example of this: ‘Pulp Fiction [...] no doubt enjoys some of its
popularity because of the way one’s understanding of the story develops
along with one’s understanding of the structure of the film’ (173). Brani-
gan (2002) has granted this conception of narrative comprehension with a
cognitive explanation:
Filmmakers employ the psychology of the everyday in order to aid

spectators in comprehending a narrative. Filmmakers also employ this
psychology against spectators when it is important that something not
be seen or fully understood during the telling of a story (e.g., to create
mystery or surprise), or when the spectator must understand in a new
way (e.g., in a metaphorical way or through a sudden revelation), or when
something disturbing or traumatic must be reconfigured by the text or
repressed. As spectators, we make mistakes in making inferences because
we are systematic in drawing inferences and authors count on that. (106)
The intriguing question implied, but left unanswered by Branigan, is: Why
do we experience pleasure when films transgress and reveal the boundaries
of logical computation and reasoning?
David Mitchell (2002) has pointed out that complex narratives can thwart
the viewers’ logical fabula construction and still provide pleasure in a move
beyond what Murray Smith (1995) terms ‘alignment’. Mitchell discusses
how M. Night Shyamalan in The Sixth Sense (1999) manages to trick the
audience to miss the ‘clues’ planted in the film to make a pre-anticipation

of its twist ending – revealing that the main protagonist Dr Malcolm Crowe
(Bruce Willis) himself is a ghost – possible. While arguing that the film
manages this in a move beyond Smithian ‘alignment’, Mitchell makes a vital
observation about the film, which it can be argued is a prominent feature
of complex narratives more generally. He writes:
To me, it seems clear that the reason we ignore such clues is that we have
become like Malcolm. We ‘see what we want to see’, we ‘don’t even know
that he’s dead’. What Shyamalan does so well is to present us with what
Malcolm sees, and no more, aligning us with him. (sec. 2)
Mitchell proposes a more direct relation between the character and the
audience than the cognitive-appraisal theory underlining the work of Ed S.
Tan, Murray Smith, and Carl Plantinga. Cinema gives us a new depersonal-
ized perspective on the world, and the pleasure stems from our immersion
in the narrative-world: ‘[W]e must take part in it; we must “interact with it
as an ongoing story, as our story”’ (sec. 2).
Such a view is supported by studies of how narration shapes our everyday
life. Kay Young & Jeffrey Saver (2001), for instance, discuss different types
of ‘dysnarrativia’, which are ‘states of narrative impairment experienced by
individuals with discrete focal damage in different regions of the neural
network subserving human self-narrative’ (75). The authors argue that
studies of ‘dysnarrativia’ have shown that narrative framing and recall
of experience is a dynamic, variable, and vulnerable process and that an
awareness of this prompts a new understanding of the brain and, conse-
quently, of human experience:
Modern neuroscience has demonstrated that retrieving memories is not

the simple act of accessing a storehouse of ready-made photos in a stable
neural album, preserved with complete fidelity to the moment of their
formation. Rather, each act of recall is a re-creation, drawing upon mul-
tiple, dynamically changing modular fragments to shape a new mosaic
[...] All memories are suspect, at the neural level. Fidelity-stable recall and
self-interpretation of the past is not a property of the human brain and
mind. The varied subjectivity of literary autobiographic productions has
inescapable subjectivity of the brain’s narrative and memory system. (79)
This seems to suggest that we do not necessarily align ourselves with charac-
ters because we share their values and beliefs, but at times also because we
as an audience ‘incorporate the narrative we are watching’ (Mitchell 2002,

sec. 2). Young and Saver draw on the work of the psychologist Jerome Bruner,
whose basic claim they summarize as follows:
[N]arrative organizes not just memory, but the whole of human experience
– not just the life stories of the past, but all of one’s life as it unfolds. Bruner
describes narrative as an instrument of mind that constructs our notion of
reality, and asserts that the experience of life takes on meaning when we
interact with it as an ongoing story, as our story. (75; cf. Bruner 1991, 2004)
Bruner imagined this to operate on a purely cognitive level, yet recent

neuroscientific findings suggest a neurobiological underpinning of the
centrality of narrative in human cognition (Young & Saver 2001, 75). In
combination with the aforementioned discovery of mirror neurons and
the research on emotions undertaken by Damasio, Smithian ‘alignment’
as based on cognitive appraisal appears to be more like the tip of the ice-
berg than the core asset of character engagement. However, these layers
remain imperceptible in films that do not produce a cognitive-emotional
dissonance between our immediate bodily preposition to the images and
our moral judgements relating to the characters’ beliefs and goals. Align-
ment on the basis of cognitive appraisal, in other words, appears to be a
valid explanation, when films follow a classical, linear trajectory bereft of
cognitive-emotional dissonances. In contemporary screen culture, however,
viewers and characters no longer necessarily share a moral compass.14 The
fragmented narrative structures prominent in contemporary cinema,
furthermore, promote an intuitive mode of spectatorship that encourages
the audience to embody (rather than comprehend) the narrative to reach
an understanding of the characters.
As I have already embarked upon, the idea that we incorporate, or
embody, the film that we are watching – as opposed to representing it
to ourselves in the mind’s eye – is closely reminiscent of the Deleuzian
contention that ‘the brain is the screen’. In a frequently cited interview
entitled ‘The Brain is the Screen’ (2000), Deleuze clarifies what neurobiology
has to offer the study of cinema:
The brain is unity. The brain is the screen. I don’t believe that linguistics
and psychoanalysis offer a great deal to the cinema. On the contrary, the
biology of the brain – molecular biology – does. Thought is molecular.
Molecular speeds make up the slow beings that we are. As Michaux
said, ‘Man is a slow being, who is only made possible thanks to fantastic
speeds.’ The circuits and linkages of the brain don’t preexist the stimuli,
corpuscles, and particles [grains] that trace them. Cinema isn’t theater;
rather, it makes bodies out of grains. The linkages are often paradoxical
and on all sides overflow simple associations of images. Cinema, precisely
because it puts the image in motion, or rather endows the image with self-
motion [auto-mouvement], never stops tracing the circuits of the brain.
Cinema not only puts movement in the image, it also puts movement in
the mind. Spiritual life is the movement of the mind. One naturally goes
from philosophy to cinema, but also from cinema to philosophy. (366)
For Deleuze, cinema sets images going in the mind, and in this sense
one ‘naturally goes from philosophy to cinema, but also from cinema to
philosophy’.15 In other words, the cinematic experience is caught up in
dynamic cinema-brain-world patterns, which can be thought of in terms
of the Möbius strip.
In Psycho (1960), Hitchcock experiments with how much he can ma-
nipulate the audience’s emotions and bodily arousal. In the beginning of
the film, Marion Crane (Janet Leigh) is entrusted with the task of bringing
USD 40,000 to a safety deposit box, yet falls for the temptation of stealing
the money and leaving both job and city behind. Before Marion makes her
decision, she weighs the pros and cons of the situation, allowing us to get
an insight into the motives that drive her immoral act. Leaving the city
at dusk, Marion eventually falls asleep in her car. She is clearly in shock
as she is woken up by a police officer, and she starts to act suspiciously,
thereby increasing the danger of being exposed. When she is asked to show
her driver’s license, she almost reveals the envelope containing the stolen
money. Hitchcock orchestrates the scene in his trademark suspenseful
manner to create an anxious sensation in the audience that mirrors Marion’s
fear of being caught.
Although Marion’s act of stealing the money is not exactly a righteous
moral one, its consequences are also not too dire (the owner, Tom Cas-
sidy, declaring: ‘I never carry more than I can afford to lose’). In cinema,
viewers have proven willing to accept certain immoral actions, as long
as the consequences for the victim are ignored (it can be argued that the
main theme of David Cronenberg’s A History of Violence [2005] is the
cinematic ignorance of the consequences of violence), marginalized (as in
this particular example from Psycho), or if the victims have displayed an
even more depraved morality (e.g. The Boondock Saints [Duffy 1999]). On
a first glance, this example appears to confirm Smith’s theory of alignment,
since our acceptance of the characters’ actions seems to be structured by
the three components of character engagement (recognition, alignment,

and allegiance). Yet, when Hitchcock later mirrors this scene from the
perspective of the film’s ‘psycho’ this theory is challenged.
After Marion is killed in the famous shower scene halfway through the
film, her murderer Norman Bates (Anthony Perkins) becomes the film’s
main character, as we follow his attempts to clear all evidence by disposing
of Marion’s corpse. He loads her dead body into the trunk of her car, and
drives to a deserted swamp, where he intends to conceal his dark crime
forever. Interestingly, Hitchcock now orchestrates the scene from the per-
spective of Norman to create suspense. As Norman has driven the car into
the swamp, the car is slowly disappearing and the evidence consequently
seems to vanish with it. Suddenly, however, the car stops sinking, as if the
swamp is not deep enough to cover the car. For a brief second, Norman’s
murky smile freezes and at the very same moment an anxious sensation of
being caught – comparable to the earlier one, when Marion was the main
character – is aroused in the audience (or at least some of us). However, this
time around it is clearly not the beliefs, morals, or our approval of the aims
of Norman that can be deemed the proper cause of this corporal-affective
sensation. Instead, the film arouses a primordial bodily sensation that
is in dissonance with our higher-level cognitive response to the scene. If
spectators become consciously aware of this dissonance, the eerie sensation
may intensify further once we feel relieved to see the car being swallowed by
the nothingness of the swamp.16 How are we to explain this phenomenon?
I believe that in the example drawn from Psycho, the viewers are
enabled to grasp – if only for a brief moment – what Deleuze terms our
‘pre-hodological’ disposition towards the ongoing action of the plot. Deleuze
(2005b) derives the concept of hodological spaces from the psychologist
Kurt Lewin. A hodological space is ‘a field of forces, of oppositions and
tenses between these forces, of resolutions of these tenses according to
the distributions of goals, obstacles, means and detours’ (125). Before our
bodily responses are formed in accordance with the action we are about to
perform and to the sensory-motor linkages that enfold this action in line
with our beliefs, judgements, and morality – i.e. according to our hodological
space – we inhabit a ‘pre-hodological’ space. This is a space with ‘overlap-
ping of perspectives, which does not allow the grasping of a given object
because there are no dimensions in relation to which the unique set would
be ordered’ (125). A ‘pre-hodological space’ is thus a bodily state yet to be
formed by higher cognitive processes. It can be perceived as a gesture that
occupies the interval between stimulus and response and prepares the
body for a later response. In the pre-hodological space, the obstacle (in our
examples from Psycho this would be the police officer or the car that will
not sink) ‘does not, as in the action-image, allow itself to be determined in
relation to goals and means which would unify the set, but is dispersed in
a “plurality of ways of being present in the world”, of belonging to sets, all
incompatible and yet coexistent’ (196).17 Deleuze develops his concept of
the ‘pre-hodological space’ in proximity to Gilbert Simondon’s concept of
individuation. Simondon argues that
the obstacle, as it is really lived, is the plurality of ways of being present

in the world. Hodologic space is already the space of a solution [...] Prior
to action the subject is firmly lodged between many worlds, between
many orders. Action is the discovery of the meaning of this disparation,
of that by which the particularities of each set are integrated in a richer,
larger, set, one possessing a new dimension.18 (Simondon in Deleuze 2001,
47, n. 1, emphasis in original)
Returning to our example, our ‘pre-hodological’ bodily response to the

scene with Marion and the police officer can easily be assimilated into
the ongoing sensory-motor linkages of the plot, insofar as the film has
already gone some way to justify Marion’s act. It is, in other words, possible
to explain our arousal by reference to cognitive dispositions, such as our
willingness to accept that Marion ought to keep the money so she can quit
her dreadful job (which gives her recurring headaches), marry well, and
start afresh. Thus, our emotional experience of the narrative information
resonates with the corporal-affective arousal instigated by Hitchcock’s
mastery of the cinematic medium. In this case, the anxiety we experience
as Marion is almost caught, is likely to be interpreted as caused by our
compassion for the character. Due to the lack of incongruity between our
emotional responses and our general cognitive dispositions, this scene
does not produce the cognitive dissonance we experience in the scene
with Norman.
The example from Hitchcock’s Psycho thus paves the way for a decon-
struction of the causal reasoning implied by cognitive theories that assert
that our emotional response is structured by our cognitive processing of
the narrative. Such can be achieved along the lines of Friedrich Nietzsche’s
deconstruction of causality as outlined by Jonathan Culler (1986). Causal-
ity has long been taken as a fundamental and universal principle of our
universe. Much of our thinking is constructed around the idea that one
event causes another, and that causes produce effects. As Culler notes, ‘the
principle of causality asserts the logical and temporal priority of cause to
effect’ (86). Yet, Nietzsche challenges the concept of causal structures as

given a priori, and asserts that it is rather the product of a precise tropo-
logical or rhetorical operation, which amounts to a reversal of chronology
(a chronologische Umdrehung). Culler provides the following example of
this. When we feel pain, we look or search for a cause of that sensation
of pain, say a pin. When a proper cause has been identified, one posits a
link that reverses the perceptual and phenomenal order. The initial order
of experience ‘pain → pin’ is now reversed into the causal sequence ‘pin →
pain’. Nietzsche writes:
The fragment of the outside world of which we become conscious comes

after the effect that has been produced on us and is projected a posteriori
as its ‘cause’. In the phenomenalism of the ‘inner world’ we invert the
chronology of cause and effect. The basic fact of ‘inner experience’ is
that the cause gets imagined after the effect has occurred. (Nietzsche
in Culler 1986, 86)
I believe such a reversal, in addition to being central for the cognitive aspect
of narrative comprehension, may prove valuable for our comprehension of
how cinema structures our emotional response according to – or in disagree-
ment with – the sensory-motor linkages that structure our higher cognitive
experience of the narrative. That is, spectators first feel a sensation, and
then search for a cause in the narrative for this sensation. Our cognitive
construction of a coherent and chronological fabula – most obviously in
cases of a non-linear or reverse chronological syuzhet – relies exactly on such
chronological reversals. Thus, rather than posing causation as an intrinsic
organizational principle of the narrative itself (or our representation of it),
causality should be understood as a basic organizational principle, with
which humans give meaning to a system.
In 1946, Baron Albert Michotte completed a series of experiments
designed to reveal how humans attribute causality to objects that are –
theoretically – not impacted by one another. By cinematic means, Michotte
demonstrated that when objects move with respect to one another within
highly limited constraints, viewers are more than likely actually to see
causality. In actuality, however, all we see is an object which moves towards
another and makes contact, and the second object is then seen moving in
the same direction, which is what produces the effect that one object is
‘launching’ the other like billiards. (Bruner 1986, 17)
Of interest here is the fact that causality is an underlying basic scheme
with which most people understand and make sense of the world. This, for
instance, becomes important when we argue that a given story is causal-

linear. It makes a difference whether this is supposed to mean that causality
is an intrinsic quality of the narrative in itself, or that causal-linearity is one
among several principles of organization that can be utilized by the audi-
ence to give meaning to the cinematic material. Nietzsche’s deconstruction
of causality ensures that causality cannot be thought of independently from
the narrative it is designated to explain in the first place. Consequently,
not only our understanding, but even more important, our application of
causality requires increased sensibility to whether we impose causality on
the narrative, or the narrative perhaps rather imposes it on us. Considering
how film analysts perceive the question of causality in relation to complex
narratives turns out to be quite revealing of the basic onto-epistemological
assumptions of their interpretative strategies.
VI. The Complexity of Complex Narratives
In spite of all this diversity and the different ways of approaching and assessing
this body of films, most theorists would agree to subsume these films under
the predicate ‘complex narratives’. However, this complexity is itself a rather
complex phenomenon, since each author seems to have different films and
different aspects of film narrative in mind.
‒ Simons 2008, 111
Complexity is not only a feature of the systems we study, it is also a matter of the
way in which we organize our thinking about those systems.
‒ Tsoukas & Hatch 2001, 979
Introduction
This chapter examines two complex narratives: Tom Tykwer’s ‘forking-

path’ narrative Lola rennt (Run Lola Run 1998) and Alejandro González
Iñárritu’s ‘mosaic’ narrative 21 Grams (2003). Despite their differences in
narrative form, national, historical and political context, audio design,
acting style, and visual expression, much of the critical attention that has
been granted these films has followed a similar line of reasoning. Both films
have been perceived as representatives of a particular mode of complexity,
which certain critics have claimed dominates contemporary narration.
The idea is that despite their intricate narrative forms, complex narratives
actually adhere to a classical, linear logic. This chapter aims to show how
this view is rooted in the linear onto-epistemological conception of the
cognitive-formalist fabula. In order to do so, a discussion will be undertaken
with a number of analytical approaches to the two selected films, which
in different manners exhibit prevalent modes of dealing with cinematic
complexity. The aim of doing so is twofold: 1) to elucidate the distinguishable
manner in which both films challenge typical analytical, narratological,
and conceptual frameworks for understanding cinematic complexity, and
2) to formulate alternative approaches that avoid the reductionist and
dichotomized principles according to which complex narratives have often
been examined.
Although embodied cognition rarely employs the actual math-
ematical or physical theories that have been developed within the vast
research f ield of dynamical system theory, chaos theory, non-linear
dynamics, or the study of complex systems (henceforth ‘complexity theory’),

this orientation makes use of concepts derived from these fields as ‘a highly
intuitive set of metaphors for thinking about physically embodied agents’
(Pfeifer & Bongard 2007, 93). Thus, within embodied cognition the brain
and its interaction with the environment are quite often thought of in terms
that have been derived from complexity theory.1 Similarly, the appeal of
complexity theory to film studies lies not in a straightforward mapping
of these concepts onto the theory of cinema. Rather our interest in these
concepts should be seen in relation to their capability of opening up the
cinematic experience to those processes that go beyond causal-linearity.
It, in other words, pertains to their ability to heighten our sensibility for
the complexity of the historical, social, environmental, and temporal mat-
ters that have largely been ignored within the long-dominant Newtonian
scientific ideals (cf. Capra 1996; Goodwin 1994; Hayles 1990; Prigogine &
Stengers 1984 1997; Shotter 1993; Toulmin 1992; Tsoukas & Hatch 2001).
Hence, complexity theory has been perceived as the scientific expression of
a more general tendency in which ‘our vision of nature is undergoing a radi-
cal change toward the multiple, the temporal and the complex’ (Prigogine &
Stengers 1984, xxvii). Significantly, it is capable of revealing the limitations
of the monolithic linear perspective of classical Newtonian science without
resorting to the non-linear side of the dichotomy.
Unlike a linear system, a dynamic system is non-linear since it cannot
be decomposed into subsystems that can be solved individually in isola-
tion from ‘disturbing’ elements, after which its pieces can be reassembled
to provide a complete picture of the system. The mathematician Steven
Strogatz (2000) has provided an often invoked real-life example of the nature
of non-linear systems. Think about your favourite two songs. According to
the linear logic, playing them both at the same time would amount to the
double pleasure. Yet, actually doing so is more likely to ruin the pleasure
of both songs.2 Complex systems display non-linear behaviour insofar as
small differences in initial circumstance can have impact on the system
elsewhere (the ‘butterfly effect’) in a manner that exceeds the perspective
of a causal-linear logic. In addition, feedback loops make the system self-
governing and unpredictable (cf. Cilliers 2002; Holland 1999; Johnson 2001).3
The ‘butterfly effect’ – also called ‘sensitive dependence on initial condi-
tions’ – is probably the most famous example of how complexity theory
challenges the classical scientific notion of causality. The butterfly effect
takes its name from the poetic example provided by Edward Lorenz, who
claimed that a hurricane in one part of the world could be influenced by
minor initial conditions such as the flapping of the wings of a butterfly
The Complexit y of Complex Narr atives 139
weeks earlier in an entirely different part of the world (cf. Gleick 1988;
cf. Lorenz 1995). However, as William Brown (2014) has pointed out, this
example is potentially misleading because it appears to suggest that the
butterfly ‘caused’ the hurricane. Yet, the lesson to take from this example
is that instead of thinking about the butterfly as a direct cause, we should
think of it as part of a vast and complex system that comprises all kinds of
potential ‘causes’ such ‘that we cannot attribute a single cause at all’ (129).
For Poulaki (2014a), complexity inevitably constitutes a challenge to clas-
sical narratology. According to her, the ‘proliferation of chance events in
a film is incompatible with classical narrative causality, because chance
cannot easily be attributed to one single cause’ (393). Consequently, the
causality we find in complex narratives is ‘better conceived as a cumulative,
non-linear and emergent effect rather than as an event-sequence of causes
and effects’ (393).
For Prigogine & Stengers (1984) complexity means that ‘we must accept a
pluralistic world in which reversible and irreversible processes coexist’ (257).
In classical science, the authors argue, the focus has been on reversible
processes in which stability, order, uniformity, and equilibrium are prevail-
ing. However, reversible processes only occur in closed systems where the
parts relate to one another linearly. Therefore, classical science has largely
ignored the irreversible processes of open systems that are considered most
common by Prigogine and Stengers.
Complexity theory thus initiated a shift in attention towards irreversible
processes governed by disorder, instability, diversity, disequilibrium, and
non-linear relationships. In reversible processes, such as, for instance, the
motion of a frictionless pendulum, no privileged direction of time exists.
Hence, the term ‘reversible’ refers to the classical scientific assumption
that ‘if the velocities of all the points of a system are reversed, the system
will go “backward in time”’ (61). Reversible processes can then be said to be
symmetrical in time. As such, the directionality of time becomes irrelevant
insofar as a temporal reversal would produce no significant material or
physical change. Irreversible processes, on the other hand, are governed
by the ‘arrow of time’, which ensures that ‘time flows in a single direction,
from past to future’ (277). Alcohol and water spontaneously mix, yet, we
never see the reverse occurring, the spontaneous separation of the mixture
into pure water or alcohol.
Interestingly, in narratology and the study of cinema reversibility and
irreversibility have become granted a set of connotations, which do not
resonate with how they are comprehended and used in complexity theory.
Especially, it should be remarked that irreversible temporality within
cinematic narratology is intimately tied to the inevitable forward movement

of the time-based medium. As Doane (2002) has observed ‘film is in some
sense popularly understood as the exemplar of temporal irreversibility’ (66,
emphasis in original). Irreversibility for Doane, however, does not merely
relate to the forward progression of the ‘arrow of time’, but is also tied to
the mechanistic nature of the cinematic apparatus, which famously caused
Bergson to refuse cinema as an example par excellence of the spatialization
of time and of the ‘snapshot’ logic of perception (cf. Bergson 1998, 296-402).
Consequently, irreversibility within cinema has been connected to the
linear and causal-deterministic understanding of the narrative at the level
of the fabula, whereas temporal reversibility has been connected to the
non-linear sequencing at the level of the syuzhet. In most narratological
studies, non-linearity is understood as reversible processes that have been
temporally disorganized and thus appear in a non-linear sequencing. The
task of the analyst thus becomes the restoration of the proper causal-linear
order of the narrative events. However, from the perspective of complexity
theory this analytical procedure amounts to a classical scientific reduction
of complex temporal processes, since the underlying assumption at stake
here is that the components of the narrative can be arranged without caus-
ing a qualitative change of it. Hence, the narrative continuum is perceived
as atemporal or temporally reversible.
Bordwell (2008) has also noticed the shift in causality in ‘network nar-
ratives’ such as 21 Grams. He argues that the plot structure present in
such films must ‘find ways to isolate or combine characters in compelling
patterns that will replace the usual arc of goal-oriented activity’ (199).
For him, these films are organized around chance and thus their causal
structure is ‘loose’. Poulaki (2014a) argues that Bordwell’s comprehension of
‘network narratives’ relies on a default understanding of causality as ‘tight’
that resorts back to the classical Hollywood film (384). As she rightly points
out, Bordwell approaches ‘complex causality’ by means of ‘classical narrative
and anthropomorphic standards – the latter in the sense of events caused
by human actors and bringing forth other events as consequences of previ-
ous actions’ (384). Ultimately, Poulaki reveals that underlying Bordwell’s
definition of ‘network narratives’ as causally ‘loose’ is an incompatibility
between the classical narratological and complexity-theoretical under-
standing of causality. In this chapter, I argue that classical narratology often
reduces complex causality and chance to a matter of predestination and
determinism. Yet, as Poulaki (2014a) reminds us, a ‘network is not caused
by the individual actions of its elements, nor by a single transcendent and
overarching cause or motivation; it rather emerges as an organization of a
multiplicity of agencies and their complex relations’ (394). Consequently,

‘causality in network films is not loose, but complex and non-linear’ (395).
Therefore, it is not sufficient simply to mix old classical narratology with
complexity theory. Instead, complexity theory should be allowed to funda-
mentally challenge our narratological tools. Until then, as Poulaki states,
complex narratives will remain insufficiently addressed (394).
21 Grams – How Much Does Life Weigh?
In the ‘Production Notes’ (2003) made available on the official website for
21 Grams, the film’s director Alejandro Gonzales Iñárritu enigmatically
declares that we are ‘all just floating in an immense universe of circum-
stances’ (par. 3). This declaration finds a visual expression in the bewilder-
ing collection of seemingly unrelated scenes satiating the opening of the
film. Here the editing is fast, the handheld camera unsteady, and the tone
of the film grainy. In this fashion, the style reflects the muddled states of
mind that the films’ three main characters Cristina (Naomi Watts), Paul
(Sean Penn), and Jack (Benicio del Toro) find themselves in respectively.
Iñárritu’s baffling description of the film also reflects the film’s narrative
structure, or rather the spectator’s experience trying to come to terms
with it. Eventually, however, the film facilitates the emerging of narrative
patterns and meaningful relations between the events.
The mosaic narrative circles around a traffic accident in which Cristina’s
husband and two girls are killed. The hit-and-run driver Jack is a devoted
Christian, whose faith in God has given him the strength to leave his
criminal past behind. 21 Grams thus presents its narrative events from
the perspective of both victim and offender. In addition, the film tells
the story of Paul – a mathematician and the unapprised receiver of the
heart of Cristina’s deceased husband. After the transplant Paul cannot
come to terms with not knowing the precise string of events that changed
his destiny. Despite being encouraged simply to accept his fortune, Paul
nurtures an obsession with learning the identity of his donor. Eventually
he succeeds in tracking down Cristina, who has returned to an old habit of
self-destructive carousing and drug abuse. Their meeting resuscitates them
both and salvages Cristina from her suicidal existence. Eventually their
mutual affection amounts to what is possibly one of the most peculiar love
affairs in the history of cinema. 4 As their relationship develops, Cristina
convinces Paul that it is his duty to kill Jack (‘You are in his house, fucking
his wife and sitting in his chair, you owe it to him!’). Paul eventually gives
in to Cristina’s demands and together they plan to kill Jack, who for them is
simply a hit-and-run driver, but who, from the perspective of the spectator,
is also a recovered alcoholic and a devoted family father. However, Paul
is unable to carry out their plan, and in a courageous act of self-sacrifice
he shoots himself instead of Jack. This act not only rescues Cristina from
pursuing her destructive path of vengeance (she would not simply take
Jack’s life but also his constant feeling of guilt), it also opens up a brief, yet
pivotal, instant in which the eyes of victim and offender meet in a moment
of recognizing their shared suffering.
The Fabula and the ‘Decomplexification’ of the Narrative Continuum
Cognitive approaches often perceive narrative complexity as being measur-

able according to the intricacy of the computational processing that it
requires its spectators to perform. Narrative complexity thus becomes
a measure of how complicated the construction of a linear, coherent,
causal, and chronological story is. Since the formal structure of 21 Grams
requires a high degree of quite knotty backward reasoning, the film would
typically be perceived as a highly complex narrative. However, Michael Z.
Newman (2006) has questioned whether such complexity automatically
guarantees an overall complex narrative design. In the present context,
Newman’s argumentation is remarkable since it allows for the uncover-
ing of a series of dominant presumptions about the nature of cinematic
complexity. In particular, Newman’s treatment of 21 Grams illustrates how
film scholars utilize the fabula-syuzhet distinction to study the complexity
of narrative designs. Furthermore, it helps to elucidate the restrictive role
cognitive theories grant emotions (since these are entirely cognitive and
as such deprived of affect) when dealing with the narrative compositions
of ‘mosaic’ films such as 21 Grams.
The hypothesis of Newman’s article is that a complex plot, or syuzhet,
often comes at the expense of complexity in other cinematic areas. In the
particular case of 21 Grams, the lack of complexity purportedly occurs at the
level of the characters. Newman’s basic argument can be summed up accord-
ing to the following line of reasoning: Due to the temporally disintegrated,
non-linear, and fragmentary narrative structure of 21 Grams, spectators will
be heavily burdened with the cognitively demanding task of reorganizing the
story material into a chronological order. This procedure ensures complexity
at the level of the syuzhet. Yet, at the same time the film must compensate for
this complexity elsewhere. In 21 Grams this is achieved by a trivial depiction
of its characters, which ensures simplicity at the level of the fabula.
By comparing 21 Grams with the more classical storytelling of Pas-

sion Fish (Sayles 1992), Newman hopes to demonstrate that ‘character
complexity is independent of plot complexity, and that it is possible that
the complexity gained through temporal reordering may even come at
the cost of complex characterization’ (91). Newman’s contention that a
multifaceted character depiction does not necessarily follow from a complex
plot is compelling. However, it is problematic to assume that narrative
complexity can be determined with a procedure that isolates characters
from the structure of the plot. Newman is aware of the uneasiness that this
modus operandi may instigate at a theoretical level, but asks the readers to
suspend their scepticism, and allow him to use this distinction for purely
analytical purposes:
If we may think of the narrative text as having a discursive level separate

from its content – a dichotomy that may not withstand theoretical scru-
tiny but that is nonetheless useful as a heuristic – then 21 Grams throws
us on the mercy of the narration in a way that Passion Fish does not. (93)
In other words, the fabula-syuzhet distinction is an analytical tool for isolat-

ing a story from the manner in which it is told, and it thereby helps the
analyst to focus more clearly on these parts separately. This, however, also
means that it is a principle designed for the reduction of complexity – in
particular evoking Morin’s third principle for the rejection of complexity
by classical science, which ‘consists in isolating and separating cognitive
difficulties from one another’ (Morin 2007, 5; cf. Chapter I).
My objection here is not that such reductions cannot serve as useful
analytical tools, merely that we should not forget the task that such concepts
have been designed to perform. In fact, the fabula-syuzhet distinction can
be a useful tool for studying the interaction between two levels of the
narrative otherwise obscured. Newman’s treatment of Passion Fish is
a good example of how an initial reduction can promote an even more
complex understanding of the workings of a narrative. In this case, Newman
carefully examines how the film – through an insightful use of classical
narrative devices – repeatedly flaunts certain viewer-expectations, regard-
less of whether these rely on cinematic conventions or (negative) social
stereotypes. The film achieves this by holding back vital information to
reveal it at a strategically important point in the narrative. In this fashion,
Passion Fish exploits what Bordwell (1985a) has called ‘suppressed gaps’
to force its viewers to re-evaluate their initial inferences and judgements
about the characters. Newman (2006) persuasively argues that the effect
of this is ‘one of intensifying the complexity of the story, its characters, and
themes’ (97). To demonstrate how this narrative strategy works in the film,
it is useful to think of the fabula and syuzhet separately. As an analytical
tool this distinction renders visible the ‘trick of narration’ played by the
film, which according to Newman is an even subtler and ‘not nearly as
intrusive or flashy an effect as those we find in 21 Grams’ (97). Thus, in
his treatment of Passion Fish, Newman’s initial heuristic reduction of
separating the fabula from the syuzhet is employed to reveal a deeper-lying,
yet not immediately detectable, complexity.
The heuristic separation of fabula from syuzhet is given another flavour
when Newman subsequently employs it to explain how, and with what
consequences, 21 Grams ‘resists its readers’ desire for it to cohere and their
cognitive effort to make it so’ (94). In this manner, the non-linear style of 21
Grams contrasts with the ‘self-effacing and unobtrusive’ classical mode of
narration that characterizes Passion Fish. Newman argues that the result
of this is that the attention of the audience is primarily invested on the
cognitively demanding task of constructing a chronological and coherent
fabula, thereby leaving little room for cognitive complexity elsewhere. In
other words, much of the spectator’s cognitive focus will be devoted to
achieving chronology and causality between the events, an effort that is
eventually rewarded insofar as underlying the film’s ‘disentangled syuzhet,
the narrative reveals itself as rather straightforward’ (91).
Beneath the initial narrative complexity of 21 Grams, Newman detects a
far more straightforward and linear logic governing the film. This becomes
especially evident, so Newman argues, once the film is compared to the
narrative design of the less flashy, yet highly complex, storytelling of Pas-
sion Fish. Therefore, Newman can conclude that when all is said and done
both 21 Grams and Passion Fish are classical narratives: ‘Even with all of
[21 Grams’] temporal manipulations, the movie has the design of a canoni-
cal narrative’ (100). Ultimately, Newman treats the narrative complexity
of 21 Grams in a classical scientific manner, which means that it has to
be ‘straightened out’ and ‘explained away’. Newman takes upon him the
analytical task of ‘salvaging’ the spectators from their confusion by offering
a dissective treatment of the film capable of revealing the superficial reality
of the ostensibly narrative complexity.
This seemingly harmless shift in the approach to the question of narrative
complexity becomes especially significant, insofar as the methodological
tool applied – i.e. the fabula-syuzhet distinction – is itself designed for the
reduction of narrative complexity. However, Newman appears to apply this
distinction to 21 Grams as a perfectly legitimate tool for determining the
overall complexity of the narrative design. Assuming that a narrative will

reveal its ‘true’ complexity at the level of the fabula, as Newman does, is
dubious. This is mostly so because the concept of the fabula presupposes
the linear, chronological, and coherent comprehension of the narrative
continuum that the analyst detects behind the temporal disjunctions of the
syuzhet. It also reduces the spectator’s comprehension and experience of the
narrative to the cognitive task of solving its narrative structure. Newman
in this manner treats the narrative continuum as consisting of isolated
narrative modules, which constitute a unidirectional relationship to the
viewer, who may find pleasure and amusement in organizing these pieces,
but whose comprehension of the underlying fabula remains unaffected by
this narrative design.
In this cognitive reductionism, the spectator’s affective experience of
the narrative is assumed to be of no relevance to the question of narra-
tive complexity. Affect is simply the result of the lack of information that
temporarily blocks the emergence of a complete representation of the
narrative. According to this view, the apparent complexity of the syuzhet
exercizes little, if any, influence on the complexity of the fabula, since
from the analytical perspective this is predestined to transform into a
causal-linear series of events. Consequently, according to this approach,
‘linearization’ is perceived to be the condition of narrative comprehension
as such, rather than an analytical tool or a mode of organizing experience.
In his analysis of 21 Grams, Newman mistakenly perceives the fabula
as an objective representation of the film’s story independent from the
spectator/analyst, who has constructed it. Consider, for instance, when
Newman in relation to 21 Grams argues that ‘the sophistication of the
storytelling functions as a screen behind which the rather unsophisticated
story material is hidden’ (100, emphases added). Or, when he concludes from
this that 21 Grams ‘ends by decomplexifying its narration’ (104). Even if
Newman probably does not believe the narrative to be able to ‘decomplexify’
itself, it nonetheless reveals that something vital is lacking in his equations.
What, more precisely, is lacking, is an understanding of how the narrative
structure facilitates an affective, emotional, and cognitive experience of the
film that cannot be reduced to a series of chronologically ordered events
or ‘facticities’.5
In earlier chapters, it has been argued that cognitivist theorists often
utilize the fabula as an analytical tool that ensures an intersubjective
perspective on the narrative as it is (cf. Chapter II & Chapter V). A pervasive
cognitive assumption is that since narrative comprehension primarily oper-
ates according to rational, disembodied, and computational information
processing, both the fabula and the manner in which we construct it can be
understood independently from the viewers’ emotional, bodily, or affective
responses (cf. Bordwell 1985a). Following this logic, Newman does not find
it imperative to study how the syuzhet influences spectators affectively or
emotionally, and their subsequent experience of the characters, since such
subjective matters are not perceived as relevant factors for a strictly objec-
tive determination of the complexity of the fabula. For Newman (2006), 21
Grams takes us on a more cognitive than emotional ride, thus the ‘pleasure
in watching [21 Grams] to a large extent, is the pleasure of working out
explanations for how people and events are connected’ (93). In other words,
the narrative pleasure lies in the computational processing of the narrative,
rather than in its ‘aesthetic’ (cf. Sinnerbrink 2011), affective, or emotional
dimensions.
To counter this position, there is certainly no need to reject the possibility
that viewers will feel a kind of satisfaction once the narrative suddenly
‘adds up’.6 However, the conception that our engagement with narratives
is largely a matter of computational processing and puzzle solving has
demonstrated little explanatory value when it comes to narrative decep-
tions and cognitive-emotional dissonances (cf. Sinnerbrink 2011, 52). Instead
of considering it in isolation, the cognitive pleasure of sense-making and
narrative order must be connected to other aspects in which the narrative
moves us (e.g. emotionally and affectively).
However, Newman evaluates the complexity of 21 Grams’ characters
entirely from the perspective of the already constructed fabula, which
presupposes a rational comprehension, and excludes a more intuitive appre-
hension, of the characters. Newman’s understanding of characters appears
to rely on the early cognitive film theory on emotions as developed in the
work of Murray Smith (1995) and in the anthology on the subject edited by
Carl Plantinga & Greg M. Smith (1999). This theory is primarily concerned
with how spectators engage with fictive characters to draw conscious,
intellectual, and voluntary conclusions that form the basis of the spectator’s
judgement of the characters. This will eventually lead spectators to form
‘allegiances’ or ‘pro-attitudes’ with certain of them while defying others
(Smith 1995, 187-227). As such, we ‘identify’ with characters in ‘aligning’ (or
possibly misaligning) ourselves with their goals, beliefs, and motifs. This
approach harmonizes with the general tendency of cognitive researchers ‘to
discuss emotion states in terms of goals, objects, characteristics, behaviors,
judgements, and motivations’ (Plantinga & Smith 1999a, 3).
To appreciate how 21 Grams moves beyond the notion of ‘alignment’,
we should not pose an affective dimension of the film isolated from the
spectator’s cognitive endeavours. Instead, we may refer to research within

the study of the classical film, which maintains that the usual concep-
tion of this paradigm as fundamentally linear should be replaced with a
more ‘complex weaving together of anticipation/culmination structures
in which our emotional reactions to present events are just as important
as our anticipatory reactions to future events’ (Keating 2006, 7). According
to this view, Hollywood storytelling does not subordinate our emotional
response to the linear progression of the narrative such as it is popularly
claimed (cf. Bordwell et al. 1985). Yet, classical cinema matches our affective,
precognitive responses (the immediate arousal caused by the image) to
our emotional and cognitive responses.7 Consequently, it is useful to think
of ‘narrative not simply as a tool for the production of coherence but also
as a tool for the creation of emotions like hope, fear, delight, and despair’
(Keating 2006, 11).
In a classical fight scene we would typically experience an affective
incipience to act, which can be linked to the immediate task of the hero,
with whom we are also cognitively aligned due to his morally righteous
aspirations. The lack of any discrepancy glues our affective, emotional,
and cognitive responses together in a dynamic process that makes these
inseparable for the viewer. Yet, this also renders the manipulations made by
the film less visible. As spectators, we are inclined to think that our compas-
sion for the hero is based solely on our rational and cognitive appreciation
of his righteous ideals. Hitchcock is, perhaps, one of the earliest classical
directors to experiment with dissonances integrated into this seemingly
‘natural’ gluing together of the spectators’ affective-emotional-cognitive
circuitry.8
Because of his acute awareness of his viewers, and his willingness to play
with their responses, Hitchcock has rightly been perceived as a significant
precursor of the contemporary complex narrative. However, whereas Hitch-
cock only experimented with momentary breaks in the automatisms of the
affect-emotion-cognition circuitry, modern(ist) cinema went a step further
and ruptured this circuitry (or, in Deleuzian terms, the sensory-motor link-
ages). I would suggest that complex narratives, on the other hand, explore
various potential reconstructions of this circuitry, which emphasizes the
embodied – rather than merely cognitive-analytical – aspects of cinematic
spectatorship. From this perspective we can expand on a number of obser-
vations that Sandy Camargo (2002) has made in relation to ‘mosaic’ cinema.9
In brief, her thesis is that the ‘mosaic’ structure works primarily to suspend
rational and moral judgements of the characters. Instead of premature and
schematic judgements, spectators are encouraged to form more intuitive
and affective bonds with the characters. In this sense, the employment
of a ‘mosaic’ narrative structure expresses a desire for going beyond the
‘moral evaluations and judgments [that] frequently underlie our emotional
reactions’ (Smith 1999, 218). In other words, the ‘mosaic’ narrative structure
destabilizes the bond between our emotional and cognitive appreciation of
the characters to enhance the relation between affect and emotion instead.
In relation to the mosaic narrative of 21 Grams, Camargo’s explanatory
framework appears more rewarding than that favoured by Newman. In the
film, character complexity is more than the sum of information provided by
the film, because it also involves the emotional and affective investments
of the spectator. The narrative structure of 21 Grams thus prompts a more
intuitive than rational bond between character and spectator. Furthermore,
the visual style of the film – the grainy and washed-out imagery and the
unsteady handheld photography – makes it possible for spectators to
descend into the messy emotional states of the characters.
Noticeable in relation to this is the appearance of Iñárritu’s star ensemble,
which includes Naomi Watts, Sean Penn, Benicio del Toro, and Charlotte
Gainsbourg. Costume designer Marlene Stewart has explained that
Iñárritu constantly ‘stressed creating looks for the characters that didn’t
overwhelm the viewer, that didn’t force the viewer to jump too quickly to
conclusions about the characters’ (Stewart in ‘Production Notes’ 2003). In
de-romanticising his cast, Iñárritu possibly wanted to prevent spectators
from identifying with the star persona of the actors, thereby enabling us to
recognize them as human beings in the midst of deep personal suffering.
The characters instead become individuals who express a human suffering,
which is capable of uniting us beyond cultural, social, and emotional dis-
tinctions related to class, race, religion, etc. At the same time, the characters
demonstrate how human beings affect each other in both positive and
negative ways. This is a key for understanding the narrative form of 21
Grams, since the fragmented ‘mosaic’ narrative emphasizes the intercon-
nectedness between individuals that otherwise only share a geographical
space (cf. Azcona 2010; Pisters 2011b; Tröhler 2006).
As an example of how the film connects affect and emotion, consider
how 21 Grams engages the spectator with the characters in the scene at
the hospital where Cristina and Jack have eye contact and an ephemeral
moment of hope and forgiveness suddenly occurs. The scene is not primarily
effective because spectators cognitively designate this scene to be ‘about’
hope or forgiveness. The emotional tone of this scene has been carefully
prepared by the film’s fragmented narrative, since it has built up a tension
in our affective registering, which finally finds its release in this moment.
Thus, by obstructing schematic, cognitive responses or accommodation,

the film attempts to tie a more emphatic than cognitive-emotional bond
between characters and spectators. In a sense, we are allowed to embody
their pain and forgiving. Due to the ‘mosaic’ structure, the scene allows us
to acquire both of the characters’ perspectives at once.
As Camargo (2002) argues, (computational) cognitivism has been criti-
cized for marginalizing the role of empathy as ‘an involuntary response to
the film’s cinematic techniques rather than as a product of the conscious
processing of a film’s narrative’ (16). In a similar line of criticism, it can
be argued that Newman exclusively searches for character complexity in
terms of cognition and cognitive emotions, and that he in doing so ignores
the role of affect. In his later work, Murray Smith appears to have grown
increasingly influenced by the embodied perspective according to which
empathy reveals a far more intricate connection between affect, emotion,
and cognition than allowed for in the computational model of mind. Murray
Smith (2011), for instance, argues:
We might regard empathy as a mechanism of the coupling between the

mind and that part of the world through which it extends itself [...]. When
we empathize with another person, we extend our mind to incorporate
part of their mind. [...] In doing so, we exploit some part of the environ-
ment around us – in this case, another human being – and thereby learn
something about the environment. (108, emphasis in original)
Thus, it would be false to assert that 21 Grams ruptures the affect-cognition

linkages, and consequently cognition is not eschewed but granted a new
role in the reconstruction of the affective-emotional-cognition circuitry
of the ‘mosaic’ film.
Consequently, to push our understanding of the complexity of these films
it is important to acknowledge another prominent aspect of the ‘mosaic’
film – namely that the non-linear, acentric, and non-hierarchical narrative
starting point usually develops gradually towards coherence, stability, and
linearity. In relation to this, 21 Grams – although widely considered a radical
example of the non-linear and fragmentary style of the ‘mosaic’ film – is
no exception. Bordwell (2006) observes that
the Setup [of 21 Grams] is tantalizingly fragmentary, but the plot becomes
steadily linear, presenting more sequential scenes and fewer flashbacks
as it proceeds. We arrive fairly soon at a stable event frame: a fatal hit-
and-run shatters the lives of the driver, the victim’s wife, and the man
who gains the victim’s transplanted heart. 21 Grams achieves closure,

and it motivates this, in the approved manner, as at once random and
determined. (112, emphasis in original)
Following Murray Smith’s theory of empathy it is, furthermore, significant

that the moment of eye contact comes at a point in the film where a broader
context about the characters has been facilitated. This is the case, since
‘it is precisely the density of information available to us that is likely to
precipitate empathic imagining’ (Smith 2011, 115). At this point, we must be
careful not to fall back into the unidirectional classical ‘sandwich’ according
to which it is primarily our cognitive-emotional comprehension of the
narrative information that causes the sensation of empathy (cf. Chapter V).
Thus, it should be clear that affect plays a central role in preparing the
emotional discharge once sufficient narrative information is facilitated.
As such, I agree with Allan Cameron (2008), who argues that ‘modular’
narratives like 21 Grams ‘oscillate between instability and uncertainty on
the one hand, and schematic structures on the other’ (26).
A Problematic Combination: Classical Narratology and Complexity

Theory
Although Cameron (2008) counts the cognitive film science of Bordwell

among his main influences, he is not satisfied with the conception of con-
temporary cinema found here. According to Cameron, Bordwell ‘ignores the
way that such modular narratives address broader questions of time and
mediation’ (5). In his study of the ‘modular’ narrative, Cameron is primarily
interested in exploring aspects that relate to the modernist elevation of narra-
tive contingency and the postmodern explorations of chaos and order (17). In
relation to 21 Grams, Cameron argues that ‘[t]he interplay between linear and
non-linear time [...] parallels complexity theory’s articulation of determinism
and chaos, reversibility and irreversibility’ (53). This thesis is alluring not only
because 21 Grams makes explicit reference to chaos-theoretical thinking,
but also given that such references have been plentiful within contemporary
cinema more generally.10 Furthermore, complexity theory may prove produc-
tive for examining how 21 Grams reconstructs the role of cognition.
Cameron assures us that he is aware that classical narratology and
complexity theory have a different understanding of certain concepts such
as ‘determinism’. As Simons (2008) acutely sums up this difference: ‘In com-
plexity theory [...] determinism is associated with reversible (Newtonian)
time, whereas in the narratological accounts [...] determinism is strongly
associated with irreversible linear time’ (124, n. 2). Although Cameron

recognizes this incommensurability, he appears to be ignorant about a
much more severe disparity between classical narratology and complexity
theory – namely their different understanding of complexity and temporal-
ity. This can be seen in Cameron’s appreciation of the narrative continuum
as consisting of reversible ‘blocks’ of temporality. From this perspective,
‘modular narratives’ ‘present themselves as made up of discrete temporal
or narrative units, arranged in ways that gesture towards non-linearity’ (5).
The term ‘modularity’ – supposedly intended as a gesture to narrative
non-linearity – thus reflects a conception of the narrative continuum as
consisting of ‘modules’ that can be broken down and rearranged without
any qualitative change to the system. As Bergson has vigorously maintained,
this classical mechanistic conception is not able to capture real movement
or time (cf. Chapter II). Paradoxically, in complexity theory non-linearity
implies that a whole (e.g. the fabula) cannot be fully understood as the
sum of its components (e.g. its narrative ‘modules’) (cf. Cilliers 2002). Given
that it implies that the narrative is composed of independent, closed, and
domain-specific entities, the term ‘module’ is itself entrenched in the clas-
sical scientific ideals.11 Ultimately, Cameron’s narratological methodology
commits to an understanding of the narrative continuum as small segments
of reversible, decontextualized, and isolatable units of narrative time. It
is this understanding of narration that Cameron wants to pair with the
concepts of non-linearity, complexity, and irreversible time such as these
are used and applied in complexity theory.
That Cameron understands the narrative continuum in terms of revers-
ibility becomes even more evident, when he aligns the concept ‘modularity’
with Paul Ricœur’s idea of the ‘configurational dimension’ of narrative.
According to Cameron (2008), the latter ‘makes of the events a meaningful
whole, defines these events in relation to an ending and, via the repeti-
tion and recollection of the story, provides an alternative to linear time,
encouraging us to “read time backwards”’ (54; cf. Ricœur 1990, 68-69). Now,
if this kind of temporality can be opposed to linear time, it is only in the
limited sense as opposed to the forward progression of the ‘arrow of time’.
However, in complexity theory the arrow of time is not an expression of
linear time but a condition of irreversible (and thus non-linear) time. Yet,
for Cameron non-linearity implies an atemporal, or temporal symmetrical,
understanding of the narrative continuum. This, however, is no different
from the classical linear conception of time rejected by Bergson. From this
classical perspective, the directionality of time is irrelevant, because a
reversal of time here would not produce any significant material or physical
change (cf. Prigogine & Stengers 1984). Hence, Cameron’s idea of ‘reading
time backwards’ expresses exactly the classical scientific understanding of
temporal processes that has been challenged by the complexity sciences. In
his preface to Order out of Chaos, Alvin Toffler (1984) explains that within
classical science, ‘events begin with “initial conditions,” and their atoms
or particles follow “world lines” or trajectories. These can be traced either
backward into the past or forward into the future’ (xx).
Consequently, Cameron’s narratology shares its predilection for read-
ing time backwards with classical science – not with complexity theory.
Cameron’s classical methodology is, further, problematic since it lacks sen-
sibility for how irreversible processes nurture an ontological determinism
(cf. Morin 2007; Prigogine & Stengers 1984). As has been argued, the fabula-
syuzhet distinction comes with a comparable predilection for ontological
determinism, because it organizes the narrative continuum causal-linearly
(cf. Chapter II). As we are about to explore, the temporal detachment of the
spectator and/or analyst from the actual cinematic experience, which is
implied by the fabula, negatively influences Cameron’s (2008) treatment of
how the ‘non-linear narrative structure [of 21 Grams] allows for a height-
ened examination of causes, effects and coincidences’ (48).
The problematic consequences of Cameron’s classical narratology be-
come apparent, when he asserts that ‘the fact that we are given information
regarding the story’s culminating events at the very beginning of [21 Grams]
seems to imply that the future is determined in advance’ (73, emphasis
added). However, rather than implying an intrinsic narrative determin-
ism, 21 Grams is disposed to causal-linearity only if our methodological,
analytical, interpretative, or habitual mode of organizing experience is
already predisposed to such patterns (cf. Chapter III). Cameron’s analysis of
the scene, where Paul at dinner with Cristina explicitly evokes complexity
theory to explain the inexplicable contingency of them having met, helps
to elucidate the reasoning underlying Cameron’s approach. At this point,
Cristina is still unaware of the fact that Paul carries her husband’s heart.
Cameron argues that the scene can be taken as exemplary of how the
events of 21 Grams may appear contingent but are in fact predetermined.
According to him, ‘their encounter was effectively pre-determined rather
than accidental, as it was orchestrated in advance by Paul’ (58).
And yet is the extent to which their meeting appears contingent not
merely a matter of the amount of information available to them respec-
tively? Thus, for Paul the element of contingency rests in the fact that it was
exactly the heart of Cristina’s husband that saved him. It is hard to see how
the film implies this event to be explainable in reversible, causal-linear, or
predeterministic terms. Granted, for Paul the accident can be seen as a cause
for their meeting, yet to a certain degree the accident has also conditioned
their meeting for Cristina, who is now widowed. The predeterminist tone
that Cameron detects in 21 Grams appears to be the brainchild of the
backward reasoning that he utilizes to comprehend the scene.
As per the definition given earlier, backward reasoning can only operate
from a future perspective in relation to the past events (in this case the
fabula). This model dispenses with time, or rather; temporality becomes a
reversible function that allows for a structural organization of the system.
As Roland Barthes (1977) has acutely declared:
Analysis today tends to ‘dechronologize’ the narrative continuum and

to ‘relogicize’ it [...] the task is to succeed in giving a structural descrip-
tion of the chronological illusion – it is for narrative logic to account for
narrative time. To put it another way, one could say that temporality is
only a structural category of narrative (of discourse), just as in language
[langue] temporality only exists in the form of a system; from the point of
view of narrative, what we call time does not exist, or at least only exists
functionally, as an element of a semiotic system. (99, emphasis in original)
The vital point to be taken from this is that the backward reasoning so
profound to narratology depends upon an abstract suspension of time from
the system. As Simons has argued, it is exactly the atemporal conception of
the narrative continuum that allows ‘database’ narratologists like Cameron
to reason forwards or backwards. The price of this is a serious reduction of
temporal complexity since the narrative must first be made to conform to
the reversible structures of classical science (cf. Simons 2008, 119). In relation
to this, we may recall that Morin’s (2007) first principle of how classical
science has traditionally reduced complexity is the principle of universal
determinism associated with Laplace (cf. Chapter I).
It is by necessity of this temporal reduction that narratologists perceive of
the past as a series of necessary events. In so doing, they systematically rule
out all the virtual possible – yet never actualized – outcomes, which could
have resulted from the initial condition. Applied to narrative comprehen-
sion this logic is often accompanied with a lack of sensibility to questions
that examine ‘what could have happened’. According to Ben Shaul (2012),
the canonical story format discourages ‘optional thinking’, i.e. it actively
discourages spectators from comprehending the narrative along the lines
of ‘what-could-have-happened’ (cf. Chapter III). Cameron’s analysis would
seem to imply that 21 Grams – as its events are causal-determined – in
this regard is ‘closed-minded’. However, I will argue that the fragmented,

non-linear, and ‘mosaic’ narrative structure of the film in fact encourages
‘optional thinking’.
The accident, around which 21 Grams circles, is an irreversible event
that completely changes and turns upside down the characters’ entire life
situations: Cristina goes from a core family existence to excessive drug
abuse, Jack suddenly turns against God, and Paul no longer finds meaning
in his relationship, or, in life as such. Cristina, Jack, and Paul are in different
manners forced to make meaning out of an ultimately meaningless event.
Relentlessly playing the last message she received from her family on the
answering machine, Cristina cannot understand why the accident had to
happen. Similarly, if ‘God knows every hair that turns on your head’, as
Jack tells a student in a scene before the accident, why must he experi-
ence so much suffering and guilt? Faced with the pure contingency of life,
Jack’s Christian worldview starts to crack. Instead of believing in God, Paul
‘believes’ in fractals. He explains to Cristina that ‘there is a number hidden
in every act of life, in every aspect of the universe [...] Numbers are a door
to understanding a mystery that is bigger than us’. It is this conviction
that leads him to Cristina, since he perceives the accident to be a message
stemming from the mysteries of the universe. However much the characters
attempt to induce a higher meaning to the accident, the fact remains that
the accident is irreversible (the characters cannot prevent it from having
happened). This nevertheless does not mean that the accident itself and the
events that follow from it have been causally predetermined (the accident
is not the outcome of causal chains already set in motion).
The narrative structure of 21 Grams, therefore, rather than encouraging
simplistic backward reasoning to reconstruct the one true, linear path
of the fabula (the narrative structured around the analytical fabula),
allows us to explore all the virtual pathways of the narrative continuum
– i.e. it allows us to embody the fabula. Rather than suggesting that the
events are predetermined, the fragmented narrative structure of the film
allows the spectator to consider the available paths, which were never
realized, but which could have given the narrative a completely differ-
ent outcome. Recall Ben Shaul’s contention that the spectator’s strive to
achieve narrative closure results in a cognitive and affective bias that
leads to premature acceptances of inadequate or incorrect hypotheses
as discussed in Chapter III. Could this not exactly be argued about the
analytical proceedings that come with the cognitive-formalist concept of
the fabula? Once taken as a representation of the actual story (rather than
as a mode of organizing experience), the fabula reduces the temporality of
the narrative continuum to a coherent causal-linear trajectory. It would

thus be the linear, interpretative tool of the classical narratologcial fabula,
and not as Cameron (2008) has it the ‘modular’ narrative, that is ‘flirting
with determinism’ (48).
Cameron uses complexity theory as a metaphorical and conceptual
toolbox because he recognizes the shared interests of contemporary cinema
and modern science. He, however, does not allow complexity-theoretical
insights to reflect on his own narratological method (cf. Simons 2008, 117).
This has unfortunate consequences, since the discrepancies are too grave
between Cameron’s narratology and complexity theory. Eventually, instead
of moving beyond the linear-non-linear dichotomy, Cameron falls into a
grey area between them. As a result, Cameron’s theory of the ‘modular’
narrative has been perceived as ‘the best example of how good old-fashioned
narratological thinking gets in the way of understanding the role and mean-
ing of temporality in the new sciences of complexity’ (Simons 2008, 117).
Alternatively, the embodied fabula allows for an appreciation of the virtual
pathways of 21 Grams. Instead of regarding 21 Grams as a series of causal-
linear and predetermined narrative events, the embodied fabula perceives
the narrative continuum as a series of complex and irreversible events
open for virtual explorations. In this fashion, this concept resonates with
recent attempts in historical studies to loosen the causal-linear conception
of past events.
Not only the study of narratives, but also the study of history, has tradi-
tionally sought to make sense of past events through a strong adherence to
backward reasoning. In history, the past is inevitably considered from the
temporally detached future perspective of the historian, who constructs
a line of important events that find a (possible) endpoint in the present
situation. To counter this methodological approach, certain historians
have proposed a ‘virtual’ alternative to the study of history (cf. Ferguson
1999). The aim is to complement the retrospective backward reasoning of
traditional historical approaches with a ‘virtual’ approach that takes the
perspective – or demands that the analyst embeds himself in the situation
– of the historical protagonists to consider ‘those alternatives which we can
show on the basis of contemporary evidence that contemporaries actu-
ally considered’ (Ferguson 1999: 86; original emphasis omitted). ‘Virtual’
historians thus compensate for their temporal displacements from the past
events by placing themselves in the midst of events, from which history can
be explored not merely from the point of view of ‘what actually happened’,
but also from the plausible, yet never realized, alternative paths of what
‘could have happened, but never did’. It is exactly their ability to allow
spectators to explore the virtual pathways of the narrative that has caused
Simons (2008) to perceive the complex narrative as the ‘filmic equivalent
of “virtual history”’ (122).
In 21 Grams the fragmented opening of the f ilm provokes a recon-
f iguration of the affective-emotional-cognitive circuitry, because the
incipience to action invoked by the images can neither find an emotional
nor a cognitive release. It, in other words, ‘short-circuits’ our perceptual
habit of selecting images that interest us only for potential action (Cole-
brook 2002, 40; Powell 2007, 3; cf. Chapter II) In this sense, the film opens
up the affective dimension by enabling spectators to explore the virtual
dimension of the images and those narrative pathways that could have
been actualized, yet never were. In relation to this, the concept of affect
in the work of Deleuze, Massumi, or Shaviro becomes an instrumental
tool, since it opens up an exploration of the virtual pathways induced by
the cinematic experience.
The affect created by the narrative structure of 21 Grams influences
the manner in which spectators comprehend the plot, the characters, and
how they emotionally connect to the narrative. In this manner, the film
integrates affect into its plot mechanics that, again, build the affective
landscape into which the film’s narrative is absorbed. As a result, it would
be mistaken to argue that 21 Grams privileges affect over emotion and
cognition. More precisely, the film reconfigures their relation to explore
new venues of cinematic spectatorship. Its fragmented non-linear structure
introduces a desire for meaning, order, and chronology, which can only
find satisfaction in an exploration of the virtual pathways of the film.
Furthermore, the narrative structure of 21 Grams is capable of inducing a
cognitive awareness of how we make sense of images, when the circuits of
affect, emotion, and cognition operate smoothly, i.e. in what we have termed
‘linear cinematic perception’. The film initially immerses the audience
in a narrative world best described by Iñárritu as an ‘immense universe
of circumstances’. Spectators will take this as an invitation to ‘solve’ the
narrative by performing cognitive tasks such as hypotheses testing, sche-
mata application, and establishing patterns and connections in order to
reconstruct the narrative according to the analytical fabula. Gradually,
patterns and linearity will emerge and bestow meaning upon the chaotic
narrative universe. In this sense, the affect-laden, fragmented narrative
structure of the film intensifies a desire for meaning and order.
At the end of the film, Duncan MacDougall’s theory that the human
soul in average weighs 21 grams is introduced. Spectators will react dif-
ferently to this emotionally and rationally. On the one hand, the theory
introduces that glimmer of hope and profound meaning of life that the
characters and spectators (in their attempts to make sense of the story)
have been yearning for. The theory of the weight of the human soul can
thus be taken as a form of explanation that provides narrative coherence
and clarity. The paradox, however, is that this can only be achieved if
spectators are willing to accept a theory whose rational and scientif ic
value is contentious. In such cases, our incipience (affect) towards meaning
(cognition) is so dominant that ‘believing becomes seeing’.12 The 21 grams
of the film’s title can be seen as providing narrative resolution, meaning,
or closure. However, the 21 grams can also be seen as an expression of
the impossibility of measuring the human ‘soul’ and as a testimony to
what will remain forever beyond human comprehension. As such, the title
reflects the film’s preoccupation with death as the place of both absolute
contingency and eternal meaning. The novel achievement of 21 Grams lies
in how it allows its spectators to explore these ‘eternal questions’ through
an embodied, immersive, affective, emotional, and cognitive experience
of the narrative.
‘Forking Paths’ – Complex Storytelling in Lola rennt
In Jorge Luis Borges’ short story The Garden of the Forking Paths, the mys-
terious figure Ts’ui Pȇn has created a labyrinth of time rather than space.13
In cinema, it can be argued that Ts’ui Pȇn’s labyrinth has materialized
in its own rights in a series of complex narratives identified as ‘forking-
path’ narratives. Characterizing these films is the narrative thread that
at one or several points departs from its linear unfolding and bifurcates,
thereby showing the different virtually possible, yet mutually exclusive,
outcomes of a given event. In Krzysztof Kieslowski’s Blind Chance (Pr-
zypadek 1987), to take an example, the main character Witek (Boguslaw
Linda) runs after a train. From here, the narrative forks into three possible
futures in each of which Witek’s life has developed in surprisingly diverse
directions. Another variation of the ‘forking-path’ narrative is Groundhog
Day (Ramis 1993) in which a pessimistic weatherman (Bill Murray) is forced
to relive the worst day of his life until he ‘learns’ to embrace the joyful-
ness of the day. In narrative terms, both films – initially at least – appear
to challenge the idea of the unified and linear progression of narration.
This is the case since the construction of a coherent fabula is rendered
problematic, because not all bifurcations, or ‘forking paths’, are possible
at once. It is, therefore, tempting to understand ‘forking-path’ narratives
as a subcategory of the Deleuzian regime of time-images insofar as they

reveal ‘the simultaneity of incompossible presents, or the coexistence of
not-necessarily true pasts’ (Deleuze 2005b, 131). However, ‘forking-path’
narratives do not dispense with cinematic linearity in a similar radical
fashion as time-images supposedly do. For reasons we are about to explore,
not ‘every model of truth collapses’ (131) in these narratives that are instead
located somewhere between the movement-image and the time-image
(cf. Martin-Jones 2006).
In relation to this, ‘forking-path’ narratives have often been studied as an
example of how complex narratives – despite incorporating certain char-
acteristics of the time-images – must be understood as belonging to the
linear paradigm of canonical Hollywood storytelling. In the ‘forking-path’
narratives of contemporary cinema, Bordwell (2002a) argues, ‘linearity
helps make these plots intelligible’ (92), while David Martin-Jones (2006)
claims that these films ‘bear testament to the reterritorialising power of
the movement-image’ (85). Given the fact that the two scholars in ques-
tion depart from the polarized theoretical frameworks of cognitive film
science and Deleuzian film-philosophy, it is startling that they arrive at
such similar conclusions. For both, cinematic ‘forking paths’ provide merely
a glance of the Borgesian labyrinth of time, which has been thoroughly
domesticated by the reterritorializing powers of the movement-image
associated with classical Hollywood. Interestingly, both authors resort to
the German experimental box-office success Lola rennt as an example
of how classical linearity pervades ‘forking-path’ narratives. The aim of
the following is to demonstrate that their respective treatments of the
film are restricted by a linear-non-linear dichotomy that the film itself
has left behind.
With extravagantly red-dyed hair and a trashy, punkish look, the title
character (Franka Potente) of Lola rennt runs the streets of Berlin to
the persistent heart-beating sound of the film’s techno score. Lola is run-
ning to save the life of her boyfriend Manni (Moritz Bleibtreu) – a foolish
antihero, who lost possession of a bag containing DM 100,000 that he was
supposed to deliver to his merciless gangster boss. Passive and incapable
of even the simplest tasks, Manni is contrasted to Lola, who literally and
visually sets things in motion, trying to get hold of the money within a
20-minute deadline. If Lola is not back in time, Manni has proclaimed to rob
the nearby supermarket. In the first of Lola’s three runs, she fails to reach
Manni in time, who then decides to rob the nearby supermarket to obtain
the money. She nevertheless does come in time to help him, and the couple
manages to get away with a large amount of money, but quickly thereafter
find themselves surrounded by the police. As Manni throws the bag with
the money to a police officer, the latter’s firearm is accidently triggered
and Lola is shot in the breast. In the second attempt, Lola fails to obtain
the money but manages to reach Manni in time. However, this time the
timing is wrong and Manni is accidently hit and killed by an ambulance. In
the third run, both Manni and Lola are successful in retrieving the money.
The ironic twist is that both are successful by chance: Lola wins on the
roulette and Manni accidently encounters the homeless person, who is still
in possession of his bag.
The relatively simplistic set up allows for the film’s complex drama-
turgical form and style. Besides from splitting into Lola’s three attempts
to retrieve the DM 100,000, the film entails an abundance of different
visual modalities (such as 35 mm, animation, black and white images,
red monochrome images, and grainy video images). These further engage
with a manifold of devices for orchestrating or manipulating cinematic
temporality (such as slow and fast motion, jump cuts, rhythmical ed-
iting, the use of split screens, etc.). Some scholars have attempted to
trace out a consistent logic able to connect the visual modalities to the
various temporal layers of the film, or, to understand these as a mode
of character focalization (Evans 2004, 107-108). Others have argued that
the film as a whole is consistent only in its constant complicating and
disorienting of any straightforward attempts to map out one-to-one
relationships between the visual modalities and the temporal layers of
the film (Wedel 2009, 131). Another prominent device for orchestrating
the temporality of the film is the deadline that is used to structure each
of Lola’s runs. A classical narrative device thus complements the more
experimental device of the bifurcations. Whereas the latter ensures
repetition and temporal non-linearity, the former provides the film with
a persistent and goal-oriented forward movement. In combination, this
mixture alludes to the narrative structure of a variety of computer games
in which the player is granted three lives/attempts to complete a level
within a given time slot.
Each run is separated by a scene – functioning as an interlude – of
the couple lying in bed. These scenes are stylistically and thematically
comparable to Hiroshima mon amour. Not least because the lovers lie
entangled, naked in bed in what suggests a pre- or post coital situation,
which marks the temporality of the scene as one ‘in-between’ events
(cf. Chapter IV). Like the lovers of Resnais’s modern(ist) landmark,
Lola and Manni talk about the contingency of their love affair. Their
conversation, too, reflects upon the film’s narrative form; however, in a
quite different way than is the case in Hiroshima mon amour. The main
difference is that Lola rennt – unlike Hiroshima mon amour – is not
constructed as a criticism of linearity, representation, narration, or history
(cf. Chapter IV), but, perhaps, rather expresses a certain mode of being
in the world. As Bianco (2004) has pointed out, this particular mode of
being in which ‘the image and we are always running, always movement
in play and always opening onto the emergent’ (378) far more resembles a
narrative game. The film introduces the metaphor of the narrative game
already in the opening prologue in which a man (later to appear in the
film as a security guard working in the same bank as Lola’s father) states:
‘The ball is round. The game lasts 90 minutes. That’s a fact. Everything
else is pure theory’. He then proceeds by kicking a ball high up in the air
to gesture that the game can now begin. What follows is a 2D animation
sequence of Lola running through a tunnel, which graphically resembles
a computer game.
Even more explicitly than 21 Grams, Lola rennt incorporates concepts
derived from complexity theory into its dramaturgical form. Especially
prominent is of course the idea of bifurcations. However, also the butterfly
effect, in which the proportions of cause and effect relate to each other in
a non-linear fashion, is prominent in the form of the film’s flashforwards.
These appear in different variations in each of Lola’s runs and reveal the
different outcomes pertaining to the minimal changes in Lola’s encounter
with the people that she quite literally runs into. The flashforwards indicate
that each moment of the narrative is a potential starting point for a bifurca-
tion, which, if it would be followed instead, would steer the narrative in
another direction. It is worth noting that in some of the bifurcations the
future scenario of the sub-characters is positive while it is tragic for Lola.
Simultaneously, in the third run Lola encounters a woman whose virtual
future is tragic, while Lola finally succeeds in solving her task (or completing
the level).
A question that has been perceived as a significant key to the comprehen-
sion of the film is how to interpret the third and final run dramaturgically.
Is it to be perceived as a narrative resolution in which the ‘right’ path has
been chosen, or, is it rather an affirmation of the pure contingency of life? It
should be obvious that the ‘either-or’ framing of this question is formulated
alongside the classical-modern(ist) dichotomy, since it inevitably desires to
categorize the film in continuation of the linear logic of classical cinema or
to position it among the ‘pure’ non-linear artworks of modern(ist) cinema.
In relation to this, Bordwell has, perhaps not surprisingly, argued that the
film adheres to the linear principles of canonical storytelling.
Are Forking Paths Linear?
For Bordwell, Lola rennt is not merely another example of the general
prevalence of classical storytelling in cinema, it is also evidence of its per-
sistency within contemporary complex narratives. In an interview, Bordwell
sums up his understanding of Lola rennt concisely:
A movie like Lola Rennt, for instance, which is very experimental in

some ways, is in many ways also very traditional. Beginning-middle-end,
she gets three chances, the last one is the right one [...] I mean this is
very much in the spirit of classical cinema. (Bordwell in Donecker 2005)
Bordwell (2002a) proposes that in the forking paths of cinema, the Borgesian
labyrinth of time has been ‘trimmed back to cognitively manageable dimen-
sions [...] designed for quick comprehension’ (91). He traces out seven conven-
tions whereby this is achieved. In the following, three of these – particularly
relevant for the recuperation of linearity in Lola rennt – are presented
and discussed. These conventions are as follows: #1) ‘Forking paths are
linear’ (92); #6) ‘All paths are not equal; the last one taken presupposes the
others,’ (97) and #7) ‘All paths are not equal; the last one taken, or completed,
is the least hypothetical one’ (100).
#1 is connected to Bordwell’s contention that ‘forking-path’ narratives
‘are built not upon philosophy or physics but folk psychology’ (90). Despite
the f ilms’ frequent and often explicit references to philosophy and/or
science, Bordwell argues that these theories can shed only modest light
on narrative conventions insofar as the latter are customized to fit ‘the
ordinary processes we use to make sense of the world’ (90). In the end, the
highly complex and abstract theories of modern physics and philosophy,
so Bordwell’s thesis, simply do not conform to the cinematic requirements
for painless consumption:
Storytellers’ well-entrenched strategies for manipulating time, space,

causality, point of view, and all the rest reflect what is perceptually and
cognitively manageable for their audiences, and the multiple worlds of
Borges and quantum mechanics don’t fit that condition. (91)
Bordwell is justified in claiming that the complexity of the philosophical

and scientific concept of ‘multiple worlds’ exceeds the boundaries of what
is cognitively comprehensible in narratives. It is indeed hard to imagine a
cinematic narrative that would do absolute justice to the complex idea of
multiple and equally possible futures. This is mainly the case, since it would
require the actualization of the virtual pathways into genuine, linear, nar-
rative trajectories. Hence, doing absolute justice to the complexity entailed
in the short story is a false requirement for determining how these films
deal with the Borgesian thought experiment. Bordwell seems to accept that
narratives must always oscillate between convention and transgression,
thus his actual line of inquiry is also not whether ‘forking-path’ narratives
actually realize a Borgesian labyrinth of time or not. The question he poses
instead is whether these films open up to the complexity of ‘multiple worlds’
or rather enclose them in a linear, classical worldview that conforms to
the idea of the one necessary and ‘true’ path. Bordwell believes the latter
to be the case.
Bordwell’s convention #1 (‘forking paths are linear’) describes how each of
the narrative forks ‘adheres to a strict line of cause and effect’ (92). Although
the overall linear trajectory may split, each of the bifurcations eventually
establishes its own linear trajectories. Bordwell assumes that the mode
of causality that pertains to each of the bifurcations is identical with the
causal-linear narrative drive of the classical film. However, it should be
noted that the fact that linearity can be restored after each bifurcation is
not necessarily a contradiction to the temporality of complexity theory.
Prigogine & Stengers (1984) explain that it can be expected that ‘near a
bifurcation, fluctuations or random elements would play an important
role, while between bifurcations the deterministic aspects would become
dominant’ (176). In other words, the contingency or non-linearity is consti-
tuted in the moment of bifurcation, whereas more causal-linear patterns
dominate each fork, when these can be observed in isolation, i.e. once the
inherent potentiality for further bifurcations of each fork is reduced from
the equation. In relation to this, the flashforwards of the film operate as
constant reminders of the endless virtual possibilities of each narrative
trajectory. Admitted, it is quite common for ‘forking-path’ narratives only
to visualize a limited amount of forks (often even the minimum of two),
and for Bordwell (2002a) this means that in Sliding Doors (Howitt 1998)
and Blind Chance the ‘narratives assume that one moment of choice or
chance determines all the rest’ (92). Although it might be the intention of
some films to invoke such an interpretation of its events (the early, more
classical ‘forking-path’ narrative It’s a Wonderful Life [Capra 1946] comes
to mind), one must be sceptical towards proclamations about the intrinsic
nature of the narrative itself (as isolated from the spectator/analyst).
Another issue with Bordwell’s theory is that it does not distinguish be-
tween different modes of causation, or, more precisely, between the linear
causality associated with classical cinema and the structural multicausality

of complex narratives (cf. Shaviro 2013). The flashforwards of Lola rennt
involve causality, but this causality is entangled in an infinitely complex
system of networks and patterns. It is, therefore, to be differentiated from
the causal-linear trajectories towards absolute knowledge that Bordwell
has argued characterize classical storytelling. In the same stroke, this mul-
ticausality diverts from the classical scientific understanding of linearity,
since it no longer implies a proportional relation between cause and effect.
The causality of Lola rennt instead forms part of a complex or non-linear
system, which is governed by a ‘startling incongruity between cause and
effect, so that a small cause can give rise to a large effect’ (Hayles 1990, 11).
Thus, although Lola rennt facilitates cognitive comprehensibility, it does
not promote determinist causal-linearity. Furthermore, as Bordwell (2008)
has argued elsewhere, the narrative device of the flashforward is ‘unthink-
able in the classical cinema’ (155).
Nevertheless, Bordwell is right to observe that the deadline device is
used to orchestrate each run temporally in a manner that is comparable
to its standard use in the classical film. However, this alone does not mean
that Lola rennt adheres to the causal-linear logic typically associated
with classical cinema (notwithstanding the question of whether this logic
is an inherent attribute of the classical film as Bordwell and several others
have suggested, cf. Chapter III). In any case, the difficulty pertaining to
Lola rennt is not the identification of a well-known temporal mode of
orchestration, but is instead to be found in the complex interplay between
the diverse temporal modes of the film. As Wedel (2009) has argued:
[T]he temporalities within each individual episode are in themselves

far too complex and convoluted – perforated and punctuated as they
are by jump cuts, discontinuities on various levels, internal repetitions,
snapshot bits and pieces, slow and fast motion shots – for one to really
speak, as Bordwell does in his first definition, of each forking path as
being organized in a strictly ‘linear’ fashion. (135)
In other words, even if certain events relate to one another in terms of

causal-linearity, it does not automatically mean that the overall temporal
design of the film is pervaded by this logic. Bordwell’s argument, however,
appears to become more persuasive with the introduction of his two last
conventions.
Bordwell’s convention #6 (‘All paths are not equal; the last one taken
presupposes the others’) is concerned with the range of knowledge that
characters and spectators have across the multiple worlds of the narra-
tive. Bordwell (2002a) asserts that ‘[t]he future shown first supplies some
preconditions for later ones, always for the audience and sometimes for the
character’ (99). That is, the information that spectators receive first will
influence how they experience later events. This, as Bordwell explains,
is ‘due to the exigencies of telling in time’ (99).14 Bordwell here touches
upon the irreversible nature of the cinematic medium, which it has been
argued cannot be equated with causal-linear reversible time. However,
this convention also concerns those incidents where characters, too, can
accumulate knowledge across the parallel universes. When this is the
case, the film introduces a linear, narrative trajectory that cuts across the
dispersed narrative continuum, and gives it a uniform direction towards
a final resolution. In Bordwell’s interpretation of the film, Lola ‘learns to
control the chance that ruined her previous futures’ (99).15
In combination with convention #7 (‘All paths are not equal; the last one
taken, or completed, is the least hypothetical one’) Bordwell has collected
strong support for the thesis that the Borgesian labyrinth is domesticated
by classical linearity within ‘forking-path’ narratives. The question is how
this linearity was achieved. In other words, does the narrative encour-
age spectators to single out the last path taken as the least hypothetical
one? Alternatively, is Bordwell analytically predisposed to single out a
privileged path due to the cognitive bias towards linear organization of
his conceptual framework? Considering Bordwell’s constructivist theory
of narrative comprehension, it is startling that he assumes linearity to be
an intrinsic quality of the narrative itself. If this is not to be perceived as
merely a personal interpretation but an intrinsic quality of the narrative
itself, it would be fair to assume a broad consensus that Lola’s last run should
be taken as the one ‘true’ path. This, however, is not the case.
Elsaesser & Hagener (2010), for instance, count Lola rennt as a ‘post-
mortem film’, where ‘while the body is (un)dead, the brain goes on living
and leads an afterlife of sorts or finds different – ghostly, but also banal,
mundane – forms of embodiment’ (165). Brockmann (2010), following a quite
different string of arguments, denies that the film’s trajectory replicates that
of the classical film, because its characters do not grow more introspective,
reflective, or enlightened over the course of the narrative, instead
the film places Lola in motion, and the point of the movie’s repetitions
is not so much to come to some sort of ultimate enlightenment as it is
the pleasure of the game itself. ‘Everything else,’ as the uniformed bank
guard in the movie’s prologue had declared, ‘is theory’. (461)
Finally, Kovács (2007) denies that any of the three bifurcations should be
regarded as the most privileged or ‘true’ one, thus denying that Bordwell’s
convention #7 is at work in the film. He instead argues that
Lola’s story exists in three alternative versions each as plausible as the

others. The role of chance here is not to confirm the rule of order by show-
ing that what should happen happens anyway, like in classical narrative,
nor to demonstrate the dramatic disaster caused by unpredictability, as
in modernist narratives that show what should happen accidentally does
not happen, or what should not happen happens accidentally. Tykwer’s
film wants to show that nothing that happens happens because that is
the way it ‘should be’. (76)
It is noteworthy that Kovács believes the film to be demonstrating that

nothing happens because it had to happen, which implicates a total rejection
of the causal-linearity that Bordwell perceives to be the constitutive logic
of the film.
Alternatively, it could be argued that the linear principles that constitute
the logic of Bordwell’s methodology become apparent in his interpretation
of the film. Bordwell first dissects the film into three smaller units, which
he then relates to one another causal-linearly. This method, however, as-
sumes the narrative continuum to be understandable in linear terms from
the very outset. In this manner, Bordwell forces causal-linearity onto the
narrative whose intrinsic logic he aims to observe objectively. In relation
to this, Mitchell (2002) has acutely contended that the difficulty Bordwell
faces in his treatment of Lola rennt, is that ‘[t]here is no problem in
understanding each story; it is their relationship that is confusing. Seeing
the whole as more than just the sum of the three separate parts is the
cognitive problem’ (5).
Bordwell’s interpretation of Lola rennt depends upon isolating and
separating cognitive difficulties from one another (cf. Morin 2007), only
to reunite them again in a linear order. In relation to this, Elsaesser (2009)
has argued that in the hands of Bordwell’s causal-linear methodology ‘the
para-normal features are given normal explanations, and the narratives are
restored to their “proper” functioning’ (21). This can be seen in extension
of the discussion of Michotte’s experiment with how humans introduce
causality to the system even in cases where there by definition is none
(cf. Chapter V). The argument is not that causal-linearity is extraneous to the
comprehension of narratives, but rather that we should be wary about con-
tentions that this exists as an intrinsic value of the narrative itself. Bordwell,
I believe, rightly argues that narratives consciously play with predictable

viewer patterns of making shortcuts, expecting stereotypes, making faulty
inferences, or jumping to erroneous conclusions (cf. Bordwell 2002a, 90).
However, could causal-linearity not in certain instances be counted among
such ‘errors’ to which we are prone due to our predisposal for organizing
our surroundings in easily accessible terms? In any case, making us aware
of our biases towards causal-linear organization is a recurring element that
can be found in complex narratives.
Embracing the Narrative Rhythm of Lola rennt
A predisposition to causal-linearity is not only to be found within cognitive-

formalist informed approaches to contemporary complex narratives.
From his film-philosophical perspective, Martin-Jones (2006) has located
a surge of ‘hybrid films’ within contemporary cinema – hybrid, because
these films ‘contain characteristics of both the movement- and the time-
image’ (39). Martin-Jones sets out to examine how within these f ilms
‘the [movement-image and time-image] interact as a mutual struggle of
deterritorialisation and reterritorialisation’ (4). Whereas the first ‘enables a
displacement of narrative into multiple labyrinthine versions’ (4), the latter
‘entails a constraining of a narrative into one linear timeline’ (4). The logic
of categorization at work here is simple:
Whether an image is a movement- or a time-image depends on the degree

to which it de- or reterritorialises time. The closer it is to establishing a
linear narrative, the more likely it is to be a movement-image. By contrast,
the more visible the labyrinth, the closer to the time-image. (27)
Rather than being interested in this mutual struggle as a dynamic and crea-
tive force in itself, Martin-Jones aspires to determine the degree to which
the selected films either de- or reterritorialize time. Exemplifying how
within contemporary cinema ‘the labyrinthine possibilities offered by the
time-image are caught in the process of reterritorialisation as movement-
images’ (113), Martin-Jones has selected two ‘forking-path’ narratives: Lola
rennt and Sliding Doors. Martin-Jones declares that he will ‘use David
Bordwell’s article, “Film Futures” [...] to demonstrate how the overriding
narrative logic that [Sliding Doors and Lola rennt] manifest is that of the
movement-image’ (85). Bordwell and Deleuze are thus united in the task of
categorizing cinema according to the linear-non-linear dichotomy. Martin-
Jones shows little concern for the discrepancies between the cognitive
and film-philosophical perception of cinema. In fact, Bordwell’s ideas on

the ‘forking-path’ narrative are assumedly translatable into a Deleuzian
vocabulary, insofar as
[i]n Deleuzian terms, Bordwell is observing how classical narrative

devices are used to order the multiple narratives of these films into the
coherent, linear schema we expect of a movement-image. Thus, their
labyrinthine visions of time are mapped in a user-friendly way, once
reterritorialised within the parameters of the movement-image.16 (87)
In effect, Martin-Jones combines Bordwell’s linear narratology with a

Deleuzian valorization of Bergsonian durée and the time-image in his
treatment of Lola rennt.
Scholars have often pointed out the implicit favouring of the time-image
in Deleuze’s cinema books (e.g. Mullarkey 2009; Rodowick 2003; Smith 2001).
This valorization of modern(ist) ideals has nurtured the establishment of
a clear hierarchy between the reterritorializing, non-linguistic, affective,
non-linear, intensive, non-narrative, artistic, and aesthetic qualities of the
time-image, and the conventionality, linearity, commerciality, easy com-
prehensibility, focus on cognitive problem-solving, and the epistemological
classicism of the movement-image. This has nurtured a too simplistic mode
of apprehending the differences of the movement-image and time-image.
An example of this is Martin-Jones’ comparison of Lola rennt to Resnais’s
L’a nnée dernière à Marienbad (1961):
Despite [Lola rennt’s] apparent use of deterministic chaos [...] the

labyrinthine possibilities this model offers are actually restrained by
the linearity of a much more classical view of time. Unlike previous films
such as L’année dernière à Marienbad (1961), which sacrificed the
clarity of a classical narrative in order to experiment with a labyrinthine
view of time, Run Lola Run – for entirely understandable, commercial
reasons – owes its financial success to a reterritorialised representation
of chaos within fairly classical (both narrative and time) parameters.
(Martin-Jones 2006, 112-113)
Ultimately, it becomes evident that Martin-Jones has chosen an approach to

cinema that stresses the differences between the classical and modern(ist)
and allows him to advocate for the superiority of a specific form of cinematic
temporality associated with the philosophy of Bergson. Due to the proces-
sual nature of cinema, the devices associated with this form, however, have
themselves been reterritorialized – or at least been conventionalized – as

modern(ist) cinematic devices have been incorporated into mainstream
cinema (cf. Mullarkey 2009, 10).
According to Martin-Jones (2006), Lola rennt ultimately adheres to
the linear logic of classical Hollywood narration. To support this claim,
the author singles out an illustrative image that ‘encapsulates the rationale
behind the film’s linear sequencing of its narratives’ (112). The image in
question is a split-screen divided in three. On the left side of the image,
Manni is waiting for Lola. If she fails to show up in time, he will rob the
nearby supermarket. On the right side of the image, Lola is seen sprinting
full speed ahead to reach Manni in time. Finally, in the lower part of the
image, a clock is ticking. According to Martin-Jones, the clock is perceived
as an objective and visual representation of time, and more precisely of the
ticking deadline that orchestrates the time pressure conveyed in the scene.
In this fashion, the film reterritorializes time in the concerned split-
screen image:
On one side of the screen, as Manni waits on his fate, time passes slowly
for him; on the other, for Lola running to beat the clock, time passes
almost too quickly. For the spectator, however, their relative movements
appear to be seen objectively, due to the presence as measure of time of
the hands of the clock that appears at the top of the screen. (112)
Hence, ‘the spectator is given the illusion of objective, spectatorial mastery

over the relative experiences of time lived by the characters’ (112). This
illusion relates to the manner in which the image renders different incom-
mensurable temporalities commensurable. That is, the presence of the
clock falsely suggests that Lola’s and Manni’s personal temporalities can be
mediated by an objective representation. Thus, Martin-Jones argues that
the true logic of the scene is causal-linear, although it has been disguised
in the form of Bergsonian durée. Therefore, the scene is exemplary of how
the film more generally domesticates the Borgesian labyrinth of time into
the linear trajectory of ‘two “wrong” narratives and one “right” one’ (112).
According to Martin-Jones, the spectator is constantly provided with a
detached narrative perspective since it is through ‘the fixing of the first
story as the originary model that we view the effects on the others as due
to Lola’s relatively faster or slower flight’ (112). In essence, Martin-Jones’ and
Bordwell’s arguments here amount to the same: the Borgesian forking paths
conform into a single, overarching, linear, narrative trajectory from which
‘the observer can judge the relative merits of the others’ (111).
Again, I am not convinced that the emergence of a single, ‘true’ temporal

logic can be considered an intrinsic quality of the narrative itself. Rather,
in adopting a Bordwellian interpretative strategy, Martin-Jones fails to see
the extent to which he himself imposes a linear logic onto Lola rennt. Due
to its multimodal temporality, the film in fact appears to be questioning
Martin-Jones’ presumption that a narrative must adhere to only one, rather
than several, temporal logics. In this fashion, the film actively undermines
Martin-Jones’ endeavour to establish the ‘true’ underlying logic behind
the appearances of the film. That Martin-Jones despite being theoretically
informed by Bergson and Deleuze chooses a narratological approach that is
rooted in the classical linear onto-epistemology (cf. Chapter II), evidences
the lack of a genuinely film-philosophical approach to narration.
In his brief treatment of Lola rennt, Slavoj Žižek (2001) encourages a
reversal of the narratological common sense, which holds that the style
has necessarily been appropriated to the narrative. This, he argues, is not
the case in Tykwer’s film, since ‘it is not that Lola [rennt]’s formal proper-
ties adequately express the narrative; it is rather that the film’s narrative
itself was invented in order to be able to practise the style’ (81). What Žižek
suggests is neither that Tykwer has chosen the ‘forking-path’ dramaturgy
for commercial reasons nor that the film owes its financial success to a
reterritorialized representation of chaos within classical parameters
(cf. Martin-Jones 2006, 112-113). It is rather that this form has been deemed
the best suitable companion to the film’s fast-motion sequences, musi-
cal editing, frozen still images, slow and fast motion, grainy images, and
above all the iconic kinetic image of the energetic, flamingly red-haired
protagonist of the film.17
For Žižek (2001) the film embodies a new life experience, which amounts
to ‘a perception of life that explodes the form of the linear, centred narrative
and renders life as a multiform flow’ (78). In this line of reasoning, all the
film’s visual modalities and temporal layers have been orchestrated in an
attempt to express the distinct rhythm of the film. The commercial appeal
of the film may, therefore, result not as much from the domestication of
linear temporality, such as Martin-Jones argues, but rather in its ability to
prompt a mode of spectatorship, which reconstitutes the classical cinematic
circuitry of affect-emotion-cognition to complement the contemporary
‘life experience’ – for better or worse. Furthermore, the success of the film
pertains to its incorporation of certain aesthetics of the music video to
highlight the audiovisual nature of the cinematic experience (cf. Chion 1994,
2009). In doing this, Lola rennt appears to give weight to Shaviro’s (2010)
argument that ‘films [...] are machines for generating affect’ (3, emphasis
in original). This view calls attention to the non-linear and virtual aspects
of the film in a manner that does not subordinate them to the narrative.
Shaviro (1993), for instance, argues:
Film’s virtual images do not correspond to anything actually present, but

as images, or as sensations, they affect me in a manner that does not leave
room for any suspension of my response. I have already been touched and
altered by these sensations, even before I have had the chance to become
conscious of them. (45)
I believe this to be a more adequate context for approaching the frequent

references that Lola rennt makes to complexity theory than the one
proposed by Martin-Jones. From this perspective, the film is driven by a
desire to produce a sensation of life as a ‘multiform flow’ in the viewers.
Here reversible and irreversible processes coexist (cf. Prigogine & Stengers
1984, 257), and it is thus tempting to describe complex narratives, such
as Lola rennt, in terms of ‘open-systems of bifurcating emergence [...]
unfolding non-linear, multi-dimensional and durational, complex, chaotic,
if not catastrophic, realities that register affectively’ (Bianco 2004, 402, n. 9,
emphasis added). Yet, the danger in taking such an approach, as has already
been argued, is that it leads to an overemphasis on affect as a dimension
of cinema entirely separated from the cognitive dimensions that interest
the likes of Newman, Cameron, Bordwell, and Martin-Jones. The challenge,
then, is not to isolate the cognitive from the affective aspects of the film,
but to group them together.
To shift our focus onto the interplay between affective and cognitive
spectatorial address in Lola rennt, it is central to accentuate both the
film’s cognitive and affective as well as linear and non-linear dimensions.
Wedel (2009) turns to the philosophy and rhythmanalysis of Henri Lefebvre
(1991, 1992, 2002, 2004) to understand the complex temporal design of the
film. Wedel (2009) is discontented with Bordwell’s reading of Lola rennt,
and feels it needs to be complemented ‘on the grounds of a reconsideration of
cinematic rhythm and sound’ (135). The advantage of Lefebvre’s framework
is its multimodal conceptualization of temporality that moves beyond the
opposition of the ‘clock’ time versus Bergsonian durée. Wedel observes:
Already in the late 1920s, Lefebvre had challenged the then dominant phil-
osophical theorization of time along Bergson’s notion of durée (duration),
and set out to develop what he himself referred to as a ‘theory of moments,’
privileging the importance of the instant. In Lefebvre’s understanding
the single instant as potentially non-calculable, as a critical moment of

virtual openness, as a contingent unit of temporal resistance encapsulates
both the difference between linear and cyclical time, and the contrast
between clock time and ‘lived’ time. (138; cf. Lefebvre 2004)
For these reasons, Wedel f inds Lefebvre’s rhythmanalysis suitable for

examining how the ‘disguised persistence of linearity’ (139) in Lola rennt
is simultaneously connected to ‘those dimensions of space, time, and subjec-
tive experience opened up by rhythms, repetitions, and intervals’ (139).
Lefebvre (2004) conceptualizes temporality in musical terms such as melody
(as the linear sequence of notes in temporal succession), harmony (the
simultaneous sounding of notes), and rhythm (as the placement of notes and
their relative length, which gives rise to issues of change, repetition, identity,
contrast, continuity, etc.) (cf. Wedel 2009, 138-139). The rhythmical approach
to temporality offers a new conceptual palette beyond the linear-non-linear
dichotomy. As such, it allows for an examination of the linear dimension
of Lola rennt as a vital component, which is nevertheless merely a part
of the larger and more complex rhythm of the film. From this point of
view, reducing the film to a linear logic would be comparable to reducing
a musical piece to a series of notes in temporal succession. Rhythmanalysis
offers an alternative to Martin-Jones’ and Bordwell’s arguments that Lola
rennt is constrained within a linear view of time, and it thus resonates
with Shaviro’s concept of ‘post-continuity’ discussed in Chapter II.
Michel Chion (2009) has suggested that in contemporary cinema ‘music
has become the ever more privileged “place” of the film and the reference
point of its editing’ (416). For this reason, we no longer ‘have the same need
for the film to construct a coherent, coordinated, and homogeneous visual
space’ (416). An analysis of the narrative rhythm of the film thus appears
to be a relevant place to understand the role of ‘post-continuity’ in Lola
rennt. In the film, the persistent techno-beat sets the pace, vitality, editing,
and visual style of each of Lola’s three runs. A dynamic that complements
the narrative rhythm is created when the techno-beat is replaced by ‘What
a Difference a Day Makes’ performed by Dinah Washington, when the film
shows Lola being shot and falling to the ground in slow motion accompanied
by Charles Ives’ ‘The Unanswered Question’, or in the complete absence of
any soundtrack in the bedroom scenes. The dynamic role of the soundtrack
becomes particularly perceptible when the recurring techno-beat fades
in over the last 20 seconds of the bedroom scene before the second run to
prepare spectators affectively and cognitively for the coming visual and
narrative explosion.
In her study of the soundscape of Lola rennt, Caryl Flinn (2004) suggests
that the choice of techno music has – at least – two plausible motivations.
First, techno music with its persistent ‘backbeat’ has a direct and bodily
appeal, which is underlined in its function as dance music. The music thus
attaches our affective engagements with the film to the movements of the
title character, and thereby constitutes a very basic and precognitive bond
between spectator and protagonist (that feeds back into and thus influ-
ences our cognitive appreciation of the film). It can be argued, as Bordwell
has done, that this is accompanied by an epistemological bond between
spectators and Lola since both are able to ‘learn’ across parallel universes
(cf. Bordwell’s convention #6). In this manner, the film connects us to Lola
cognitively as well as affectively. The music, however, does not only tie
spectators to the film’s main character, but equally connects them to the
persistent and riveting flow of the film. As Flinn argues, the ‘musical beats
dictat[e] much of [Lola rennt’s] rhythm, pace, editing, and energy’ (202).
The techno music thus sets the overall rhythm of the film.
It would, however, be a mistake to assume that this means a complete
reintroduction of the affective-emotional-cognitive circuitry associ-
ated with classical cinema. The music does not lead up to the inevitable
climax of the story, since ‘unlike traditional tonal music, techno has no
clear beginning, patterns of development, or resolution; unchanging and
energetic, it is repetitive without standing still’ (202). Rather than simply
inducing a moment of progression, the film attempts to enfold or embed
us into its (narrative, audiovisual, stylistic, etc.) rhythm (cf. Chapter V).
This is primarily achieved at the level of affection since the spectator here
is connected to the musical rhythm of the film and to Lola’s relentlessly
forward-moving body. At times, the rhythm of the film, the movement of
the main character, and the affectedness are so enfolded into one another
as to appear inseparable. The film, for instance, achieves this in a scene
where Lola’s heartbeat synchronizes with the beat of the techno music to
create one united pumping rhythm.
When the audiovisual aspects of the film are considered, the dramaturgi-
cal ‘forking-path’ structure is given a dimension that has been obscured by
Bordwell’s and Martin-Jones’ obsession with linearity and continuity. The
narrative structure, the film’s multimodal visual style, and the dynamics
of the soundtrack all simultaneously complement Lola’s restless forward
movement and the film’s repetitive loops in order to accentuate recogniz-
able patterns of change and difference. It is in this manner that Lola rennt
contains, yet cannot be reduced to either the linear logic of continuity or the
non-linear ruptures of discontinuity. Wedel (2009) turns to Flinn to rehearse
a similar string of arguments when he asserts that the techno music acts
‘[a]s a musical idiom and temporal marker which is both a linear forward
driving force that vectorizes movement and unleashes a pulsating energy
and, at the same time, constitutes a highly “repetitively structured form,
organized around beats per minute”’ (137; cf. Flinn 2004, 202).
In a similar fashion, the split screen that for Martin-Jones encapsulates
Lola rennt’s linear rationale is perhaps better understood as comprising
several modes of temporality – including linearity and non-linearity. Once
the split screen is seen as part of the overall narrative rhythm of the film,
the linearity of the scene incorporates the spectator, who takes part and
actively forms this rhythm. This does not happen in the form of a struggle
between reterritorializing and deterritorializing powers, but is perhaps
better perceived as ‘unities in opposition’ formed between different spatio-
temporal logics (Lefebvre 2004; Wedel 2009). Whereas the relation between
these in the first strives towards dominance, the latter brings these elements
into synthesis.
The split screen should be seen in unison with the narrative structure,
the soundtrack, the bedroom scenes, the flashforwards, etc. All are operat-
ing towards ‘synthesizing linear and cyclical temporalities into the one
common logic of rhythm’ (Wedel 2009, 137). The common logic of rhythm
should not mistakenly be understood as a ‘true’ logic, since it itself implies
a multiplicity of many logics. Instead, it is the logic of persistent feedback
loops in the system that connects non-linear affects with cognitive strives
for meaning. We are here approximating the logic of the embodied fabula.
This differs from the binary logic as exemplified by Martin-Jones, who
perceives the clock in the split-screen image to be a symbol of the disembod-
ied, objective, and temporally detached perspective that the film proffers
(cf. Martin-Jones 2006, 112). Here the film is ultimately reduced to its linear
dimension. However, once the spectator is actually perceived to embody
the various temporalities of the films – in the split screen, the spectator
embodies the dissonance in temporal perspectives – the temporal logic
conveyed in the scene appears multilayered and complex.
This is not a recuperation of linear temporality, since according to the
logic of rhythm ‘[t]ime and space, the cyclical and the linear, exert a recipro-
cal action: they measure themselves against one another; each one makes
itself and is made a measuring-measure; everything is cyclical repetition
through linear repetitions’ (Lefebvre 2004, 8). The film ties us affectively
and cognitively to the characters, but also to the narrative and audiovisual
rhythm of the film. This ‘synchresis’ (Chion 1994) (an acronym formed in
the merging of the words synchronism and synthesis to describe the forging
between an audio impression and a visual impression, when those occur at

exactly the same time) ties the visual image to the soundscape of the film. In
Lola rennt, the ‘synchresis’ underlines the embodied cognitive-affective
mode of spectatorship that characterizes the film. In this fashion, the film
does what Marshall McLuhan (1994) has famously argued electronic tech-
nology does more generally, it ‘dethrones the visual sense and restores us
to the dominion of synesthesia, and the close interinvolvement of the other
senses’ (111). Wedel’s application of Lefebvre’s rhythmanalysis provides an
example of how an embodied methodological approach can go beyond the
linear-non-linear dichotomy, and in doing so enhance our understanding
of the temporal logic of Lola rennt.
VII. Memento and the Embodied Fabula
In post-theory, the ‘beautiful’ is not necessarily consilient with goodness,

the romantic, or transcendent notions, but to a feeling of duration, movement
and continual process: what Deleuze refers to as ‘haecceity’ or ‘intensity’.
[...] ‘Beauty’ thus pertains to a process that takes empirical precedence over form.
[...] The determination of beauty becomes temporal, not reflective:
an open-ended process, a feeling of flowing, rhythm, or ‘becoming’.
Indeed, a refreshing concern with sensation, rather than desire or pleasure,
requires us to think about sensation as a rhythmical experience,
not one of static shock of excitations on the nervous system.
‒ Kennedy 2000, 30-31
What will you find once you have completed the intricate
and exhausting path and found the centre of the maze?
Yourself – in the middle of a maze.1
Introduction
Christopher Nolan’s Memento (2000) – based on Jonathan Nolan’s short

story Memento Mori – is perhaps the most debated complex narrative to
date. In addition to countless formal dissections, the film has been the
subject of innumerable critical examinations. It has been perceived as
exemplary of various different philosophical and theoretical positions, and
the philosopher Andy Clark (2010) has even suggested that the film’s main
character can be perceived as the embodiment of his extended mind thesis.2
On account of this, Robert Sinnerbrink (2011) has humorously asserted
that Memento should now be considered the philosophers’ favourite film,
thereby suggesting that it has trumped such beloved examples as Rashômon
(Kurosawa 1950) or more recently The Matrix (Andy & Larry Wachowski
1999) (48). The extended attention devoted to the film demonstrates that
‘Memento is a very good example of what a complex narration can be’
(Ghislotti 2009, 87). Although Memento affirms the inferential nature of
cinematic spectatorship, it simultaneously renders difficult the cognitive
processes that organize narratives into chronological series of events.
This chapter argues that Memento thereby questions the narratological
coupling of realist and constructivist epistemologies and defamiliarizes
linear cinematic perception . In so doing, the film enfolds or embeds the
spectator into the narrative, and thus spectator and film become deeply
intertwined. To better capture this aspect of the cinematic experience
induced by Memento an embodied reconfiguration of spectatorship is
required.
In contrast to our everyday experience of time as forward progression,
Memento opens with a sequence in which the ‘arrow of time’ has been
reversed (cf. Chapter VI). Here a Polaroid photo appears on the screen,
which is only later to be identified with the film’s main character Leonard
Shelby (Guy Pearce), who suffers from anterograde amnesia – a condition
that makes him unable to create new long-term memories. Curiously,
rather than gradually developing, the Polaroid dissolves each time it
is shaken. Suddenly, the Polaroid ‘jumps’ back into the camera, blood
‘runs’ up the wall, a gun ‘springs’ into Leonard’s hand, and a cartridge
‘flies’ back into the gun, before a man – later to be identified as Teddy
(Joe Pantoliano) – ‘comes alive’ with a scream. The scene contains a
media-conscious play on the anxieties and desires associated with the
cinematic and photographic ability to ‘preserve the lifelike movements
of loved ones after their death’ (Doane 2002, 3).3 Yet, Memento goes even
further in this respect, since it has integrated an awareness of how the
cinematic medium structures human perception and cognition into its
dramaturgical arrangement.
The innovative and experimental narrative structure of Memento has
been instrumental for the extensive attention devoted to the film by fans,
critics, and academics alike. Following the suggestion of film critic Andy
Klein (2001), the narrative structure can be illustrated according to an
intricate, yet systematic, scheme where the coloured reverse chronological
scenes (A, B, C, etc.) are separated from the black-and-white, chronologi-
cal scenes (1, 2, 3, etc.). Klein’s suggestion provides us with the following
visualization of the narrative structure:
Credits, 1, V, 2, U, 3, T, 4, S, 5, R, 6, Q, 7, P, 8, O, 9, N, 10, M, 11, L, 12, K, 13 J,

14, I, 15, H, 16, G, 17, F, 18, E, 19, D, 20, C, 21, B, 22/A.
Each reverse chronological sequence is interrupted by a black-and-white

scene from which a new reverse chronological sequence follows, etc. Scene
22/A is pivotal in relation to this, since the two narrative threads merge
in this scene. The black-and-white scene 22, as Klein observes, ‘almost
imperceptibly slips into color and, in an almost vertiginous intellectual
loop, becomes (in real-world order) scene A, the first of the color scenes’
(par. 30).
Memento and the Embodied Fabul a 177
With the merging of the two narrative threads, the film allows for the
construction of a causal-chronological continuum, which can be translated
into the following structure:
1, 2, 3, 4 ,5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22/A, B, C, D, E, F,
G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, Credits (in reverse).
In applying this narrative logic, Klein continues by providing an extensive

and meticulous account of the film’s events once these have been organized
causal-linearly.
Stefano Ghislotti (2009) has examined the difficulties that this narrative
structure causes to memory and cognition. According to him, ‘the difficulties
we experience [in comprehending the narrative of Memento] are deeply
rooted in the skillful way we normally watch ordinary movies’ (88). Ghislotti
is intrigued by the viewers’ mental capacities and their ability to comprehend
the narrative despite its demanding structure: ‘[H]ow one can cope with a film
so intricate and difficult to understand is a very stimulating question’ (87).
Interestingly, Memento first becomes truly exciting once our cognitive
processes can no longer operate in the unattended and automatic manner that
cognitivists refer to as ‘successful’ (cf. Bordwell 1989, 12). Because it makes us
aware of the standard quotidian processes that constitute what is experienced
as ‘normal’ or ‘natural’ cinematic perception, Memento is ideal for exploring
the distinction between the cognitive-formalist and the embodied fabula.
Ghislotti’s study makes it evident that Klein’s analytic dissection of the
film’s fabula does not equal the cinematic experience of the spectator, who
during the screening is trying to work out the narrative structure mentally.
This owes to the reverse chronological dramaturgy of Memento, which
allows spectators to experience their active contribution to the construction
of the narrative. The film thus questions the fabula as a representation of
the narrative via an initial affirmation of core cognitivist assumptions
about cinematic spectatorship. In this fashion, Memento allows us to
grasp how the ‘sensory data of the film at hand furnish the materials out
of which inferential processes of perception and cognition build meanings’
(Bordwell 1991, 3). Hence, Memento makes visible the inferential nature of
cognition and perception, whereby it becomes evident that the ‘perceiver
is not a passive receiver of data but an active mobilizer of structures and
processes’ (3). Touching upon this, Sinnerbrink (2011) has contended that
Memento ‘chimes with cognitivist theories of narrative that emphasize the
roles of rational inference-making, the testing and adjusting of beliefs, and
the cognitive matching of affective tone with perceptual awareness’ (48).
The reverse chronological dramaturgy is thus not only connected to the

film’s discursive exploration of memory, it also proves a demanding task
for the spectator’s own memory capacities. Both reflect upon Leonard’s
inability to build new memories, despite his ability to remember the past
before the accident normally.4 Consequently, much of the narrative centres
on Leonard’s ‘memory system’ – consisting of mementos in the form of
annotated photographs, notes, and tattoos for the most important ‘facts’
– that aids him in his strive for revenge. Leonard harbours an unrelenting
belief in the capability of his methodological system to acquire facts that
are superior to those that a normal working memory would be able to
conjure:
Facts, not memories: that’s how you investigate. I know, it’s what I used
to do. Memory can change the shape of a room or the color of a car. It’s
an interpretation, not a record. Memories can be changed or distorted
and they’re irrelevant if you have the facts.
Since the film’s difficult narrative structure challenges the spectator’s

own memory capacities, it can be argued that it substantiates Leonard’s
disbelief in the accuracy of memory. However, as viewers will progressively
come to recognize it is not just memory that is deemed unreliable, but also
the ‘facts’ of Leonard’s ‘memory system’. Viewers become aware of how
easily Leonard is manipulated in a scene where he is tricked into helping
Natalie (Carrie-Anne Moss) – the femme fatale whose boyfriend was killed
by Leonard, who, nevertheless, has forgotten about the incident. Natalie
hides all pencils in her home, then she provokes Leonard into hitting her (in
passing, it should be noted how revealing it is that violence is his instinctive
reaction to being provoked). Natalie leaves only to return moments after, yet
long enough for Leonard, who has been struggling to find a pen, no longer
to remember the episode. Seeing the marks on her face, he agrees to help
Natalie pacify the man whom he now firmly believes to be responsible for
harming her.
From such scenes, it becomes clear that all the characters surround-
ing Leonard are in fact interested in taking advantage of his condition,
which makes an accurate version of the story hard to conjure. Yet, for
Bordwell (2006) the difficulties of the film are merely the result of the
order in which information is presented:
If story events had been presented in chronological order, Teddy would at

the start have explained fully how he duped Leonard into killing Jimmy
and others, providing us with all necessary exposition. In reverse order,

however, Teddy’s explanation serves as a climactic revelation, resolving
many uncertainties. (80)
As argued in Chapter III, the linear interpretative method implied by the

fabula rests on an elimination of ambiguities. Consequently, Bordwell is
willing to accept Teddy’s account of the events presented at the end of the
film.5 In short, Teddy denies that Leonard’s wife was killed in the accident,
and claims that she died from an overdose of insulin. According to him,
Leonard did in fact lose his memory in the accident, yet his surviving wife
could not live with her ‘new’ husband and in a final test to ‘shake him out
of his condition’, she made him give her repeated injections of insulin. This
explanation suggests that Leonard’s story about Sammy Jankis (Stephen
Tobolowsky), a former client who suffered from the same condition, is in
fact a revised version of Leonard’s own story. Leonard explains that the story
of Sammy ‘helps [him] understand [his] own story,’ but believes that his
‘memory system’ will prevent him from suffering the same fate as Sammy:
Sammy Jankis wrote himself endless notes. But he’d get mixed up. I’ve
got a more graceful solution to the memory problem. I’m disciplined and
organized. I use habit and routine to make my life possible. Sammy had
no drive. No reason to make it work.
When Teddy suggests that Sammy is merely Leonard’s psychological projec-

tion of himself, this can be interpreted as providing the ‘solution’ to the
narrative. Since this interpretation is supported by a subliminal image,
where Leonard takes the place of Sammy in a mental hospital, it has often
been perceived as the unambiguous solution to the narrative.6
However, nothing ensures the absolute reliability of Teddy’s version. As
the film ends, spectators have witnessed Leonard killing twice (Natalie’s
boyfriend Jimmy and Teddy), and we are left with the impression that his
search for revenge/justice has become instrumentalized; mostly for drug
dealing. In what we come to realize is just a Sisyphean loop, the question
we thought was central – who killed Leonard’s wife – no longer bears the
promise of epistemological clarity.7 The original poster for the film plays
with the mise en abyme, or infinite regression, of its narration, hinting
that our attempts to linearize and rationalize the narration will result in a
journey down the rabbit hole.8 However, this journey is also the attraction
of the ‘game’ Memento plays with its spectators (cf. Elsaesser 2009), and
should be seen in the context of the particular kind of self-reflexitvity to
be found in complex narratives. As Poulaki (2014b) observes, this differs

from its Brechtian and modern(ist) kind. Thus, ‘instead of alienating the
viewers, or making them reflect upon the film (by means of a subject-object
relationship and positivist epistemology), self-reflexive techniques engage
them, as already broached, in an increasingly complex communication’ (41).
Realism and Constructivism: A Narratological Deadlock
To better appreciate how Memento engages us with the narratological para-

dox that lies in the double role of the spectator as both creator and discoverer
of the fabula, it is useful to differentiate between realist and constructivist
epistemological positions. Bebe Speed (1991) has provided a concise descrip-
tion of the realist position, as that according to which ‘reality exists, can
be discovered by people in an objective way and thus strongly determines
what we know’ (396). This is countered by various forms of constructivism.
According to Ernst von Glasersfeld (1991), constructivism in its ‘radical’ form
was conceived as an attempt to circumvent the paradox of traditional

epistemology that springs from a perennial assumption that is inextri-
cably knitted into Western philosophy: the assumption that knowledge
may be called ‘true’ only if it can be considered a more or less accurate
representation of a world that exists ‘in itself,’ prior to and independent
of the knower’s experience of it. (13, emphasis in original)
In Memento the realist position is expressed with reference to Leonard’s

‘memory system’ and the manner in which he takes upon him the role of
the classical detective, who must unravel the true nature of events in the
name of an altruistic justice (altruistic insofar as Leonard himself will
be unable to remember that justice has been served). In relation to the
cognitive-formalist fabula, the realist aspect is expressed in the search for
a ‘true’ account of the events, whereas the constructivism is emphasized in
the inferential nature of narrative comprehension (perception, cognition,
and memory are all inferential).
Memento brings attention to the latter insofar as its ‘unusual composi-
tion hinders some basic functions of memory’ (Ghislotti 2009, 88). From this
perspective, the difficulties pertaining to the construction of a coherent
fabula involve: the formation of a mental representation of the state of
affairs; the processing of the audiovisual stream of data; the sorting of im-
portant aspects into mental schemes; the storing of important information
and order of events; the constant revision and reintegration of new fabula
events; and the capacity of sensing time (89). The film thus overloads the
mechanical operations of our memory systems, insofar as ‘[i]t is impossible
to keep more than a few elements in our short-term memory’ (97). This
renders difficult the realist search of a coherent, meaningful, and unified
‘truth’. Thus, the narrative structure of Memento both activates our realist
desire for knowing the ‘truth’ while spelling out the imperfect nature of
knowledge based on the unreliable nature of memory.
Consequently, ‘[w]hile the film continues to present new elements to be
processed, we cannot keep the complete ordered fabula structure in mind’ (97).
This introduces a gap, since the solidity of the fabula available to the analyst
after a critical retrospective dissection of the film (cf. Klein 2001) can no longer
meaningfully be said to correspond with the story that spectators operate
with during the screening of the film. This renders evident that the fabula
can be seen as an analytically detached representation of the narrative (the
analytical fabula), which is generally more systematic and linear than the
spectator’s mental experience of the narrative (the embodied fabula). My point
is that Memento activates our desire for constructing an analytical fabula,
yet in doing so demonstrates that narrative experience extends far beyond
the ‘problem-solution’ or ‘question-answer’ models of cognitive formalism.
Volker Ferenz (2009) has pointed out that Memento stages a clash
between the realist and the constructivist positions on epistemology.
Besides carrying the traces of a realist epistemology, Ferenz (2009) has
contended that Memento and similar f ilms simultaneously ref lect
upon the ‘modern discourse of epistemological scepticism because they
playfully stage the epistemic limitations experienced by their respective
character-narrators’ (260). Since Memento does not deny the existence of
an external reality, but restricts the possibility of knowledge to the sphere
of that which ‘the autopoietic human being makes of it’ (274), Ferenz ar-
gues that Memento can be perceived as a ‘textbook case of psychological
constructivism’ (274). Although the constructivist theory appears to be a
suitable framework for understanding Leonard, Ferenz argues that the film
ultimately demonstrates how this epistemological position can ‘easily lead
to the erosion of personal responsibility’ (277).
On this interpretation, the film actually contends that ‘a philosophical re-
alism, where we discover features of objective reality rather than construct
them, is preferable in the real world’ (277). This argument should be seen in
the light of how Leonard actively constructs his memories according to a
classical and autopoietic narrative, where he plays the role of the innocent
victim, who must fight an unjust world against all odds. In the pivotal
scene at the end of the film (22/A) Leonard manipulates himself into killing
Teddy (we see Leonard shoot Teddy in the opening scene, but it is only at
the end of the film that this act can be put into a more rewarding context).
As Leonard scribbles down Teddy’s number plate in his notebook with the
addition ‘fact number 6’, his voiceover is heard: ‘Do I lie to myself to be
happy? In your case, Teddy... Yes, I will’. For these reasons, Ferenz argues
that Memento expresses the modern anxieties connected to a radical
constructivist epistemology and the loss of stable truths.
However, it could equally be argued that the moral deprivation of
Leonard neither originates from his realist nor constructivist tenets but
their unfortunate combination. Leonard acts and reacts as an ‘affective
machine’, because the mnemonic devices around which he structures his
environment are unable to connect with deeper layers of memory (cf. Bianco
2004). This means that Leonard’s tattoos, Polaroids, notes, etc., prompt
immediate actions. His facts are bereft of their complex context, and his
methodological and realist system thus relies on a problematic deductive
reasoning. As Bianco (2004) has observed, ‘the injunction inscribed on the
Polaroid of Teddy, “Don’t believe his lies,” becomes a generalized, “don’t
believe Teddy,” since there can be no distinction made between perception
and memory, Teddy and his lies’ (384).
Due to Leonard’s misuse of the constructivist epistemology to justify his
immoral actions, Ferenz concludes that the film demonstrates the moral
implications of the radical constructivist position:
For [Leonard], because (a) in such an objective and mind-independent

world somebody else killed his wife (a subjective interpretation), his goal
of (b) revenging her death is just (it is a self-righteous concept). Therefore,
(c) it is okay for him to hunt down that mysterious John G. (a subjectively
understandable form of justice). However, the assumption that some
junkie killed his wife has already been shown to be untrue, and thus this
whole line of reasoning collapses. On the level of the film’s story, Leonard’s
line of reasoning is but an excuse of sorts to remain a serial killer. (276)
Paradoxically, this interpretation is itself enunciated from a realist narrato-

logical perspective, since it levels out the carefully constructed ambiguity
of the film – the question of whether Teddy can be trusted – to provide
an authoritative fabula. Moreover, this interpretation is conducted from
a rationalist position that posits to gain knowledge about the narrative
world through an analytically detached gaze. It thus fails to acknowledge
how Memento implements the spectator’s and analyst’s desire for ‘solving’
the narrative. The film’s reverse chronological structure makes us aware of

our own ongoing linear, cognitive reconstruction of the story. In accentuat-
ing the spectator’s active construction of the narrative (constructivism), the
film thus simultaneously questions the idea of settling on an intersubjective
narrative truth (realism).
Ferenz’s approach is representative of the difficulties that cognitive
formalism confronts, when it fails to properly take into account the dynamic
relationship between narrative and spectator/analyst. In the case of Me-
mento, spectators are aligned to the protagonist in a shared desire to learn
about the ‘true’ nature of the events. This alignment is further established
by an epistemological bond stemming from the reverse chronological
dramaturgy. Consider how Nolan explains his motive for telling the story
in reverse chronology:
One day I drank too much coffee and said to myself, ‘Well, if you tell
the story backwards, then the audience is put in the same position as
Leonard. He doesn’t know what just happened, but neither do we.’ (Nolan
in Molloy 2010, 45)
In this fashion, the narrative structure not only creates an epistemological

bond between the spectators and Leonard, it also accentuates the similarity
between Leonard’s situation and narrative interpretative strategies that aim
to conjure up a coherent fabula. Unlike Leonard, the spectators are capable
of retroactively connecting the disjunctive elements into a fairly coherent
string of events. In doing so, they realize the unreliability of Leonard’s realist
‘memory system’ and his ‘autopoietic narrative’. This causes a destabilization
of the epistemological bond, which in turn renders visible the similarities
between Leonard’s endeavours and a cognitive-analytical comprehension
of the narrative. As Simons (2008) explains, ‘the narratologist does what
[...] the main character of Memento do[es]: [both] reason backwards from
the end to identify the events that necessarily had to occur to make the
“terminal event”, the ending possible’ (119; cf. Barthes 1977, 99).
Consequently, it can be argued that the epistemological scepticism of
Memento goes beyond Leonard’s manipulation of his mementos (and
hence his memory construction) to also include the narratological mode
of reasoning backward. Leonard’s moral deprivation is thus not the result
of an epistemological constructivism per se, but arises when such is im-
plemented into a classical, realist narrative of an innocent man seeking
justice. Without his realist ‘memory system’ Leonard would not be able
to design the paths that repeatedly send him on a Sisyphean hunt for a
murderer who may or may not exist – but whose existence in any case has
ceased to be important. In this fashion, Memento explicates the mise en
abyme of cinematic narrative comprehension as a paradoxical oscillation
between constructivism (we actively construct the narrative) and realism
(to establish an approximately ‘true’ fabula).
In doing so, Memento playfully stages a modern predilection for reconcil-
ing constructivist insights within a realist epistemology. Nowhere is this as
explicitly pronounced as in Bordwell’s theory of narrative comprehension.
According to Bordwell (1996a), radical or strong constructivist positions
risk the lure of absolute relativism:
If beliefs are relative to culture, then belief in relativism must be relative

to our culture; but then that doctrine cannot claim true insights into
the beliefs, relative or not, of other cultures. As far as I can tell, no film
theorists have addressed the self-contradictions haunting the radical
constructivist premise. A radical constructivism is also empirically
limiting. Universal or cross-cultural regularities can play an important
role in our explanation of human action. (13)
Consequently, Bordwell’s (1991) version of constructivism ‘assumes that it

is possible to arrive at inferences which are at least approximately true;
it is thus compatible with a critical realist epistemology’ (277, n. 9). Nick
Redfern (2005) has not only convincingly argued that Bordwell’s objections
to radical constructivism arise from a (common) misunderstanding of this
position, he has also presented a precise criticism of Bordwell’s attempt to
reconcile the constructivist and realist positions.9
Redfern explains that radical constructivism denies neither the existence
of an ontological reality nor the existence of a mind-independent reality.
However, insofar as radical constructivism contends that an ontological
reality is not reachable given that human knowledge is constructed, this
position should be understood as ‘second-order knowledge’, i.e. a form of
knowledge about knowledge itself. For the radical constructivist, cognition
thus serves the organization of the experiential world, and not the discovery
of an ontological reality (Glasersfeld 1988, 83). According to Glasersfeld (1991),
this distinguishes the ‘radical’ from the ‘trivial’ constructivist position:
Those who merely speak of the construction of knowledge, but do not

explicitly give up the notion that our conceptual constructions can or
should in some way represent an independent, ‘objective’ reality, are still
caught up in the traditional theory of knowledge. (16)
In being knowledge about knowledge, radical constructivism is not

concerned with the ‘true’ nature of things. Redfern (2005) explains that
‘the key to evaluating competing knowledge claims [...] is not to seek to
compare them to a mind-independent reality that cannot be known, but
to assess their cognitive viability or functional fitness’ (sec. 3, emphases in
original).
In relation to Bordwell’s theory of cognitive spectatorship, Redfern argues
that it falls short on two points. First, Bordwell assumes that spectators
are able to conjure inferences that are ‘at least approximately true,’ but he
fails to account for the circumstances that would ensure such proximity.
Redfern (2005) reasons as follows:
Confronted with such a model the following question remains: How do

we know that our inferences are ‘at least approximately true?’ An answer
to this question has eluded Western philosophy since Socrates, and,
without a solution to the ontological problem, constructivism remains
trapped within the epistemological trap of realism. (sec. 3, emphasis
in original)
Secondly, Bordwell lacks a convincing theory of communication, since he

perceives the fabula to be constructed from the ‘mind-independent’ cues
that are perceived to be inherent entities of the film. Redfern rejects the
proposition that the cues can act as the ‘facts’ from which the fabula can
be constructed in more or less proximity to the film’s ‘true’ story. Instead,
‘“facts” in the cinema are [...] not the pre-existing cues in a film described
by Bordwell, but are the result of an active process of observing on the part
of the spectator. These facts are then put into a context by a spectator (i.e.
they are interpreted)’ (Redfern 2004, 45-46, emphasis in original).
The result of the combination of realism and constructivism is that
cognitive approaches tend to assume that the cinematic experience is
restricted by the formal properties of a given film. Ben Shaul (2012), for
instance, criticizes Memento for being a f ilm that forces viewers to
construct the narrative along a closed-minded path (cf. Chapter III). He
reasons as follows:
[T]he disjointed dynamic of the backtracking narrative, figuring and

formally replicating the mentality of its mentally unstable protagonist,
rather ends up encouraging in viewers an intense attention, heightening
their need for closure, along with an intense fear of invalidity due to the
random and dissociative narrative. While this process is intriguing, in
that it evidences the cognitive ability of viewers to manage to nevertheless

follow and figure out the narrative, it nevertheless ends up encouraging
loopy thinking both in the required and repetitive back-and-forth puzzle-
solving activity, and in the latter’s short-circuiting at the end by a loop
that returns the viewer to the film’s opening shot. (106)
Here ‘loopy thinking’ describes the act when individuals, fearing a lack of
closure, ‘fix on a mental loop whereby options are reduced to a narrow and
recurring set of irresolvable alternatives, a mental maze from which they
cannot escape’ (9).
In proposing a ‘fear of invalidity’, Ben Shaul assumes a spectator for whom
the main cinematic and cognitive pleasure resides in the achievement of
closure, i.e. of learning the ‘truth’ about the narrative. This assumption reso-
nates with Bordwell’s theory of narrative comprehension, which assumes
that a great deal of cinematic pleasure lies in the cognitive reorganization of
the narrative continuum into a causal-linear and coherent series of events.
As has been argued elsewhere, this account has revealed its explanatory
limitations, since it cannot account for the narrative pleasures invoked by
films that disrupt the urge for narrative certainty or otherwise upset our
expectations (Sinnerbrink 2011, 52). Consequently, cognitive theories have
displayed a general lack of sensitivity for how the spectator experiences
the (successful or failed) attempts to construct a fabula, and the possible
reflections to be drawn from this experience are lost as a result.
In fact, it can be argued that Ben Shaul’s criticism of Memento reflects
back upon his own restricted conception of narrative comprehension. Since
the linear interpretative model that Ben Shaul presupposes is uneasy with
narrative ambiguity, the spectator is bound to a ‘closed-minded’ diminishing
of ambiguity by prematurely settling on one of the possible interpretations
as the correct one. Otherwise viewers are caught in a catatonic state of
‘loopy’ thinking in which no progress is being made. However, perceiving
the narrative maze as a limitation to thought – rather than a productive and
creative force – appears to be an erroneous belief, which, I believe, stems from
the marriage of constructivism and realism found in cognitive film science.
In order to avoid the problematic union of realism and constructivism,
Redfern proposes an alternative to Bordwell’s unidirectional communica-
tion model. Redfern’s alternative is a self-referential model that relies on
constant feedback loops (cf. Fig. 3). While Redfern’s model can be criticized
for not allowing emotion or affect into the model of knowledge construction
(cf. Chapter V), its advantage over Bordwell’s model is that it does not presup-
pose the existence of inherent narrative meanings or cues. This model
Figure 3.
Stages in the Construction of Knowledge (Redfern 2004, 46)
Theorising, abstracting conceptual

structures to construct general
explanations and general
taxonomies.
Interpreting, putting facts in to an

Testing, applying a theory to new
experiential context in order to
experiential contexts.
build up conceptual structures.
Observing, constructing empiric

all facts on the basis of regularities
in the subject’s experience.
accounts for how cinematic images can be meaningful without succumbing

to the epistemological fallacy of understanding these as inherent features
of the artwork itself. Hence, meanings do not reside in the cinematic images
but arise from the encounter between the perceiver and the perceived world.
Redfern’s model allows for a more open examination of cinematic nar-
ratives, since it does not rely on pre-existing meanings that the spectator
must uncover – an idea that contributes to the limited freedom of the
cognitive spectator (cf. Chapter III). According to Redfern’s model, cinema
is constituted in the encounter between the cinematic material and the
cognitive-perceptual system of the spectator, from which structures and
interpretation are conjured. Due to the feedback loops of this model, the
viability of these is then continuously tested and evaluated against the
world (or narrative universe) it is supposed to reflect upon. As such, this
model avoids the problem of representation, since it ‘uses the world as its
own model’ (cf. Brooks 1999, 79; Chapter II).
From this perspective, cinema does not communicate meanings to the
spectator, but structures pathways to be explored in a myriad of manners.
In relation to cinematic complexity, this means that the formal properties
of a film do not determine the meanings that can be constructed from
it (Redfern 2004, 47). Although Redfern’s model does not account for the
affective and emotional role in the construction of meaning, it is useful for

understanding how the embodied fabula can operate with an intersubjec-
tive comprehension of narratives without succumbing to a realist epistemol-
ogy. While the embodied fabula denies the realist and representational
assumptions of Bordwell’s cognitive model, it acknowledges the inferential
and constructivist nature of spectatorship.
A desire for establishing the ‘true’ account of the narrative events can also
be found in Deleuzian film criticism. Here this endeavour has, however, been
reversed. The task thus becomes to deem the narrative puzzle unsolvable.
In her article on Memento, Melissa Clarke (2002) accentuates how the film
arouses a desire in spectators and analysts to figure out the ‘truth’ about
‘the circumstances surrounding Leonard’ (168). For Clarke, it is central that
the film does not render it possible to arrive at an unambiguous solution.
Consequently, her main interest becomes determining the indeterminable
nature of the film, and in her efforts to do so she spells out a paradox that
haunts the Deleuzian time-image.
Clarke explains that time is revealed anywhere there is ‘an indiscernibil-
ity between the actual and the virtual, where there is an ambiguity between
the real and imaginary on the screen’ (172). However, this indiscernibility
cannot ‘be a simple confusion to which there will soon be a resolution in
the next scene’ (172). Consequently, determining this indiscernibility read-
ily becomes a classical narratological question. In this Deleuzian version,
however, the narratological task becomes the establishment of the fabula
as predominantly incoherent rather than as an unambiguous narrative
solution. Hence, the film-philosophical discernment of time-images comes
to rest upon a method of inquiry that is Bordwellian in essence, since it
proposes the most fundamental question of the viewer to be whether a
coherent fabula can be constructed.
Clarke’s (2002) treatment of Memento is thus representative of a tendency
to pursue classical narratological questions within a Deleuzian framework
with the aim of categorizing films along the opposition of movement-image
and time-image. Consider, for instance, the following passage:
Was the story about Sammy ‘true’? Or was the story rather about Leonard,
and was it Leonard’s wife rather than Sammy’s wife who was diabetic and
killed herself to try to snap Leonard out of it? Was Leonard’s wife even
really killed in the accident? These questions are raised by Ted, but can Ted
be trusted to tell the truth? In any event, Ted himself does not necessarily
know the truth and claims to be only speculating. And so on. The ‘truth’
of the past is undecidable for these and other issues; for example, we never
do find out whether Leonard did or did not kill ‘John G.’ At this level, the
viewer becomes aware of the indiscernibility between the truth or falsity of
layers of past. And again, the structure of time is directly revealed. If instead
we were confident in a single determinate truth of the past, we would be
falling into the more traditional understanding of time as linear progress,
and of the present as the only reality, and the past as no longer existent. (177)
Here Clarke perceives the opposition of movement-image and time-image

in terms of a reductive formula according to which the film either allows for
the construction of a coherent fabula and consequently confines to a linear
temporality (movement-image) or it does not allow such to be conjured and
thus opens up for a non-linear, Bergsonian time (time-image).
Once time-images are perceived as arising from the impossibility of deter-
mining a ‘single determinate truth of the past’, these can only be discerned
in retrospect. Whether the indiscernibility was permanent can, therefore,
first be determined after the fact, i.e. once the narrative has ended. The
paradox is that time-images in this manner can only be discerned from a
perspective that is temporally detached from the actual viewing experience.
Since Clarke addresses the problem of movement-image versus time-image
from the linear onto-epistemology of classical narratology, she, too, makes it
the central concern to ‘search, behind those appearances, the hidden order
that is the authentic reality of the universe’ (Morin 2007, 6; cf. Chapter II).
Although she presumes ‘reality’ not as linear but Bergsonian and non-linear
(and this obviously entails a series of philosophical self-contradictions).
Rendering visible how the cognitive-formalist fabula continues to inform
Deleuzian approaches to cinematic narration, and replacing this concept
with the embodied fabula may help overcome a tendency, as observed by
Sinnerbrink (2008), within post-Deleuzian film-philosophy to bifurcate
into two competing directions of interest. On the one hand, there are those
who foreground ‘nonsubjective affect or sensation (as Powell and Coleman
do)’ (95) or phenomenological studies that pay extensive attention to the
visceral aspects of the cinema experience (cf. Sobchack 1992, 2004; Marks
2000, 2002; Barker 2009). The other group of Deleuze-inspired scholars
mentioned by Sinnerbrink (2008) emphasizes ‘thought or the cinema as
brain (as Lambert and Flaxman do)’ (98). Sinnerbrink observes that one of
the difficulties pertaining to contemporary Deleuzian film-philosophy is
that these two camps – the ‘affective’ and the ‘brain’ – ‘rarely meet on the
same conceptual plane’ (98). He bemoans this situation, since ‘this is surely
one of the most interesting questions facing Deleuzian film-philosophers:
what is the cinematic relationship between time, affect, and the brain?’ (98).
Similarly, the cinematic experience conveyed by Memento demonstrates

the importance for film-philosophers to embrace both the cognitive and
affective dimensions of cinema. The concept of the embodied fabula allows
us to unite the affective and cognitive dimensions of the cinematic experi-
ence on the same conceptual plane, and to conceptualize the multilayered
temporality that arises from this encounter.
In relation to this, the clear separation of the actual from the virtual, the
linear from the non-linear, which at times follows from the Deleuzian dif-
ferentiation of movement-images from time-images, becomes problematic.
Whereas Deleuze accounts for the distinct nature of the movement-image
and the time-image from a (film)philosophical perspective, Rancière (2006)
reminds us that ‘in practice this opposition between two logics is almost
indiscernible’ (122). Perhaps in order to avoid a clear distinction between
these image-regimes, Deleuze (2005b) denies that the actual and virtual
aspects of the image should be understood in terms of a mutual exclusion.
He explains that there is ‘no virtual which does not become actual in rela-
tion to the actual, the latter becoming virtual through the same relation:
it is a place and its obverse which are totally reversible’ (67). In the words
of Rancière (2006), ‘[t]he gesture that frees the potentialities remains, as
always, the gesture that chains them up again’ (118).
To come to grips with the actual experience of time conveyed in the
cinematic experience, it is thus incisive not to equate the temporality associ-
ated with the analytical fabula with that of the embodied fabula; the latter
arising from the immediate encounter of film and spectator. In relation to
this, Anne Friedberg (1993) has acutely observed:
Deleuze’s descriptions border on a theoritization of where – in time – the

spectator is, but his discussion of the ‘time-image’ ultimately relies on a
conception of diegetic film time, not the alternations in the spectator’s rela-
tionship to temporality produced by film-going. (129, emphases in original)
The tendency to conflate the temporality conveyed in the film with that
pertaining to the spectator’s experience is thus not reserved to cognitive
and classical narratological studies (cf. Chapter VI). The embodied fabula
circumvents this conflation by providing a conceptual tool capable of dif-
ferentiating between these layers of temporality. It can thus reflect how
spectators get caught up in their own attempts to render the cinematic
continuum comprehensible in causal-linear terms. This in a fashion that
does not seek to reconcile a realist narratological desire for establishing the
determinable (or, indeterminable) nature of the classical fabula.
I believe the reason Memento has proven so stimulating to f ilm-

theoretical and f ilm-philosophical debates is exactly because it is not
‘about’ current topics such as memory, narration, and personal identity,
but offers a cinematic experience that allows for an embodied and film-
philosophical reflection of these themes. Memento has become a site for
rethinking the complexity of the cinematic medium, yet both cognitive
and film-philosophical examinations tend to primarily perceive the film
as a cognitive puzzle (cf. Panek 2006; Buckland 2009b) that can or cannot
be solved by a ‘straightening out’ of the narrative continuum. According
to Scott Timberg (2001), who interviewed Nolan shortly after the film’s
US release, the director ‘insists he knows the movie’s Truth – who’s good,
who’s bad, who can be trusted and who can’t – and insists that close
viewing will reveal all’ (16). Thus, the director clearly sees a point in
engaging us with a narrative puzzle. Having meticulously dissected the
analytical fabula, Klein (2001) nonetheless concludes that he no longer
believes Nolan in his above-quoted remarks as ‘[t]he only way to reconcile
everything is to assume huge inconsistencies in the nature of Leonard’s
disorder’ (par. 60).
Therefore, I suggest to redirect our inquiries from the analytical-realist
question of whether a ‘true’ fabula can be established to the embodied
question of how this quest affects the experience of the viewer. Consider,
for instance, how Carrie-Anne Moss, who plays the femme fatale Natalie,
describes her numerous viewings: ‘I’ve seen [Memento] five times, [...] and
I’ve seen it differently each time now’ (Moss in Timberg 2001, 16). Rather
than structuring our perception of the film towards the construction of a
unified story, the embodied fabula honours the film’s carefully constructed
ambiguities and the singularity of each viewing experience.
In studying the film from the perspective of the embodied fabula, it
becomes possible to carve out the multilayered temporality of the film.
Although Mary Ann Doane (2002) herself perceives Memento to be an
example of narrative cinema’s tendency to recuperate temporal irrevers-
ibility and linearity (252, n. 49), it can be argued that the film exemplifies the
double temporal commitment that she observes in relation to early cinema.
That is, although Doane rightly detects a linear narrative drive in the reverse
chronology of Memento, she fails to acknowledge its non-linear aspects.
Non-linearity here does not simply spell out the backward or ‘non-linear’
telling of the film. Once confronted with the idea of Memento being a
‘non-linear film’, Christopher Nolan promptly answered as follows: ‘You
referred to the film as non-linear but in fact it’s very linear just in reversed
chronology’ (Nolan 2012).
The linear nature of the film is made explicit on the special edition of the
DVD-version, since it includes a chronological version of the film. This ver-
sion reveals how meticulously the film has been constructed to accentuate
the causal-linear relation of the individual scenes. Yet, the chronological
version not only affirms the linearity of the film (as ‘effect-cause’ rather
than ‘cause-effect’), but also the non-linearity of the cinematic continuum.
Although the events of the film remain the same, anyone who has seen
both the original and chronological version of the film is likely to have had
two qualitatively different cinematic experiences. Given that a non-linear
system undergoes a qualitative (rather than quantitative) change once its
components have been reassembled (cf. Chapter VI), the addition of the
chronological version draws our attention to the non-linear dimension of
the cinematic continuum. Again, Nolan himself has been the first to point
out this particular non-linearity of the film. In an interview from 2001 with
New Times Los Angeles, published shortly after the film’s initial release,
Nolan explains:
The whole idea was to make a film that bled into the mind a little bit,
spun in your head, that you constructed very much yourself. And when
I listen to [Radiohead’s album Kid A], no matter how many times I listen
to it, I don’t know what comes next. (Nolan in Timberg 2001, 14)
Here it becomes evident that Nolan does not envision the narrative as an
inherent feature to be discovered in the film. Instead, narration is a process
that occurs in an assemblage with the spectator. It would thus be a mistake
to assume that since the film is constructed according to a strict effect-cause
pattern, the cinematic experience it gives rise to must necessarily also
succumb to this logic. The film rather balances two separate logics at the
same time. First, there is the causal-linear logic according to which the
narrative continuum can be reassembled. This influences, yet cannot be
reduced to, the temporality conveyed in the complex cinematic experience
of the spectator trying to straighten out the cinematic continuum. The
film’s complexity owes a lot to letting its viewers experience the differ-
ences between these two layers of cinematic temporality. In carving out
the temporal complexity that is conveyed in the interplay of these, realist
interpretative methods reduce the temporal experience of the film to the
construction of a coherent fabula (cf. Bordwell 1985a, 33).
As suggested in previous chapters, the embodied approach within the cog-
nitive sciences may prove an important cornerstone for bridging cognitive
and film-philosophically-oriented positions (cf. Protevi 2010; Pisters 2012;
Brown 2013; Chapter V). Memento allows for a film-philosophical reflection

of this, since it provides the spectator with the possibility to sense or feel
time in a manner that arises from the cognitive difficulties in organizing
the narrative continuum linearly. In this fashion, it installs in the viewers
a unique mode of perception and a sensitivity towards temporal layers that
otherwise remain imperceptible. Memento can thus be perceived as a
‘meta-time travel story’, which is to say that it is ‘not one told by cinema, but
one enacted by the film-viewing experience’ (Baguer 2004, 250). The film is
a perfect expression of cinematic narration as a process not constructed by
the detached, computational, and cognitive machinery of the spectator with
the aim of establishing a narrative ‘truth’, but a process in which spectators
become cognitively, emotionally, and affectively immersed in the narrative
environment unfolding before them. It, in other words, demonstrates how
in contemporary complex cinema ‘narratives are not in us, but it is we who
move in a Being-narrative, a world-narrative’ (cf. Chapter V).
The Defamiliarization of Narrative Perception
Memento’s narrative complexity is not primarily caused by the film’s with-

holding of information or attempts to otherwise disrupt the spectator’s
construction of a chronological account of the narrative events. Actually,
Memento is very communicable (in spite of being ambiguous at certain
points), as evidenced by its abundance of temporal markers and cohesion
devices that help the audience to properly connect the events of the film
causal-linearly (cf. Bordwell 2006, 78-79). Yet, Memento presents its nar-
rative information in an order that complicates our ongoing attempts to
render experience linear. In this sense, Memento defamiliarizes the linear
cinematic perception that cognitive-formalism perceives as the consistent,
‘normal’, and ‘natural’ background against which other modes of cinema
can be evaluated (cf. Chapter III).
In carving out the workings of defamiliarization in Memento, the dif-
ference between the analytical and the embodied fabula proves valuable
(cf. Chapter V). In relation to this, it is vital to recall that the analytical
fabula structures our perception around the cognitive task of solving or
answering the questions of the narrative. Thus, Memento does not simply
defamiliarize cinematic perception as such. More precisely, it defamiliarizes
linear cinematic perception as structured according to the analytical fabula
(cf. Chapter III). In doing so, the film-philosophical encounter facilitated by
Memento plays a crucial role in the discernment of an alternative mode
of narrative perception, which is structured around the embodied fabula.

Here the spectator inhabits and enacts the narrative environment, which
is to say that cognitive problems, emotional states, and affective intensities
combine not primarily to solve but rather to explore the narrative universe.
As an extension of this, the concept of the embodied fabula makes it possible
to argue that Memento does not so much dissolve (Cartesian) subjectivity
as it reconfigures and alters its foundations.
Due to its defamiliarization of linear cinematic perception, Memento
calls for an approach that is able to combine the cognitive understanding
of active spectatorship with a film-philosophical sensitivity for the cin-
ematic encounter as direct, embodied, and affective. Unlike Stage Fright,
Memento does not limit its defamiliarization to a single cinematic device
(i.e. the flashback). In targeting cinematic perception more generally, the
film pushes the boundaries of defamiliarization beyond its restricted use in
cognitive-formalism, i.e. beyond ‘refamiliarization’ (cf. Chapter III). Taking
defamiliarization a step further allows spectators to feel and perceive those
automatic linear processes that structure cinematic perception to provide
the impression that we perceive a coherent, linear, and unified story.
It would be tempting to suggest that Memento in this manner brings
forth a ‘pure’ or ‘direct’ time-image. However, something quite different
is at stake since the perception of time here is neither facilitated by the
breakdown of the sensory-motor system nor from short-circuiting the
connections that tie affect, emotion, and cognition together. Memento
differs from the Deleuzian time-image by actively encouraging rather
than discouraging the linear processes that constitute ‘normal’ narrative
comprehension as described by Bordwell et al. In fact, Memento compels
the viewer to be more engaged with linear reasoning than the majority of
classical Hollywood narratives.
In connection to this, it has been noted that with Memento the ‘act of
recollection seems to be incorporated into [its] narrative strategy’ (Hes-
selberth & Schuster 2008, 98). In dramaturgically incorporating the act of
recollection, Memento departs from a classical cinematic use of flashbacks
as recollection-images (cf. Chapter IV). The film consciously plays on its
narrative reversal in a flashback scene, where Leonard has a hard time
accepting that his wife can read the same book repeatedly. ‘I always thought
the pleasure of a book was in finding out what happens next,’ Leonard as-
serts. Paradoxically, due to his condition, Leonard is constantly confronted
with the inverse of the typical narrative drive towards ‘what happens next’,
namely ‘how did I arrive here’, and the film’s narrative structure makes it
possible to maintain that this is the condition for its spectators, too.
Dark City (Proyas 1998), Eternal Sunshine of the Spotless Mind

(Gondry 2004), The Butterfly Effect (Bress & Gruber 2004), and In-
ception (Nolan 2010) are but a few examples of contemporary films in
which memory has become a central concern. Memento integrates the
contemporary fascination with memory into a narrative structure that
relies on the complex interrelation between remembering and forgetting.
Memento first requires us to adapt to a mental scheme of temporal integra-
tion without which the narrative remains incomprehensible (cf. Ghislotti
2009; Bordwell 2006, 78-81). The film further prompts us to pay attention
to certain details that enable the restoring of the lost causality of the story.
However, due to the difficulties in keeping the narrative structure organized
mentally – this also owes to the unreliability of the main protagonist and
the people trying to take advantage of his condition – spectators are forced
to constantly discard, overwrite, or reinterpret their version of the film’s
story. For these reasons, the narrative structure of Memento does not
merely challenge us to remember but also requires us to overwrite and
forget (cf. Cameron 2008, 107).
In relation to Bergson’s model of the cone (cf. Chapter IV, Fig. 2), it can
be argued that Memento oscillates between the most contracted regions
(of recollection) and the least contracted regions (of ‘pure’ virtual memory).
Here memory is conceived of as a palimpsest that carries traces or ‘memen-
tos’ of all earlier virtual possible pasts of the narrative. Hence, memory is
not to be understood in terms of a ‘container’ in which a part of the past
can be ‘stored’. This is in line with recent studies that suggest envision-
ing the memory faculty as a particular attitude towards, or structuring
of, the environment. As the embodied cognitivists Rolf Pfeifer and Josh
Bongard (2007) explain, ‘when we ask, Where is memory? we should perhaps
be looking not only inside the brain but at specific relationships between
the agent, its task, and its environment’ (321). In this sense, memory is a
particular mode of subtracting relevant information from the environment,
not a container for storing elements of the past in the brain.
This conception of memory is acutely expressed in a scene in Memento,
where Leonard has hired a prostitute to do the morning routines of his de-
ceased wife. In the scene, Leonard constructs his environment in a specific
manner to conjure a particular memory. Given that the film is about a man
who has lost his memory, it is interesting to note how pervasively the narrative
universe is structured around (forgotten) memories, whether in the form of
Leonard’s tattoos, his Polaroids, his wife’s items, or countless other ‘mementos’.
The double nature of memory is manifested, as Leonard while burning some
of his wife’s old possessions states: ‘Can’t remember to forget you’.
Janet Harbord (2007) has argued that contemporary reverse chronologi-

cal narratives couple epistemological questions (‘how do I as a spectator
construct the narrative’) with ontological questions (‘how would it be to
move backwards in time’) to facilitate awareness of the extent to which ret-
rospective knowledge informs (narrative) experience (127; cf. Hesselberth &
Schuster 2008, 99-100). Thereby, this type of film renders visible the crucial
role that the spectator’s cognitive-perceptual and memory capacities play in
the construction of the narrative, while questioning this narrative condition
at the same time. In demonstrating that memory is not a container of the
past but a ‘dynamic process of appropriation, association and translation of
the past (in)to the present’ (Hesselberth & Schuster 2008, 100), Memento
not only questions the cognitive processes that constitute our ability to
comprehend cinematic narratives, it moves cognition outside the rigid
boundaries of the individual brain.
Elsaesser perceives Memento as an example of the ‘post-mortem cinema’,
which is a cinematic mode related to the ‘post-humanist’ philosophy of Deleuze
and Foucault (cf. Elsaesser 2004b; Elsaesser & Hagener 2010). According to
Elsaesser & Hagener (2010), films like Memento challenge the Cartesian and
classical cognitive scientific idea of a unified and self-assured cogito presup-
posed by the classical narratological fabula. In the post-mortem cinema,
[t]he mental and conceptual images [...] have to do with the limits of
classical identity formation, where we assure ourselves of who we are
through memory, perception and bodily self-presence. When these
indices of identity fail, or are temporarily disabled, as in conditions of
trauma, amnesia or sensory overload, it challenges the idea of a unified,
self-identical and rationally motivated individual, assumed and presup-
posed by humanist philosophy. (155-156)
Jamie Skye Bianco (2004), too, perceives Memento as an affective dissolu-

tion of the Cartesian subject. For her, ‘[t]he image/matters of Memento
affectively force actions of forgetting: forgetting how we have conditioned
sensation, thought, and action; forgetting chronology and the structures of
narrative genres; and, forgetting our acts of centring’ (384). In accentuating
what has become automatic and familiar in the process of forgetting, Bianco
touches upon the defamiliarization of Memento. However, from her per-
spective this defamiliarization amounts to a total dissolution of subjectivity.
Cinema is an explosion of energetic flows, and the spectator can no longer be
reduced to a cognitive-hermeneutic machinery engaged in the reconstruc-
tion of causal-linear patterns and trajectories on the basis of the cinematic
input. In her inspiring study of the ‘de-subjectivizing’ affectivity of modern

‘techno-cinematic events’, Bianco however, fails to also acknowledge the
continued importance of ‘subjectivizing’ cognitive processes.
Whereas cognitive film theory can be criticized for largely ignoring the
cinematic and aesthetic experience, Hayles (2004) has summed up a problem
that I believe could apply more generally to Deleuzian ‘affect theorists’. She
maintains that ‘Bianco rightly observes that an analytical approach does
not explain [Memento’s] affective dimension, but her affective enactment
does not account for the film’s structural precision, including the enfolding
of the two chronologies at the film’s conclusion’ (314).
This neglect can be traced back to the prominent idea of Deleuze’s cinema
books that cinema’s potential to liberate time from its subordination to
movement lies in the ability of the time-image to rupture the sensory-motor
system. Massumi has taken inspiration from this notion to formulate his
comprehension of affect as more or less disconnected from the cognitive
spheres of signification, subjectivity, and reason (cf. Chapter II). Yet, I would
contend that Memento demonstrates how affect and time can become
perceptible in cinema not only through modern(ist) ruptures, but also by
a reconfiguration of the interrelation of cognition, emotion, and affect and
a defamiliarization of linear cinematic perception.
To appreciate how this works, we must again turn our attention to how
the narrative structure of Memento hinders some of the basic functions
of memory. The film in this way achieves an embodied mode of alignment,
since the spectator will experience some of the difficulties of the film’s main
protagonist. Due to films like Memento, Elsaesser (2009) proposes that
contemporary complex narratives
imply and implicate spectators in a manner not covered by the classical

theories of identification, or even of alignment and engagement, because
the ‘default values’ of normal human interaction are no longer ‘in place,’
meaning that the film is able to question and suspend both the inner and
outer framing of the story. (30)
From the perspective of the embodied fabula, cinema is perceived as a nar-

rative rhythm, whose temporal flows have not been predetermined by the
linear trajectory of a story. In Memento’s backward flow of narration, the
linkages between cognition, emotion, and affect are not ruptured, broken,
or disassembled (as in the Deleuzian time-image), rather their interrela-
tion is constantly being reconfigured and renegotiated. One of the great
achievements of modern art and philosophy has, according to Maurice
Merleau-Ponty (2004), been to ‘allow us to rediscover the world in which

we live, yet which we are always prone to forget’ (39). The rediscovering
of the world is not the result of purely intellectual or cognitive capacities,
but concerns the reawakening of embodied perception. Therefore, rather
than merely arousing intellectual and reflective detachment, artworks like
Memento provide an empirical possibility to embody another mode of
experiencing the world:
[T]he transition from classical to modern was marked by what might be

thought of as a reawakening of the world of perception [...] So the way we
relate to the things of the world is no longer as a pure intellect trying to
master an object or space that stands before it. Rather, this relationship is
an ambiguous one, between beings who are both embodied and limited
and an enigmatic world. (69-70)
Contemporary complex cinema enables an embodied mode of perception

often ignored by the classical linear onto-epistemology. Merleau-Ponty
states: ‘It is well known that classical thought has little time for animals,
children, primitive people and madmen’ (70). Elsaesser has observed how
especially the latter of these has been the subject of extensive attention
within contemporary cinema. What appears to fascinate contemporary
cinema about these ‘madmen’ is the extent to which they are able to grant us
an estranged perception of the world. Observing this Elsaesser (2009) writes:
Indeed, the point of giving such subjectivities-in-action the format of a

mind-game film would be to draw the audience into the protagonists’
world in ways that would be impossible if the narrative distanced itself
or contextualized the hero via his or her (medical) condition. (30)
Contemporary complex narratives hereby challenge the classical conditions of

cinematic spectatorship described by Bordwell. Furthermore, these films take
us beyond a Smithian engagement or alignment with the characters, since the
set of ‘normal’ cognitive processes associated with narrative comprehension
is encouraged and simultaneously defamiliarized. However, this does not
necessarily mean that the spectator’s sense of subjectivity disappears. Rather,
it can be argued that the notion of subjectivity must instead be redefined
as emerging out of our bodily engagement with the narrative environment.
Karen Renner (2006) has examined the emotional markers of Memento,
and proposed that rather than being a result of textual processing, they ‘condi-
tion’ viewers to respond with an appropriate emotional response (here sadness)
to the images. Aside from the recurring musical theme of the film, Renner
mentions the close-ups of Leonard’s sad facial expressions as he mourns his
dead wife as examples. Unlike traditional cognitive theories that highlight
the rational dimensions of our emotional responses to cinema, Renner (2006)
is interested in those cognitive processes that are largely ‘unconscious and
that occur somewhat independently of the film’s plot or outcome’ (107). She
argues that ‘emotion cues’ do more than act as instigators and sustainers of
mood (cf. Carroll 2001), and maintains that these can operate more directly
to condition the viewer to a particular state of mind (Renner 2006, 107).
Thus, Memento demonstrates the intricate nature of emotions as situ-
ated in the gap between affect and cognition. Initially the film structures
our emotional experience along a classical pattern, insofar as it invites us to
sympathize with Leonard, who has been the victim of a terrible crime from
which he still suffers greatly. Thereby, the film allows us to experience a
rational underpinning to our emotional states by playing with the cinematic
conventions pertaining to the righteous hero. This initial alignment is
supported by what Greg M. Smith (2003) has termed ‘mood cues’. According
to him, the act of ‘generating brief, intense emotions often requires an
orienting state that asks us to interpret our surroundings in an emotional
fashion’ (42). In addition, ‘film structures seek to increase the film’s chances
of evoking emotion by first creating a predisposition toward experiencing
emotion: a mood’ (42).
Renner (2006) explores four moments in Memento whose primary func-
tion she argues is the ‘creation of sympathetic sadness’ (108). The central
focal points of these scenes are Leonard’s grief over the loss of his wife, and
his inability to move on. As Leonard describes in a scene designed to create
sympathy for his character:
I know I can’t have her back, but I want to be able to let her go. I don’t want
to wake up every morning thinking she’s still here then realizing that she’s
not. I want time to pass, but it won’t. How can I heal if I can’t feel time?
For Leonard the memory of his wife continues to pervade his perception,
while no new memories are allowed to enter his conscious state. Leonard’s
condition is thus misconstrued in terms of a loss of memory, insofar as
memories are what define his existence. Towards the end of Memento,
Leonard explains his condition to the undisclosed person (presumably
Teddy) on the telephone: ‘You know the truth about my condition, officer?
You don’t know anything. You feel angry, you don’t know why. You feel guilty,
you have no idea why.’ Here we find an acute description of the rupture
in Leonard’s cognitive-emotional-affective circuitry, which has left him

bereft of the ability to connect his feelings of guilt or anger to a subjective
and causal-linear experience. Leonard’s ‘memory system’ acts instead in
accordance with the ‘autonomic nature of affect’ (Massumi 2002, 28).
Similarly, although the film gradually renders problematic our initial
apprehension of Leonard, our original sympathetic stance continues to
linger on. An affective compassion with Leonard thereby continues to
resonate in our emotional-affective systems, despite not being cognitively
sustainable any longer. Our allegiance to Leonard thus becomes permeated
by dissonances in the feedback loops of affect, emotion, and cognition.
Emotion and feeling here take on central roles, because they ‘provide the
bridge between rational and nonrational processes, between cortical and
subcortical structures’ (Damasio 1995, 128). However, in destabilizing our
emotional allegiance with Leonard, Memento does not so much break the
linkages that tie cognition, emotion, and affect together, as it defamiliarizes
the invisible underpinnings of these processes. The film thereby expands
the boundaries of what is cinematically perceptible.
In this fashion, the defamiliarization of Memento takes us beyond
the normal and successful operations of human cognition. Insofar as it is
‘difficult to keep its stimuli distinct from our mental construction’ (Ghis-
lotti 2009, 98), the actual experience facilitated by Memento is marked by
cognitive dissonances and (temporal) confusion. From the perspective of
the embodied fabula, such dissonances come with a productive and creative
potential to open up for an awareness of the deeper-lying non-conscious
processes that constitute linear cinematic perception.
Consider the following three viewer reactions that Ghislotti (2003) ac-
centuates in an earlier text on the film:
Viewer 1: I’ve seen the movie three times now and may have to watch it ten
more times until I get it all straightened out.
Viewer 2: This movie was brilliant because it totally got me dizzy... never be-
fore can I recall concentrating so hard on what was going on... eventually,
I hit a mind warp and got totally lost forgetting how things ended thus
making the facts in the beginning a dizzying of feelings and a distortion
of my OWN memory.
Viewer 3: I loved this movie because it made me feel as if I had a short-term
memory deficit.
These comments illustrate the interrelation between aspects of the cin-

ematic experience that are cognitive-analytical (the ‘straightening out’ of
the narrative), affective-embodied (a ‘dizzying of feelings and a distortion
of my OWN memory’), or induce film-philosophical reflections about the

cinematic experience (as if ‘I had a short-term memory deficit’).10
Although the comments stem from three different viewers, I believe
each alludes to the same interrelated experience. Yet, the first describes
the experience from the perspective of the analytical fabula, while the
latter two describe their experience from the perspective of the embodied
fabula (cf. Chapter V). For the cognitively and analytically detached viewer
the important task is to straighten out and solve the cinematic narrative,
whereas the latter two perceive the cognitive difficulties as part of the
unique cinematic experience that the film propounds.
Following the Deleuzian differentiation of Riemannian and Cartesian
space, it can be argued that spectators further reflect upon their cinematic
experience from different spatio-temporal registers. Brown (2013) explains
the differences between these as follows:
While consciously I perceive movement and ‘automatically’ infer where

the moving figures will go, and/or work out how they got from point A
to B in space, nonconsciously I am aware of the whole flux of becoming
which lies ‘between’ geometrical ‘points in space’ that consciously are
apprehended only ‘after the fact’ (Riemannian, as opposed to Cartesian,
space), or which, in the case of predicting where figures are headed and/
or what they will do, allow me to project causality into the movement
in order to arrive at its effect [...] This nonconscious perception is not of
point-to-point geometric space, then; nor is it of figures changed and/or
presumed states of change that will result from the present conscious
perception. This nonconscious perception is instead of time itself, which
lies outside of classical cause and effect. (133-134)
Memento allows us to grasp how the cinematic experience is constituted

in constant feedback loops between these layers of experience. In this
fashion, the film opens up our senses to – and thus defamiliarizes – the
underlying processes that traditionally ensure our experience of causal-
chronological cinematic perception as ‘natural’. This process does not differ
fundamentally from that whereby Stage Fright defamiliarizes the classical
flashback that over time had come to appear natural (cf. Chapter III).
In the context of how Leonard’s voice-over ‘mirrors what constructivists call
the endless autobiographical dialogue we hold with ourselves’ (Ferenz 2009,
274), the film renders visible the automatic cognitive processes that constitute
linear cinematic perception. These give direction to time by automatically
extending the past into the future making the narrative continuum appear
unified and continuous. Yet, they also ascribe meaning to our experiences in
a manner that accommodates our ongoing autobiographical or ‘autopoietic’
narratives (cf. Bruner 1991, 2004; Young & Saver 2001). Thus, it can further be
argued that Memento in constructing an allegiance between Leonard and
the spectators defamiliarizes the autobiographical processes that structure
perception according to the ongoing narrative of the individual.
The alignment between Leonard and the spectator’s narrative construc-
tion is further facilitated through the film’s extensive use of voice-over. This
technique causes the main character’s efforts to restore spatio-temporal
coherence to mirror those of the spectators trying to navigate the narrative
universe unfolding before them. The repetition in a variety of scenes where
Leonard – presumably in synchrony with the audience – asks: ‘So where
are you?’ marks the manner in which his voice-over runs parallel with the
spectator’s cognitive efforts to construct narrative sense (cf. Bianco 2004).
The voice-over thus connects with, reflects upon, and eventually obstructs
the spectator’s own ‘autobiographical dialogue’.
To capture the deeper-lying strata of the cinematic experience, defamil-
iarization must be able to go beyond the cognitive-formalist separation of
‘habitual’ from ‘aesthetic’ perception. Kristin Thompson (1988) renders this
separation explicit, when she contends:
Because everyday perception is habitual and strives for a maximum of

efficiency and ease, aesthetic perception does the opposite. Films seek
to defamiliarize conventional devices of narrative, ideology, style, and
genre. Since everyday perception is efficient and easy, the aesthetic film
seeks to prolong and roughen experience – to induce us to concentrate
on the processes of perception and cognition in and of themselves, rather
than for some practical end. (36)
The problem with this differentiation is that it presumes the formal aspects of
a film to be determinant of the range of (cognitive) responses available to the
audience. Having strongly positioned her theory as one of active spectatorship,
it is noteworthy that Thompson assumes that the viewer can only ‘respond
to a film in the way that the film wants her to, i.e. passively’ (Brown 2013,
134). This separation harbours a reductive conception of perception as being
generally automatic, linear, and efficient-prone. Given the constant feedback
loops in our cognitive-perceptual registers, we nevertheless do not always
experience our environment in an automatic manner. Changes in our ha-
bitual environment should therefore be accepted as being – in principle at
least – capable of defamiliarization to the same degree as art.
In this context, defamiliarization is understood in a more complex,

entangled, and dynamic manner than in the cognitive application of the
term. In my use, the term thus operates where the borders between ‘life’
and ‘art’ constantly overlap. As Jerome Bruner (2004) has argued:
The mimesis between life so-called and narrative is a two-way affair: that
is to say, just as art imitates life in Aristotle’s sense, so, in Oscar Wilde’s, life
imitates art. Narrative imitates life, life imitates narrative. ‘Life’ in this sense
is the same kind of construction of the human imagination as ‘a narrative’
is. It is constructed by human beings through active ratiocination, by the
same kind of ratiocination through which we construct narratives. (692)
From this perspective, it no longer makes sense to talk about a strict distinc-
tion between the spectator and the medium, the object or the subject, the
body or the mind. Instead, narratives and cinema have entered the domain of
perception ‘proper’, and become active players in structuring perception, also
beyond the ‘proper’ cinematic situation. As Elsaesser (2004a) has remarked:
In making much of human life and history ‘visible,’ the cinema has also
created new domains of the ‘invisible.’ Key elements of cinematic percep-
tion have become internalised as our modes of cognition and embodied
experience, such that the ‘cinema effect’ may be most present where its
apparatus and technologies are least perceptible. (76)
In rendering difficult the automatisms of temporal integration and autobio-

graphical narration, Memento does not simply defamiliarize the analytical
fabula as a coherent entity; it defamiliarizes how this concept structures
experience as a mode of perception. The film thereby demonstrates that
the fabula is ill-conceived as an inherent feature of the narrative itself and
instead reflects how cinema is capable of producing its own subjectivity.11
From the perspective of the embodied fabula, the narrative is thus not
constituted when a detached cognisor masters the film to impose upon it a
‘true’ representation of its inherent story. Instead, the subject is formed out
of the cinematic experience itself. According to Damasio (1995):
Perceiving the environment [...] is not just a matter of having the brain
receive direct signals from a given stimulus, let alone receiving direct
pictures. The organism actively modifies itself so that the interfacing can
take place as well as possible. The body proper is not passive. Perhaps
no less important, the reason why most of the interactions with the
environment ever take place is that the organism requires their occur-
rence in order to maintain homeostasis, the state of functional balance.
The organism continuously acts on the environment (actions and explora-
tion did come first), so that it can propitiate the interactions necessary for
survival. But if it is to succeed in avoiding danger [...] appropriate actions
can be taken in response to what is sensed. Perceiving is as much about
acting on the environment as it is about receiving signals from it. (225)
The circuits and linkages of the brain do not pre-exist the stimuli, corpus-
cles, and particles that trace them, as Deleuze (2000) has argued, and thus
cinema ‘makes bodies out of grains’ (366). In comprehending Memento,
spectators initially utilize the analytical fabula, but in doing so our attempt
to induce the film with causal-linear order turns back on itself, and in this
manner, we are enfolded or embedded into the narrative universe. In this
fashion, the film restructures our perception in terms of the embodied
fabula as a corporal-mental tool that guides the immediate interaction with
the narrative according to a Möbius strip logic such that our cognitive infer-
ences and visceral affectedness form a feedback loop. It is in this loop, rather
than by cognitive inferences alone, that spectators are able to construct and
embody the narrative universe or ‘film world’ (cf. Yacavone 2015). Following
this logic, the narrative is no longer something we merely impose on the
film as analytically detached spectators from the outside (‘narrative is not in
us’), but something which is created in the encounter with the film (‘we are
in a Being-narrative’). Ultimately, narrative comprehension can no longer
be detached from our bodily, affective, or emotional responses to the film.
Simultaneously, the cinematic experience emerges as something more
than just a ‘machine for generating affect’ (cf. Shaviro 2010, 3). In the theory
proposed here, the fabula transforms into an embodied, operational, and
structuring tool that guides the spectator’s exploration of the narrative
environment cognitively, emotionally, and affectively. This provides
us with a new narratological approach to the aesthetic and beauty of
Memento as something not primarily a function of the film’s narrative
puzzle (cf. Buckland 2009b), but rather something that pertains to how
viewers become absorbed into the temporalities, narrative rhythms, and
the sensational flows of what asks to be conceptualized as an embodied
cinematic experience.
VIII. Conclusions
So we can follow this movement, from beauty to the world to everything that
is behind the world. All of it enters through our eyes and our ears, vibrating
directly on us. What this says is: there is a world out there and it is huge and I am
in it. Leviathan [Castaing-Taylor & Paravel 2012] doesn’t ask us to bring anything
to it; it doesn’t need our stories, our knowledge, our prejudices. In its indifference
to our presence it’s a nice reminder of why the movies can be so affecting in the
first place, of what it feels like to be confronted with something that is truly
outside of everything our experiences have taught us.
‒ Coldiron 2012, par. 12
The aim of this book has been to offer a framework for comprehending
the transformative nature of cinematic spectatorship, especially as it has
recently been altered by the fluid boundaries of contemporary media culture
(cf. Pisters 2012, 11). Contemporary complex cinema forces us to rethink and
reconfigure the linear-non-linear dichotomy of film studies that harks back to
the opposition of classical cinema and the tradition of (European) art cinema.
As a consequence, complex cinema is no longer well accounted for by the
overarching Deleuzian categories of the movement-image and the time-image.
Instead of emphasizing the difference between the classical and the modern,
contemporary cinema explores their mutual inextricability. Required is thus
an approach that emphasizes the interrelation of what has traditionally been
kept isolated, such as the linear and non-linear; the affective, emotional,
and cognitive investment of the audiences; the contingent from the causal-
determined; the body from the mind; etc. In their experiments with different
modes to invoke a unique narrative experience, complex films render the
limitations of reducing the spectator’s narrative engagement to the cogni-
tive processes involved in organizing the narrative events causal-linearly
visible. Similarly, once confronted with the ‘will to complexity’ that defines
contemporary cinema, several prevalent concepts (e.g. the fabula, background,
or defamiliarization) have begun to reveal their explanatory limitations.
Following the main argument raised by this book, contemporary cinema
demands a reformulation of cinematic spectatorship by revealing the
limitations of the classical narratological fabula insofar as this concept
1) insulates spectators’ cognitive responses from their emotional and af-
fective responses to the film and 2) conflates the multilayered cinematic
temporality to its causal-linear predispositions. Although the fabula as
a f ilm-narratological concept is mostly associated with the cognitive

formalism of Bordwell et al., the core assumptions and the analytical
procedure that inevitably accompany this concept rehearse an influence
on Deleuzian approaches in their lack of a more suitable theory of nar-
ration, too. In the case of the latter, however, the fabula is mostly used
implicitly along the linear-non-linear dichotomy to demonstrate how a
given film either adheres to an underlying linear logic (and thus allows
for the construction of a coherent fabula) or completely refutes such logic
(and thus no unambiguous account of the fabula can be constructed). In
this fashion, the complexity of contemporary cinema is reduced to the
complex telling of a more or less complex story. Therefore, I have argued
that cognitive-formalist and Deleuzian film-philosophy could both profit
from a notion of the fabula not founded on the philosophical principles
of computation and representation. I have proposed the amalgamation of
embodied cognition, Deleuzian (film)philosophy, and complexity theory as
the basis of my embodied reconceptualization of the fabula. This book has
drawn the outlines of this new narratological concept, which nonetheless
still remains to be further explicated and its value to a broader embodied
re-examination of cinematic narration needs to be proven.
The embodied fabula has been film-philosophically formulated with ref-
erence to a series of complex narratives. It has thus initially been conceived
as a tool for cutting across the ‘classical-modern divide’ in cinema (cf. Kovács
2007, 33-48) and as a means for exploring cinema beyond the linear-non-
linear dichotomy. In relation to this task, complex narratives have been
granted a formative role insofar as they reinvigorate cinematic complexity
by seeking to combine elements that have traditionally been kept apart and
in doing so call for a reconceptualization of basic narratological concepts.
This counters the onto-epistemological roots of our preferred interpretative
procedures, according to which the prototypical task of the analyst is a
‘straightening out’ or a ‘decomplexification’ of the narrative continuum.
Instead, the embodied fabula has been designed to explore cinema between
the lines of traditional cinematic binaries. As a temporally dynamic concept
based on the idea of feedback loops in our affective, emotional, and cogni-
tive circuits, the embodied fabula will hopefully uncover new venues of
cinematic spectatorship. It aims to do so by drawing attention to the actual
(temporal, spatial, and cinematic) experience of the viewers that has been
neglected in the era of the cognitive-formalist fabula.
In their clear separation of ‘nature’ and ‘culture’, cognitive-formalist
explanations obscure the feedback loops according to which the ‘natural-
ness’ of Hollywood continuity and the broader manner in which humans
Conclusions 207
organize their experiences according to a causal-linear onto-epistemology

constantly reaffirm each other. However, I have insisted that the film-
philosophical alternative, the valorization of the non-linear dimension of
cinema, risks reversing the dichotomy in foregrounding a non-semantic,
autonomous, virtual, singular, and unpredictable dimension of cinema,
which maintains to be disconnected from any fixed or conventional mean-
ings (cf. Leys 2011, 449; cf. Chapter II & Chapter IV). In doing so, I have
expanded on the arguments that can be found in the growing body of
research that has embraced the challenge of contemporary cinema to think
beyond the prevalent binaries of film studies (Brown 2013; Pisters 2012;
Shaviro 2010; Yacavone 2015); an important task as the linear-non-linear
dichotomy is rooted and implemented in our conceptual, narratological,
and theoretical frameworks. In this and other respects, I hope to have
proposed an alternative to the dominant narratological adherence to clas-
sical scientific principles designed for the reduction of complexity.
Classical narratological tools such as the fabula, however, should not
simply be abandoned in favour of a turn to affect as that which ‘eludes
form, cognition, and meaning’ (Leys 2011, 450; cf. Chapter II). To dispute
the claim that our bodily connection to the world can be separated from
our cognitive comprehension of it – a tendency within (Deleuzian) ‘affect
theory’ (cf. Massumi 2002; Shouse 2005) – I have turned to embodied cogni-
tive neuroscience and complexity theory. Alfred Hitchcock’s Stage Fright
and Alain Resnais’s and Marguerite Duras’s Hiroshima mon amour in
different manners reveal certain limitations involved in overemphasizing
the linear or the non-linear dimension of cinema. In relation to the former,
the widespread acceptance of the classical film as ‘linear’ not only stems
from a lack of sensitivity towards the inherent ambiguities that reside in the
cause-effect dramaturgy of classical cinema, but also to an unawareness
of the degree to which this linearity is produced by our own analytical
procedures and conceptual tools. On the contrary, Hiroshima mon amour
expresses an inherent paradox of Deleuzian film-philosophy: the combina-
tion of a non-linear valorization with an uncritical acceptance of classical,
linear narratological tools such as the fabula.
The critical dissection of the linear-non-linear dichotomy thus needed
to be accompanied with a rethinking of core conceptual tools such as the
fabula and defamiliarization. 21 Grams, Lola rennt, and Memento have
been examined as film-philosophical ‘encounters’ capable of demonstrating
how the fabula as an analytical and narratological concept inevitably struc-
tures our comprehension of cinema according to the classical scientific aim
to ‘search, behind those appearances, the hidden order that is the authentic
reality of the universe’ (Morin 2007, 6). The result of this process appears
to be either a reduction of the complex spatio-temporal experience of the
viewers to a linear principle, or the determination of the indeterminate
nature of the narrative in question (cf. Chapter VI & Chapter VII). Missing
in today’s film theoretical landscape is a theoretical and methodological
alternative that emphasizes the cinematic experience in favour of a classical
narratological search for a coherent, chronological, and unified narrative.
By challenging the representational and computational foundation of
the fabula, it has been possible to conjure up a complementary take on
this imperative narratological tool; sensible to both the higher cognitive
processes and the aesthetic dimension of cinema.
In relation to this, it has been incisive to allow the films themselves
to take an active part in the reconf iguration of our analytical devices
and interpretative strategies. Complex narratives, as I have argued, do
not privilege affect over emotion or cognition; however, they do call for a
reconfiguration of their relation. To rethink the cinematic experience in
terms of the interactions of cognition, emotion, and affect, this study has
turned to the emerging research field of embodied cognition (cf. Chapter V).
Insofar as such interactions are ill-understood from a context-independent
framework designed to separate cognitive difficulties from each other
(cf. Morin 2007), I have followed the proposal of John Protevi (2010) to draw
upon Deleuzian ontological insights to further strengthen the philosophical
claims of embodied cognition and complexity theory. In this manner, cin-
ema has not only been granted a formative role in revealing the limitations
of traditional narrative approaches structured around the establishment of
a coherent, chronological, and unified fabula, but also provides intriguing
film-philosophical ‘encounters’ that enable us to re-explore highly relevant
philosophical, psychological, and neurological issues with reference to the
cinematic experience.
Going beyond the linear-non-linear dichotomy thus also entails a move
beyond the usual limitations of the sphere of cinema. Consequently, de-
bates concerning the temporal and narrative complexity of contemporary
cinema are not merely of film-theoretical relevance, since cinema’s mode
of engaging the spectator grants us with a renewed perspective on human
perception, cognition, and affect. This is especially evident in relation to the
growing neurological interest in cinema that allows us to frame questions
concerning the human brain differently. Nowhere are the ties between
cinema, philosophy, and science, as well as current societal, political,
and moral questions, as tightly woven together as in the questions that
contemporary complex cinema raises. As Pisters (2012) has suggested,
Conclusions 209
in contemporary media culture the brain and the images converge (and
become ‘neuro-images’). Antonio Damasio (2011) presents a kindred idea
from a neurological perspective, when he argues that ‘[m]ovies are the
closest external representation of the prevailing storytelling that goes on in
our minds’ (188). In extension, Gallese and Guerra (2012) have argued that
‘[f]ilm is a possible target of investigation for cognitive neuroscience, and for
a variety of good reasons’ (183). While cinema and neuroscience may shed
valuable light on each other, it is, however, vital not to reduce the cinematic
experience to the evolutionary hardwired biological machinery of the brain
according to a moderate interpretation of embodiment (cf. Chapter V). The
risk is that we forget the actual films and the cinematic experience that
we opted to explain in the first place. Before turning to neuroscience and
related fields, we should thus turn to the films themselves.
One possibility is to turn to the fragmented opening of 21 Grams to
examine how it provokes a reconfiguration of the ‘affective-emotional-
cognitive’ circuitry of the spectator, since the affective ‘incipience to act’
(cf. Massumi 2002) invoked by the images can find neither emotional nor
cognitive discharge. Another possibility is to examine how Lola rennt
does something comparable by tying us affectively and cognitively to the
kinaesthetic of its main protagonist, to its energetic soundscape, and to
its dynamic, narrative rhythm (cf. Chapter VI). We could also turn to the
backward narrative structure of Memento, which arguably does not so
much dissolve (Cartesian) subjectivity as it reconfigures and alters its
foundations to demonstrate its connection to a constant ongoing pro-
cess of defamiliarization (cf. Chapter VII). The point being that cinema
contains an endless cabinet of examples that could be used to rehearse
film-philosophical ‘encounters’ allowing the complexity of the films to
challenge the monopoly of classical narratological concepts.
In this fashion, the temporally complex cinematic experience offered by
Memento has been incisive for the formulation of the embodied fabula as a
concept operating in the interstices of the linear/cognitive and non-linear/
affective dimensions of cinema. As a corporeal-mental tool, whose mode of
communication is bidirectional and dynamic (due to feedback loops in the
circuits of cognition, emotion, and affect), this concept has in turn enabled
me to study how Memento restructures cinematic perception by taking
advantage of our attempts to ‘linearize’ the narrative in order to fold us into
the narrative according to a logic, which can be carved out in reference to
the feedback loops of the Möbius strip (cf. Chapter V & Chapter VII). Here
narrative is no longer something we impose on the film from the outside
(‘narrative is not in us’), but something which is created in the encounter
with the film (‘we are in a Being-narrative’). One of the main advantages
of the embodied fabula is that it offers a conceptualization of the narrative
experience that does not insulate our cognitive efforts from the affective
or emotional manner in which we ‘inhabit’ the narrative.
This perspective also reconfigures the traditional notion of the fabula,
which I have attempted to capture with the notion of the analytical fabula.
It is incisive, however, to note that the analytical fabula should not be
understood as a ‘representation’ of the ‘actual’ story of the film (as is often
implied in the classical narratological use of the fabula). It instead desig-
nates a prominent mode of structuring cinematic perception described by
Bordwell (1985a) as a strive for the construction of a chronological, unified,
and coherent story (i.e., linear cinematic perception).
Consequently, the embodied fabula cannot simply be incorporated into
a cognitive-formalist framework; it actively demands a reconfiguration
of the cinematic experience, because it no longer assumes that narrative
comprehension can be adequately explained from the perspective of a
detached spectator who cognitively masters the film in order to impose
upon it a ‘true’ representation of its inherent story. Hence, the embodied
fabula no longer places as essential for the narrative experience the pos-
sibility of the construction of a coherent and unified story; neither does it
presuppose an analytical and temporal displacement between the analyst
and the cinematic experience (cf. Chapter II). Being an embodied cognitive
tool that guides the spectator’s immediate interaction with the narrative,
the embodied fabula allows us to explore the complex temporal layers of a
given film from the perspective of the actual cinematic experience.
The Embodied Fabula and Beyond
As the concept of the embodied fabula has been developed from a film-
philosophical encounter with contemporary cinema and how it deals with
questions of complexity, this concept should be seen in the broader context of
a more comprehensive reconceptualization of cinematic spectatorship. While
the embodied fabula, as it has been formulated here, is intimately associated
with complex narratives, I believe it could also contribute to our understand-
ing of how cinema is currently undergoing a transformation driven by a ‘will
to immersion’ supported by the technological inventions of digital imagery
and 3D cinema. I will briefly refer to Alfonso Cuarón’s box-office success
Gravity (2013) and the experimental documentary Leviathan (Castaing-
Taylor & Paravel 2012) to elucidate two interesting developments within the
Conclusions 211
cinematic medium, whose novel mode of engaging spectators may be usefully

excavated with the concept of the embodied fabula.
A product of the Harvard Sensory Ethnography Lab (SEL), Leviathan
presents a fascinating new take on the documentary that leaves us ‘with
only our senses to follow in the dark’ (Coldiron 2012, par. 2). In doing so,
the film becomes an audiovisual expression of the declared aim of SEL
to ‘support innovative combinations of aesthetics and ethnography, with
original nonfiction media practices that explore the bodily praxis and
affective fabric of human existence’ (Sensory Ethnography Lab: Harvard
University 2010). With its extended use of GoPro cameras that have been
tied to the fishermen’s bodies, thrown into the waves, and caught like the
fish in the nets, the film literally embodies the naval existence off the coast
of New Bedford – the ‘Whaling City’ of Herman Melville’s Moby Dick. Due
to its creative use of GoPro cameras, Leviathan is an expression of the
innovative powers of digital cinema and has been described as ‘less surreal
than hyperreal, flooding the senses, and fashioning an almost nightmarish
environment with an assault of digital information’ (Goldsmith 2013, par. 2).
The film expresses the desire of digital cinema to be both immediate and
hypermediate (cf. Bolter & Grusin 2000), ‘marked both by the capacity to
create new worlds through motion-capture and 3-D rendering, and by its
intimacy with the real world, its physical proximity to our bodies and our
experience’ (Goldsmith 2013, par. 3).
In a much more radical manner than complex narratives, Leviathan
defamiliarizes linear cinematic perception as structured around the ana-
lytical fabula. Consequently, film critic Scott MacDonald (2012) comes to
the conclusion that ‘[c]ritical detachment will remain difficult at least for a
few more screenings’ (par. 7). Leviathan thereby overthrows the rational
cognition associated with the analytical fabula to provide MacDonald
with an experience he describes as being ‘as overwhelming, thoroughly
immersive, and unpredictable as the ocean itself’ (par. 7). Leviathan is a
vivid example of a cinematic exploration of ‘haptic visuality’ in which the
‘eyes themselves function like organs of touch’ (Marks 2000, 162; cf. Deleuze
2005a, 2005b; Marks 2002; Beugnet 2007; Elena del Rio 2008). In relation
to this, the embodied fabula could contribute to explain more concisely
how in watching films like Leviathan ‘we are certainly not in the film,
but we are not entirely outside it, either. We exist and move and feel in
that space of contact where our surfaces mingle and our musculatures
entangle’ (Barker 2009, 12, emphases in original). Understanding the cin-
ematic experience as embodied does not suggest that cinema gives rise to an
entirely affective or bodily modality that can be (theoretically) disengaged
from our cognitive and intellectual capacities. To better comprehend how

Leviathan replaces the linear narrative movement of the analytical fabula
with an immersive and bodily sensation of temporality, a conception of
narration and cognition that is not predisposed to linearity is thus required.
In having explored the theoretical and methodological foundation of such
a framework – most evidently with the development of the embodied
fabula – this book has paved a potential way of studying the immersive
spectatorship of films like Leviathan, too.
The embodied cinematic experience is also the central magnetic force in
the 3D blockbuster-adventure Gravity. Yet, in this case the film’s capabil-
ity of allowing the spectator to ‘embody’ its cinematic universe cannot
be separated from its 3D technology. In this sense, the film is not just an
exploration of the sensation of loss of gravity in outer space; it is also an
exploration of the potentials of 3D cinema as a newly rediscovered medium.
As Stanley Cavell (1979) has argued, ‘[o]nly the art itself can discover its
possibilities, and the discovery of a new possibility is the discovery of a new
medium’ (32). The film-philosophical task in relation to the rediscovery of
3D cinema is to excavate productive concepts capable of shedding light on
how this type of cinema structures our sense of time and space: the two
dimensions of which the sensation of weightlessness imposed by Gravity
allows a potentially novel experience. The loss of connection to Earth and
the endless slip out into the vastness of space can be taken as a gesture
towards the ultimate paradox of spatio-temporality: its limitlessness. I
believe the embodied fabula could prove a productive tool for thinking
about Gravity’s bodily gesture towards this infinitude. As Deleuze (2005b)
writes, ‘[t]he attitude of the body relates thought to time as to that outside
which is infinitively further than the outside world’ (182-183).
At the same time as the film disconnects, it also stages the limitations
of mainstream narration as Ryan Stone’s (Sandra Bullock) attempt to
re-establish connection with Earth could be said to mirror the specta-
tor’s attempt to reconnect with the classical, Euclidean spatio-temporal
coordinates of the narrative experience. As such, the floating in space and
constantly failing endeavours to return safely back to the world we know
become an expression of how the cinematic experience facilitated by the
film progressively unties itself from the Euclidean space-time of the analyti-
cal fabula to provide instead a cinematic sensation of a universe devoid of
(narrative, spatio-temporal) gravity. As William Brown (2013) has argued,
analogue cinema, insofar as it is dependent on the cut, operates by spatial
fragmentation. Following Brown, this strengthens the alignment between
quotidian human perception and analogue cinema, since space in both
Conclusions 213
marks clear limitations. The ‘virtual’ camera of digital cinema, however,

is capable of entering brain-spaces, far away galaxies, it can move through
walls or buildings, and can provide us with perspectives not available to
the human eye (42-50). Although the same could be claimed for analogue
cinema, I believe this cinematic potential is particularly accentuated and
embraced by digital cinema. In this context, Brown regards digital cinema
to be ‘an inhuman or posthuman form, which posits a world in which we
can pass through solid objects as easily as we do through empty spaces’ (47).
Leaving aside the discussion of whether or not digital cinema definitely
poses an ‘inhuman’ form, it does appear to have left behind the classical
cinematic anchoring in quotidian human perception. Following this line
of reasoning, Gravity is a reflection upon how digital cinema is gradually
losing its foundation in the Euclidean spatio-temporal coordinates of clas-
sical cinema and quotidian perception. Its use of 3D technology to induce
in the viewers an experience of weightlessness reflects how digital cinema
‘allows us to transcend our limited human perception, which must fragment
and divide’ (47).
Throughout this book, I have argued that the cognitive-formalist fabula
is committed to linear temporality and Euclidean space. Both Leviathan
and Gravity support the claim that digital and 3D imagery are untying the
traditional cinematic foundation in the quotidian human experience. The
spectator’s narrative comprehension can, therefore, no longer be thought
of in terms of a disengaged cognisor involved with ‘problem-solution’
models. Nor can we think narrative comprehension as something that can
be theoretically isolated from the emotional and affective responses of the
spectator. One promising way of understanding the current transformations
in cinematic spectatorship is by developing new tools apt for the challenges
of contemporary cinema. This book has attempted to do so by challenging
the classical scientific assumption of the cognitive-formalist fabula to
propose the far more dynamic concept of the embodied fabula.
Bibliography
Allen, Richard. 2001. ‘Cognitive Film Theory’. In Wittgenstein, Theory and the Arts, edited by
Richard Allen & Malcolm Turvey, 175-210. London: Routledge.
Anderson, Joseph. 1998. The Reality of Illusion: An Ecological Approach to Cognitive Film Theory.
Carbondale, IL: Southern Illinois University Press.
Anderson, Michael L. 2003. ‘Embodied Cognition: A Field Guide’. Artificial Intelligence 149 (1):
91-130. doi:10.1016/S0004-3702(03)00054-7.
Andrew, Dudley. 2010. What Cinema Is: Bazin’s Quest and Its Charge. Malden, MA: Wiley-Blackwell.
Aristotle. 1996. Poetics. Revised. London/New York: Penguin Classics.
Aronofsky, Darren. 2001. ‘Aronofsky, Darren: Requiem for a Dream’. Urban Cinefile. February 8.
http://www.urbancinefile.com.au/home/view.asp?a=4516&s=interviews.
Ashton, Dyrk. 2008. ‘Feeling Time: Deleuze’s Time-Image and Aesthetic Effect’. Rhizomes, no.
16 (Summer). http://www.rhizomes.net/issue16/ashton/index.html.
Azcona, Maria del Mar. 2010. The Multi-Protagonist Film. 1st ed. Malden, MA; Oxford: Wiley-Blackwell.
Baguer, Ignacio Domingo. 2004. ‘What Time Is It? New Temporal Regimes in Contemporary Sci-
ence Fiction Cinema’. In Memory, Imagination and Desire in Contemporary Anglo-American
Literature and Film, edited by Constanza Río-Álvaro & Luis Miguel García-Mainar, 245251.
Heidelberg: Universitätsverlag Winter.
Barker, Jennifer M. 2009. The Tactile Eye: Touch and the Cinematic Experience. Berkeley: University
of California Press.
Barthes, Roland. 1977. Image-Music-Text. Translated by Stephen Heath. London: Fontana Press.
Bazin, André. 2005. What Is Cinema? Translated by Hugh Gray. Vol. 1. 2 vols. Berkeley/Los Angeles/
London: University of California Press.
Bell, Jeffrey A. 2006. Philosophy at the Edge of Chaos: Gilles Deleuze and the Philosophy of Differ-
ence. Toronto Studies in Philosophy. Toronto: University of Toronto Press.
Bellour, Raymond. 1998. ‘Thinking, Recounting: The Cinema of Gilles Deleuze’. Translated by
Melissa McMahon. Discourse 20 (3): 56-75.
—. 2010. ‘Deleuze: The Thinking of the Brain’. Cinema: Journal of Philosophy and the Moving
Image 1: 81-94.
Bergson, Henri. 1991. Matter and Memory. Translated by Nancy Margaret Paul & William Scott
Palmer. 8th Printing. New York: Zone Books.
—. 1998. Creative Evolution. Translated by Arthur Mitchell. Unabridged. Mineola, NY: Dover.
Berliner, Todd. 2010. Hollywood Incoherent: Narration in Seventies Cinema. Austin: University
of Texas Press.
Beugnet, Martine. 2007. Cinema and Sensation: French Film and the Art of Transgression. Car-
bondale, IL: Southern Illinois University Press.
Bianco, Jamie Skye. 2004. ‘Techno-Cinema’. Comparative Literature Studies 41 (3): 377-403.
doi:10.1353/cls.2004.0030.
Bickhard, Mark H., & Loren Terveen. 1996. Foundational Issues in Artificial Intelligence and
Cognitive Science. 1st ed. Amsterdam: Elsevier.
Boden, M.A. 2006. Mind as Machine: A History of Cognitive Science. Oxford/New York: Oxford
University Press.
Bolter, Jay David, & Richard Grusin. 2000. Remediation: Understanding New Media. Cambridge,
Mass: MIT Press.
Bondebjerg, Ib. 2015. ‘The Embodied Mind: When Biology Meets Culture and Society’. Palgrave
Communications 1 (July): 1-5. doi:10.1057/palcomms.2015.15.
Bordwell, David. 1985a. ‘Part 1: The Classical Hollywood Style: 1917-60’. In The Classical Hollywood
Cinema: Film Style and Mode of Production to 1960, edited by David Bordwell, Janet Staiger,
& Kristin Thompson. New York: Columbia University Press.
—. 1985b. Narration in the Fiction Film. Madison, WI: University of Wisconsin Press.
—. 1989. ‘A Case for Cognitivism’. IRIS 9: 11-40.
—. 1990. ‘A Case for Cognitivism: Further Reflections’. IRIS 11: 107-112.
—. 1991. Making Meaning: Inference and Rhetoric in the Interpretation of Cinema. . Cambridge,
MA: Harvard University Press.
—. 1996. ‘Contemporary Film Studies and the Vicissitudes of Grand Theory’. In Post-Theory:
Reconstructing Film Studies, edited by David Bordwell & Noël Carroll, 3-36. Madison, WI:
University of Wisconsin Press.
—. 1997. On the History of Film Style. Cambridge, MA: Harvard University Press.
—. 2002a. ‘Film Futures’. SubStance 31 (1): 88-104. doi:10.1353/sub.2002.0004.
—. 2002b. ‘Intensified Continuity: Visual Style in Contemporary American Film’. Film Quarterly,
55 (3): 16-28.
—. 2006. The Way Hollywood Tells It: Story and Style in Modern Movies. Berkeley/Los Angeles/
—. 2008. Poetics of Cinema. New York: Routledge.
—. 2010. ‘The Part-Time Cognitivist: A View from Film Studies’. Projections 4 (2): 1-18. doi:10.3167/
proj.2010.040202.
—. 2011. ‘Common Sense + Film Theory = Common-Sense Film Theory?’ David Bordwell’s Website
on Cinema. http://www.davidbordwell.net/essays/commonsense.php.
Bordwell, David, & Janet Staiger. 1985. ‘Part 7: Historical Implications of the Classical Hollywood
Cinema’. In The Classical Hollywood Cinema: Film Style and Mode of Production to 1960, edited
by David Bordwell, Janet Staiger, & Kristin Thompson, Paperback. New York: Columbia
University Press.
Bordwell, David, Janet Staiger, & Kristin Thompson. 1985. The Classical Hollywood Cinema: Film
Style and Mode of Production to 1960. New York: Columbia University Press.
Borges, Jorge Luis. 2000. ‘The Garden of Forking Paths’. In Labyrinths: Selected Stories and Other
Writings, translated by Donald A. Yates & James E. Irby. London: Penguin Books.
Brady, Terrence J. 1998. ‘Dial H for Hitchcock’. Third Millennium Entertainment. http://www.
teako170.com/dial36.html.
Branco, Patrícia Silveirinha, ed. 2012. Cinema: Journal of Philosophy and the Moving Image. Vol.
3: Embodiment and the Body. Lisbon: IFILNOVA ‒ NOVA Philosophy Institute.
Branigan, Edward. 1992. Narrative Comprehension and Film. London/New York: Routledge.
—. 2002. ‘Nearly True: Forking Plots, Forking Interpretations: A Response to David Bordwell’s
“Film Futures”’. SubStance 31 (1): 105-114. doi:10.2307/3685811.
Brockmann, Stephen. 2010. A Critical History of German Film. Rochester, NY: Camden House.
Brooks, Rodney Allen. 1999. Cambrian Intelligence: The Early History of the New AI. 1st ed.
Cambridge, MA: MIT Press.
Brown, William. 2013. Supercinema: Film-Philosophy for the Digital Age. New York/Oxford:
Berghahn Books.
—. 2014. ‘Complexity and Simplicity in Inception and Five Dedicated to Ozu’. In Hollywood Puzzle
Films, edited by Warren Buckland, 125-139. AFI Film Readers. New York/London: Routledge.
Bruner, Jerome. 1986. Actual Minds, Possible Worlds. Cambridge, MA: Harvard University Press.
—. 1991. ‘The Narrative Construction of Reality’. Critical Inquiry 18 (1): 1-21.
Bibliogr aphy 217
—. 2004. ‘Life as Narrative’. An International Quarterly 71 (3): 691-710.

Buchanan, Ian, & Claire Colebrook, eds. 2000. Deleuze and Feminist Theory. Edinburgh: Edin-
burgh University Press.
Buckland, Warren. 2003. Cognitive Semiotics of Film. Cambridge, MA: Cambridge University Press.
—. 2009a. ‘Introduction: Puzzle Plots’. In Puzzle Films: Complex Storytelling in Contemporary
Cinema, edited by Warren Buckland, 1-12. Malden, MA: Blackwell-Wiley.
—. , ed. 2009b. Puzzle Films: Complex Storytelling in Contemporary Cinema. Malden, MA:
Blackwell-Wiley.
—. , ed. 2014. Hollywood Puzzle Films. AFI Film Readers. New York/London: Routledge.
Camargo, Sandy. 2002. ‘“Mind the Gap”: The Multi-Protagonist Film Genre, Soap’. M/C: A Jounral
of Media and Culture 5 (5). http://journal.media-culture.org.au/0210/Carmago.php.
Cameron, Alan. 2008. Modular Narratives in Contemporary Cinema. Basingstoke/New York:
Palgrave Macmillan.
Campora, Matthew. 2014. Subjective Realist Cinema: From Expressionism to Inception. New York/
Oxford: Berghahn Books.
Capra, Fritjof. 1996. The Web of Life: A New Scientific Understanding of Living Systems. New York:
Anchor/Doubleday.
Caracciolo, Marco. 2011. ‘Narrative, Meaning, Interpretation: An Enactivist Approach’. Phenom-
enology and the Cognitive Sciences 11 (3): 367-384. doi:10.1007/s11097-011-9216-0.
Carroll, Noël. 1996. Theorizing the Moving Image. Cambridge/New York: Cambridge University Press.
—. 2001. Beyond Aesthetics: Philosophical Essays. Cambridge/New York: Cambridge University Press.
—. 2003. Engaging the Moving Image. Yale Series in the Philosophy and Theory of Art. New
Haven, CT: Yale University Press.
—. 2006. ‘Film, Emotion, and Genre’. In Philosophy of Film and Motion Pictures: An Anthology,
edited by Noël Carroll & Jinhee Choi, 217-233. Malden, MA/Oxford: Blackwell Publishers.
—. 2008. The Philosophy of Motion Pictures. Malden, MA/Oxford: Blackwell Publishers.
Caruth, Cathy. 1995. ‘II. Recapturing the Past: Introduction’. In Trauma: Explorations in Memory,
edited by Cathy Caruth, 151-157. Baltimore, MD/London: John Hopkins University Press.
Casetti, Francesco, & Luciane Bohne. 1986. ‘Antonioni and Hitchcock: Two Strategies of Narrative
Investment’. SubStance 15 (3) (51): 69-86.
Cavell, Stanley. 1979. The World Viewed: Reflections on the Ontology of Film. Cambridge, MA/
London: Harvard University Press.
Celeste, Reni. 2005. ‘Love and Catastrophe: Filming the Sublime in Hiroshima Mon Amour’.
Film-Philosophy 9 (4). http://www.film-philosophy.com/index.php/f-p/article/view/851.
Chatman, Seymour. 1978. Story and Discourse: Narrative Structure in Fiction and Film. Ithaca,
NY/London: Cornell University Press.
Chemero, Anthony. 2009. Radical Embodied Cognitive Science. Cambridge, MA/London: MIT Press.
Chion, Michel. 1994. Audio-Vision: Sound on Screen. Translated by Claudia Corbman. New York:
Columbia University Press.
—. 2009. Film, a Sound Art. New York: Columbia University Press.
Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press.
Clark, Andy. 1998. Being There: Putting Brain, Body, and World Together Again. Cambridge, MA:
MIT Press.
—. 2000. Mindware: An Introduction to the Philosophy of Cognitive Science. Oxford/New York:
Oxford University Press.
—. 2010. ‘Memento’s Revenge: The Extended Mind, Extended’. In The Extended Mind, edited by
Richard Menary, 43-66. Cambridge, MA/London: MIT Press.
Clarke, Melissa. 2002. ‘The Space-Time Image: The Case of Bergson, Deleuze, and Memento’. The
Journal of Speculative Philosophy 16 (3): 167-181. doi:10.1353/jsp.2003.0003.
Clough, Patricia Ticineto, & Jean Halley, eds. 2007. The Affective Turn: Theorizing the Social. 1st
ed. Durham, NC: Duke University Press.
Coëgnarts, Maarten, ed. 2014. ‘Metaphor, Bodily Meaning, and Cinema’. Image [&] Narrative
15 (1). http://www.imageandnarrative.be/index.php/imagenarrative/issue/view/37.
Coëgnarts, Maarten, & Peter Kravanja. 2012. ‘Embodied Visual Meaning: Image Schemas in
Film’. Projections 6 (2): 84-101. doi:10.3167/proj.2012.060206.
—. , eds. 2015. Embodied Cognition and Cinema. Ithaca, NY: Cornell University Press.
Colebrook, Claire. 2002. Gilles Deleuze. Routledge Critical Thinkers. London/New York: Routledge.
Connolly, William. 2002. Neuropolitics: Thinking, Culture, Speed. Minneapolis, MN/London:
University of Minnesota Press.
Culler, Jonathan. 1986. On Deconstructivism: Theory and Criticism after Structuralism.. Ithaca,
NY: Cornell University Press.
Cummins, Richard. 1991. Meaning and Mental Representation. Cambridge, MA: MIT Press.
—. 1996. Representations, Targets, and Attitudes. Cambridge, MA: MIT Press.
Currie, Gregory. 1995. Image and Mind: Film, Philosophy, and Cognitive Science. Cambridge, MA/
London: Cambridge University Press.
Curtis, James M. 1976. ‘Bergson and Russian Formalism’. Comparative Literature 28 (2): 109.
doi:10.2307/1769653.
Curtis, Quentin. 1995. ‘The Thrill of The Usual Suspects Is That It Re-Mythologises the Crime
Movie’. The Independent, August 27. http://www.independent.co.uk/life-style/the-critics-
cinema-confused-you-will-be-1598236.html.
d’Allonnes, Fabrice Reveault. 1994. Pour Le Cinéma ‘Moderne’: Du Lien de L’art Au Monde. De
Parti Pris. Liège, Belgium: Editions Yellow Now.
Damasio, Antonio. 1995. Descartes’ Error: Emotion, Reason, and the Human Brain. New York:
Penguin Books.
—. 1999. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. London:
Vintage.
—. 2003. Looking for Spinoza: Joy, Sorrow, and the Feeling Brain. London: William Heinemann.
—. 2011. Self Comes to Mind: Constructing the Conscious Brain. New York: Random House.
DeLanda, Manuel. 2002. Intensive Science and Virtual Philosophy. London/New York: Continuum.
Deleuze, Gilles. 1991. Bergsonism. Translated by Hugh Tomlinson & Barbara Habberjam. New
York: Zone Books.
—. 1995. Negotiations 1972-1990. Translated by Martin Joughin. New York: Columbia University Press.
—. 2000. ‘The Brain Is the Screen: An Interview with Gilles Deleuze’. In The Brain Is the Screen:
Deleuze and the Philosophy of Cinema, edited by Gregory Flaxman, translated by Marie
Therese Guirgis, 365-374. Minneapolis, MN: University of Minnesota Press.
—. 2001. ‘Review of Gilbert Simondon’s L’individu et Sa Genese Physico-Biologique (1966)’.
Translated by Ivan Ramirez. Pli: The Warwick Journal of Philosophy 12: 43-49.
—. 2004a. Difference and Repetition. Translated by Paul Patton. London: Continuum.
—. 2004b. Logic of Sense. Translated by Michael Lester. London: Continuum.
—. 2005a. Cinema 1: The Movement-Image. Translated by Hugh Tomlinson & Barbara Habberjam.
London: Continuum.
—. 2005b. Cinema 2: The Time-Image. Translated by Hugh Tomlinson & Barbara Habberjam.
London: Continuum.
Bibliogr aphy 219
Deleuze, Gilles, & Felix Guattari. 1987. A Thousand Plateaus: Capitalism and Schizophrenia.
Translated by Brian Massumi. Minneapolis, MN: University of Minnesota Press.
del Rio, Elena. 2008. Deleuze and the Cinemas of Performance: Powers of Affection. Edinburgh:
Edinburgh University Press.
—. 2011. ‘Cinema’s Exhaustion and the Vitality of Affect’. In Media Res. August 29. http://medi-
acommons.futureofthebook.org/imr/2011/08/29/cinemas-exhaustion-and-vitality-affect.
Dennett, Daniel C. 1991. Consciousness Explained. Boston, MA: Little, Brown, and Company.
Doane, Mary Ann. 2002. The Emergence of Cinematic Time: Modernity, Contingency, the Archive.
Cambridge, MA/London: Harvard University Press.
Domarchi, Jean, Jacques Doniol-Valcroze, Jean Luc Godard, Pierre Kast, Eric Rohmer, & Jacques
Rivette. 1985. ‘Hiroshima, Notre Amour’. In Cahiers Du Cinéma, the 1950s: Neo-Realism,
Hollywood, New Wave, edited by Jim Hillier, translated by Liz Heron, 1: 59-70. Cambridge,
MA: Harvard University Press.
Donecker, Simone. 2005. Narrative Constructions in Tom Tykwer’s Run, Lola, Run. Munich: GRIN
Verlag.
Dreyfus, Hubert L. 1991. Being-in-the-World: A Commentary on Heidegger’s Being and Time,
Division I. Cambridge, MA: MIT Press.
Duncan, Seth, & Lisa Feldman Barrett. 2007. ‘Affect Is a Form of Cognition: A Neurobiological
Analysis’. Cognition & Emotion 21 (6): 1184-1211. doi:10.1080/02699930701437931.
Duras, Marguerite. 1961. Hiroshima Mon Amour. Translated by Richard Seaver. New York: Grove
Press.
Elliott, Paul. 2010. ‘The Eye, the Brain, the Screen: What Neuroscience Can Teach Film Theory’. Excur-
sions 1 (1). http://www.excursions-journal.org.uk/index.php/excursions/article/viewArticle/2.
—. 2011. Hitchcock and the Cinema of Sensations: Embodied Film Theory and Cinematic Reception.
London/New York: I.B. Tauris & Co.
Elsaesser, Thomas. 2004a. ‘The New Film History as Media Archaeology’. Cinémas: Journal of
Film Studies 14 (2-3): 75-117.
—. 2004b. ‘Was Wäre, Wenn Du Schon Tot Bist? Vom “Postmodernen” Zum “Postmortem” Kino Am
Beispiel von Christopher Nolans Memento’. In Zeitsprünge. Wie Filme Geschichte(n) Erzählen,
edited by Christine Rüffert, Karl-Heinz Schmid, & Alan Tews, 13-41. Berlin: Bertz + Fischer.
—. 2009. ‘The Mind-Game Film’. In Puzzle Films: Complex Storytelling in Contemporary Cinema,
edited by Warren Buckland, 13-41. Malden, MA: Blackwell-Wiley.
Elsaesser, Thomas, & Malte Hagener. 2009. Film Theory: An Introduction through the Senses. 1st
edition. New York: Routledge.
Elsaesser, Thomas, Alexander Horwath, & Noel King, eds. 2004. The Last Great American
Picture Show: New Hollywood Cinema in the 1970s. Film Culture in Transition. Amsterdam:
Amsterdam University Press.
Engell, Lorenz. 2005. Bilder der Endlichkeit. Weimar: VDG Weimar.
Engell, Lorenz, & Oliver Fahle, eds. 1997. Cinéma Selon Deleuze. Weimar: Verlag der
Bauhaus-Universität.
Evans, Owen. 2004. ‘Tom Tykwer’s Run Lola Run: Postmodern, Posthuman or “Post-Theory”?’
Studies in European Cinema 1 (2): 105-115.
Fahlenbrach, Kathrin. 2016. Embodied Metaphors in Film, Television, and Video Games: Cognitive
Approaches. Routledge Research in Cultural and Media Studies. London/New York: Routledge.
Fahle, Oliver. 2005. Bilder der Zweiten Moderne. Weimar: VDG Weimar.
Feagin, Susan. 1999. ‘Time and Time’. In Passionate Views: Film, Cognition, and Emotion, edited
by Carl Plantinga & Greg M. Smith, 168-179. Baltimore, MD:Johns Hopkins University Press.
Flaxman, Gregory, ed. 2000. The Brain Is the Screen: Deleuze and the Philosophy of Cinema.
Minneapolis, MN: University of Minnesota Press.
Flinn, Caryl. 2004. ‘The Music That Lola Ran to’. In Sound Matters: Essays on the Acoustics of Modern
German Culture, edited by Lutz Koepnick & Nora Alter, 197-213. New York: Berghahn Press.
Fludernik, Monika. 2010. Towards a ‘Natural’ Narratology. London: Taylor & Francis.
Fodor, Jerry A. 1983. The Modularity of Mind: An Essay on Faculty Psychology. Cambridge, MA:
MIT Press.
Frampton, Daniel. 2006. Filmosophy. London/New York: Wallflower Press.
French, Sarah. 2008. ‘From History to Memory: Alain Resnais’ and Marguerite Duras’ Hiroshima
Mon Amour’. Melbourne Art Journal 3: 1-13.
Friedberg, Anne. 1993. Window Shopping: Cinema and the Postmodern. Berkeley, CA: University
of California Press.
Gallagher, Shaun. 2015. ‘Invasion of the Body Snatchers: How Embodied Cognition Is Being
Disembodied’. The Philosopher’s Magazine, April: 96-102.
Gallagher, Shaun, & Dan Zahavi. 2012. The Phenomenological Mind. 2nd edition. London/New
York: Routledge.
Gallese, Vittorio, Luciano Fadiga, Leonardo Fogassi, & Giacomo Rizzolatti. 1996. ‘Action Recogni-
tion in the Premotor Cortex’. Brain 119 (2): 593-609. doi:10.1093/brain/119.2.593.
Gallese, Vittorio, & Michele Guerra. 2012. ‘Embodying Movies: Embodied Simulation and Film
Studies’. Cinema: Journal of Philosophy and the Moving Image 3: 183-210.
Gardner, Howard. 1985. The Mind’s New Science: A History of the Cognitive Revolution. New York:
Basic Books.
Gargett, Adrian. 2002. ‘Nolan’s Memento, Memory, and Recognition’. CLCWeb: Comparative
Literature and Culture 4 (3): 3.
Ghislotti, Stefano. 2003. ‘Backwards: Memory and Fabula Construction in Memento by Chris-
topher Nolan’. Film Anthology. Internet Review of Film and Cinema. http://www.unibg.it/fa.
—. 2009. ‘Narrative Comprehension Made Diff icult: Film Form and Mnemonic Devices in
Memento’. In Puzzle Films: Complex Storytelling in Contemporary Cinema, edited by Warren
Buckland, 87-106. Malden, MA: Wiley-Blackwell.
Gifford, Barry. 1997. ‘Interview: “Lost Highway Screenwriter Barry Gifford”’. Film Threat. http://
www.lynchnet.com/lh/lhgifford.html.
Gilbert, Jeremy. 2004. ‘Signifying Nothing: “Culture,” “Discourse” and the Sociality of Affect’.
Culture Machine 6 (0). http://www.culturemachine.net/index.php/cm/article/view/8.
Gillespie, Michael P. 2006. ‘The Chaos and Complexity of Classic Hollywood Cinema’. Non-linear
Dynamics, Psychology, and Life Sciences 10 (1): 123-141.
Glasersfeld, Ernst von. 1988. ‘The Reluctance to Change a Way of Thinking’. Irish Journal of
Psychology 9 (1): 83-90.
—. 1991. ‘Knowing without Metaphysics: Aspects of the Radical Constructivist Position’. In
Research and Reflexivity, by Frederick Steier. London: SAGE Publications.
Gleick, James. 1998. Chaos: Making a New Science. London: Vintage.
Goldsmith, Leo. 2013. ‘The Violence of Sensation’. Reverse Shot 33 (New Releases). http://reverse-
shot.org/reviews/entry/1682/leviathan.
Goodwin, Brian. 1994. How the Leopard Changed Its Spots: The Evolution of Complexity. New
York: C. Scribner’s Sons.
Bibliogr aphy 221
Gregg, Melissa, & Gregory J. Seigworth, eds. 2010. The Affect Theory Reader. Durham, NC: Duke
University Press Books.
Grodal, Torben. 1999. Moving Pictures: A New Theory of Film Genres, Feelings, and Cognition.
Oxford/New York: Oxford University Press.
—. 2009. Embodied Visions: Evolution, Emotion, Culture, and Film. Oxford/New York: Oxford
University Press.
Gunning, Tom. 1986. ‘The Cinema of Attraction: Early Film, Its Spectator and the Avant-Garde’.
Wide Angle 8 (3-4): 63-70.
Harbord, Janet. 2007. The Evolution of Film: Rethinking Film Studies. Cambridge, UK/Malden,
MA: Polity.
Hardt, Michael, & Antonio Negri. 2000. Empire. Cambridge, MA/London: Harvard University Press.
—. 2004. Multitude: War and Democracy in the Age of Empire. New York: The Penguin Press.
Hayles, Kathrine. 1990. Chaos Bound: Orderly Disorder in Contemporary Literature and Science.
Ithaca, N.Y: Cornell University Press.
—. 2004. ‘Refiguring the Posthuman’. Comparative Literature Studies 41 (3): 311-316.
Hesselberth, Pepita, & Laura Schuster. 2008. ‘Into the Mind and Out to the World: Memory
Anxiety in the Mind-Game Film’. In Mind the Screen: Media Concepts According to Thomas
Elsaesser, edited by Jaap Kooijman, Patricia Pisters, & Wanda Strauven, 96-111. Amsterdam:
Amsterdam University Press.
Hogan, Patrick Colm. 2012. Affective Narratology: The Emotional Structure of Stories. Lincoln,
NE: University of Nebraska Press.
Holtmeier, Matthew A. 2014. ‘The Wanderings of Jia Zhangke: Pre-Hodological Space and Aimless
Youths in Xiao Wu and Unknown Pleasures’. Journal of Chinese Cinemas 8 (2): 148-159. doi:
10.1080/17508061.2014.915565.
Hurley, Susan. 2002. Consciousness in Action. 1st edition. Cambridge, MA: Harvard University Press.
—. 2008. ‘The Shared Circuits Model (SCM): How Control, Mirroring, and Simulation Can
Enable Imitation, Deliberation, and Mindreading’. Behavioral and Brain Sciences 31: 1-58.
doi:10.1017/S0140525X07003123.
Hven, Steffen. 2010. Facing the Lost Highway: Three Approaches to the Cinematic Encounter.
Aarhus: Aarhus University. http://library.au.dk/fileadmin/www.aestetik.au.dk/uddannelse/
specialer/DRA_steffen_hven_speciale_udlaan_www.pdf.
—. 2014. ‘Narrative Is Not in Us; but It Is We Who Move in a Being-Narrative, a World-Narrative’.
Edited by Susana Viegas & Maria Teresa Teixeira. International Conference on Philosophy
and Film eProceedings 1: 260-277.
—. 2015. ‘Memento and the “Embodied” Fabula: Narrative Comprehension Revisited’. Acta
Universitatis Sapientiae. Film and Media Studies 11 (2015): 93-110.
Iacoboni, Marco. 2008. Mirroring People: The New Science of How We Connect with Others. New
York: Farrar, Straus & Giroux.
Iacoboni, Marco, & Roger W. McHaney. 2009. ‘Applying Empathy and Mirror Neuron Concepts
to NeuroLeadership’. NeuroLeadership Journal, no. 2: 1-7.
Ingold, Tim. 2000. The Perception of the Environment: Essays on Livelihood, Dwelling and Skill.
London/New York: Routledge.
Johnson, Mark. 1990. The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason.
Chicago, IL: University of Chicago Press.
Johnson, Steven. 2001. Emergence: The Connected Lives of Ants, Brains, Cities, and Software. New
York: Scribner.
Jones, Kent. 2015. ‘Hiroshima Mon Amour: Time Indef inite’. Criterion Essays. https://www.
criterion.com/current/posts/291-hiroshima-mon-amour-time-indefinite.
Jullier, Laurent. 2010. ‘Should I See What I Believe? Audiovisual Ostranenie and Evolutionary-
Cognitive Film Theory’. In On ‘Strangeness’ and the Moving Image: The History, Reception, and
Relevance of a Concept, edited by Anna van den Oever, 1: 119-140. Amsterdam: Amsterdam
University Press.
Kallay, Jasmina. 2013. Gaming Film: How Games Are Reshaping Contemporary Cinema. Basing-
stoke: Palgrave Macmillan.
Keating, Patrick. 2006. ‘Emotional Curves and Linear Narratives’. The Velvet Light Trap 58 (1):
4-15. doi:10.1353/vlt.2006.0029.
Kennedy, Barbara. 2000. Deleuze and Cinema: Deleuze and the Aesthetic of Sensation. Edinburgh:
Edinburgh University Press.
Kessler, Frank. 2010. ‘Ostranenie, Innovation, and Media History’. In On ‘Strangeness’ and the
Moving Image: The History, Reception, and Relevance of a Concept, edited by Anna van den
Oever, 1: 61-79. Amsterdam: Amsterdam University Press.
Kinder, Marsha. 2002. ‘Hot Spots, Avatars, and Narrative Fields Forever: Buñuel’s Legacy for New
Digital Media and Interactive Database Narrative.’ Film Quarterly 55 (4): 2-15. doi:10.1525/
fq.2002.55.4.2.
King, Geoff. 2002. New Hollywood Cinema. New York: Columbia University Press.
Kiss, Miklós. 2010. ‘The Perception of Reality as Deformed Realism’. In On ‘Strangeness’ and the
Oever, 1: 165-172. Amsterdam: Amsterdam University Press.
Klein, Andy. 2001. ‘Everything You Wanted to Know about “Memento”’. Salon, June.
Kovács, András Bálint. 2000. ‘The Film History of Thought’. In The Brain Is the Screen: Deleuze
and the Philosophy of Cinema., edited by Gregory Flaxman, translated by Sandor Hervey,
153-170. Minneapolis, MN: University of Minnesota Press.
—. 2007. Screening Modernism: European Art Cinema, 1950-1980. Chicago, IL: University of
Chicago Press.
Kozloff, Sarah. 1988. Invisible Storytellers: Voice-over Narration in American Fiction Film. Berkeley/
Los Angeles/London: University of California Press.
Kreidl, John F. 1978. Alain Resnais. 6th ed. Boston, MA: Twayne Publishers.
Kretzschmar, Laurent. 2002. ‘Is Cinema Renewing Itself?’ Film-Philosophy 6 (15). http://www.
film-philosophy.com/vol6-2002/n15kretzschmar.
Kruglanski, Arie W. 2004. The Psychology of Closed Mindedness. New York: Taylor & Francis.
Lakoff, George. 1987. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind.
Chicago, IL/London: University of Chicago Press.
Lakoff, George, & Mark Johnson. 1999. Philosophy in the Flesh: The Embodied Mind and Its
Challenge to Western Thought. New York: Basic Books.
—. 2003. Metaphors We Live By. Chicago, IL/London: University of Chicago Press.
Laplace, Pierre Simon. 1951. A Philosophical Essay on Probabilities. Translated by Frederick Wilson
Truscott & Frederick Lincoln Emory. 1st ed. New York/London: John Wiley & Sons. https://
archive.org/stream/philosophicaless00lapliala#page/n5/mode/2up.
Bibliogr aphy 223
Lawlor, Leonard, & Valentine Moulard Leonard. 2013. ‘Henri Bergson’. Edited by Edward N.
Zalta. The Stanford Encyclopedia of Philosophy. http://plato.stanford.edu/archives/win2013/
entries/bergson/.
Lazarus, Richard S. 1984. ‘On the Primacy of Cognition’. American Psychologist 39 (2): 124-129.
Lefebvre, Henri. 1991. The Production of Space. Translated by Donald Nicholson-Smith. Hoboken,
NJ: Wiley-Blackwell.
—. 1992. Critique of Everyday Life. Translated by John Moore. Vol. 1. 2 vols. New York: Verso.
—. 2002. Critique of Everyday Life. Translated by John Moore. Vol. 2. 2 vols. New York: Verso.
—. 2004. Rhythmanalysis: Space, Time and Everyday Life. Translated by Gerald Moore & Stuart
Elden. London: Continuum.
Lemmen, Ronald. 1997. ‘Towards a Non-Cartesian Cognitive Science : In the Light of the Phi-
losophy of Merleau-Ponty.’ Ph.D. Dissertation, University of Sussex.
Leys, Ruth. 2011. ‘A Turn to Affect: A Critique’. Critical Inquiry 37 (Spring): 434-472.
Lorenz, Edward. 1995. ‘Predictability: Does the Flap of a Butterfly’s Wings in Brazil Set off a
Tornado in Texas?’ Essence of Chaos 139th Annual Meeting of the American Association for
the Advancement of Science (29 Dec 1972): Appendix 1.
MacDonald, Scott. 2012. ‘Leviathan’. Framework: The Journal of Cinema & Media. New York
Film Festival 2012.
Marks, Laura U. 2000. The Skin of Film: Intercultural Cinema, Embodiment, and the Senses.
Durham, NC/London: Duke University Press.
—. 2002. Touch: Sensuous Theory and Multisensory Media. Minneapolis, MN/London: University
of Minnesota Press.
Martin-Jones, David. 2006. Deleuze, Cinema and National Identity: Narrative Time in National
Contexts. Edinburgh: Edinburgh University Press.
—. 2011. Deleuze and World Cinemas. London/New York: Bloomsbury Academic.
Massumi, Brian. 2002. Parables for the Virtual: Movement, Affect, Sensation. Durham, NC/London:
Duke University Press.
McCullers, Carson. 2012. The Mortgaged Heart. London: Penguin Books.
McLuhan, Marshall. 1994. Understanding Media: The Extensions of Man. 1st MIT Press Edition.
Cambridge, MA/London: MIT Press.
Merleau-Ponty, Maurice. 2002. Phenomenology of Perception [Phénomènologie de La Perception].
Translated by Colin Smith. London/New York: Routledge.
Mitchell, David. 2002. ‘Twisted Tales: Cognitivism and Narrative Distortion’. Zenoshrdlu. http://
www.zenoshrdlu.com/zenotwist.htm.
Mittell, Jason. 2015. Complex TV: The Poetics of Contemporary Television Storytelling. New York:
New York University Press.
Molloy, Claire. 2010. Memento. Edinburgh: Edinburgh University Press.
Moulard-Leonard, Valentine. 2008. Bergson-Deleuze Encounters: Transcendental Experience
and the Thought of the Virtual. SUNY Series in Contemporary French Thought. Albany, NY:
State University of New York Press.
Mulhall, Stephen. 2008. On Film. 2nd ed. London/New York: Routledge.
Mullarkey, John. 2009. Refractions of Reality: Philosophy and the Moving Image. Basingstoke/
New York: Palgrave Macmillan.
Murray, John P., Mario Liotti, Paul T. Ingmundson, Helen S. Mayberg, Yonglin Pu, Frank Zamar-
ripa, Yijun Liu, Marty G. Waldorff, Jia-Hong Gao, & Peter T. Fox. 2008. ‘Children’s Brain
Activations While Viewing Televised Violence Revealed by fMRI’. Media Psychology 8 (1): 25-37.
Nadel, Alan. 2005. ‘Second Nature, Cinematic Narrative, the Historical Subject, and Russian
Ark’. In A Companion to Narrative Theory, edited by James Phelan & Peter J. Rabinowitz,
427-440. Oxford: Blackwell Publishers.
Nannicelli, Ted, & Paul Taberham. 2014. ‘Introduction’. In Evolutionary Film Theory, edited by
Ted Nannicelli & Paul Taberham, 1-23. New York/London: Routledge.
Neale, Steve. 1981. ‘Art Cinema as Institution’. Screen 22 (1, Spring): 11-39.
Neupert, Richard John. 1995. The End: Narration and Closure in the Cinema. Detroit, MI: Wayne
State University Press.
Newell, Allen, & Herbert Alexander Simon. 1972. Human Problem Solving. Upper Saddle River,
NJ: Prentice-Hall.
Newman, Michael Z. 2006. ‘Character and Complexity in American Independent Cinema: 21
Grams and Passion Fish’. Film Criticism 31 (1-2): 89-106.
Noë, Alva. 2004. Action in Perception. Representation and Mind. Cambridge, MA/London: The
MIT Press.
Nolan, Christopher. 2012. ‘Interview: Christopher Nolan: A Visionary’s Memento Pt. 1’. Soothsayer. https://
moviesoothsayer.wordpress.com/2012/06/27/christopher-nolan-a-visionarys-memento-pt-1/.
Núñez, Rafael. 1999. ‘Could the Future Taste Purple? Reclaiming Mind, Body and Cognition’. In
Reclaiming Cognition: The Primacy of Action, Intention and Emotion, edited by Rafael Núñez
& Walter J Freeman. Exeter: Imprint Academic.
Oever, Anna van den. 2010. ‘Introduction: Ostran(n)enie as an “Attractive” Concept’. In On

‘Strangeness’ and the Moving Image: The History, Reception, and Relevance of a Concept, edited
by Anna van den Oever, 1: 11-18. Amsterdam: Amsterdam University Press.
Panek, Elliott. 2006. ‘The Poet and the Detective: Defining the Psychological Puzzle Film’. Film
Criticism 31 (1-2 (Special Double Issue: Complex Narratives)): 62-88.
Patton, Paul. 1996. ‘Introduction’. In Deleuze: A Critical Reader, edited by Paul Patton, 1-17. Oxford/
Cambridge, MA: Blackwell Publishers.
Persson, Per. 2000. Understanding Cinema: Constructivism and Spectator Psychology. PhD dis-
sertation. Stockholm University.
Pfeifer, Rolf, & Josh Bongard. 2007. How the Body Shapes the Way We Think. Cambridge, MA/
London: MIT Press.
Pickover, Clifford A. 2007. The Möbius Strip: Dr. August Möbius’s Marvelous Band in Mathematics,
Games, Literature, Art, Technology, and Cosmology. New York: Basic Books.
Pisters, Patricia. 1998. ‘From Mouse to Mouse – Overcoming Information’. Enculturation 2 (1).
—. 2011a. ‘The Mosaic Film: Nomadic Style and Politics in Transnational Media Culture’.
Thamyris/Intersecting: Place, Sex and Race 23 (1): 175-190.
—. 2011b. ‘Flashforward: The Future Is Now’. Deleuze Studies 5 (November): 98-115. doi:10.3366/
dls.2011.0039.
—. 2012. The Neuro-Image: A Deleuzian Film-Philosophy of Digital Screen Culture. Stanford, CA:
Stanford University Press.
Plantinga, Carl. 1997. ‘Notes on Spectator Emotion and Ideological Film Criticism’. In Film Theory and
Philosophy, edited by Richard Allen & Murray Smith, 372-390. Oxford: Oxford University Press.
—. 2001. ‘Movie Pleasures and the Spectator’s Experience: Toward a Cognitive Approach’. Film
and Philosophy 11 (4 (originally published in 1995)). http://www.angelfire.com/film/eliab/
visceral.html.
—. 2002. ‘Cognitive Film Theory: An Insider’s Appraisal’. Cinémas Revue D’études cinéma-
tographiques/Cinémas Journal of Film Studies 12 (2): 15-37.
Bibliogr aphy 225
—. 2009a. ‘Emotion and Affect’. In The Routledge Companion to Philosophy and Film, edited by
Carl Plantinga & Paisley Livingstone. London/New York: Routledge.
—. 2009b. Moving Viewers: American Film and the Spectator’s Experience. Berkeley/Los Angeles/
Plantinga, Carl, & Greg M. Smith. 1999a. ‘Introduction’. In Passionate Views: Film, Cognition,
and Emotion, edited by Carl Plantinga & Greg M. Smith. Baltimore, MD: Johns Hopkins
University Press.
—. , eds. 1999b. Passionate Views: Film, Cognition, and Emotion. Baltimore, MD: Johns Hopkins
University Press.
Polkinghorne, Donald. 1988. Narrative Knowing and the Human Sciences. Albany, NY: State
University of New York Press.
Poulaki, Maria. 2011. Before or beyond Narrative? Towards a Complex Systems Theory of Contem-
porary Films. PhD Dissertation. Amsterdam: Rozenberg Publisher.
—. 2014a. ‘Puzzled Hollywood and the Return of Complex Films’. In Hollywood Puzzle Films,
edited by Warren Buckland, 35-53. AFI Film Readers. New York/London: Routledge.
—. 2014b. ‘Network Films and Complex Causality’. Screen 55 (3): 379-395. doi:10.1093/screen/hju020.
Powell, Anna. 2006. Deleuze and Horror Film. Edinburgh: Edinburgh University Press.
—. 2007. Deleuze, Altered States and Film. Edinburgh: Edinburgh University Press.
—. 2009. ‘Jack the Ripper’s Bodies-without-Organs: Affect and Psychogeography under the
Scalpel in From Hell’. The Irish Journal of Gothic and Horror Studies 7.
‘Production Notes’. 2003. Focus Features. http://www.21-grams.com/index.php?c=2.
Protevi, John. 2010. ‘Adding Deleuze to the Mix’. Phenomenology and the Cognitive Sciences
9 (3): 417- 436.
Pylyshyn, Zenon Walter. 1986. Computation and Cognition: Toward a Foundation for Cognitive
Science. Cambridge, MA: MIT Press.
Rajchman, John. 2010. ‘Deleuze’s Time, or How the Cinematic Changes Our Idea of Art’. In
Afterimages of Gilles Deleuze’s Film-philosophy, edited by David Norman Rodowick, 283-306.
Minneapolis, MN/London: University of Minnesota Press.
Ramachandran, Vilayanur. 2000. ‘Mirror Neurons and Imitation Learning as the Driving Force
behind the Great Leap Forward in Human Evolution’. Edge. http://edge.org/conversation/
mirror-neurons-and-imitation-learning-as-the-driving-force-behind-the-great-leap-
forward-in-human-evolution.
Rancière, Jacques. 2006. Film Fables. Translated by Emilliano Battista. Oxford/New York:
Bloomsbury Academic.
Redfern, Nick. 2004. ‘Communication and Meaning in the Cinema’. Constructivism in the Human
Sciences 9 (2): 39-48.
—. 2005. ‘Film as Text: Radical Constructivism and the Problem of Narrative in Cinema’.
Amsterdam International Electronic Journal for Cultural Narratology (AJCN) 2 (Autumn).
http://cf.hum.uva.nl/narratology/a05_redfern.html.
Renner, Karen. 2006. ‘Repeat Viewings Revisited: Emotions, Memory, and Memento’. Film
Studies 8: 106-115.
Resnais, Alain. 2008. ‘Resnais Is Not An Auteur’. Interview with Alain Resnais excerpted
from the French TV programme ‘Cinepanorama’, 1961. https://stipeticf ilms.wordpress.
com/2008/07/29/resnais-is-not-an-auteur/.
Richter, David H. 2005. ‘Your Cheatin’ Art: Double Dealing in Cinematic Narrative’. Narrative
13 (1): 11-28. doi:10.1353/nar.2005.0006.
Ricœur, Paul. 1990. Time and Narrative, Volume 1-3. Translated by Kathleen McLaughlin & David
Pellauer. Revised. Chicago, IL: University of Chicago Press.
Rizzolatti, Giacomo, & Laila Craighero. 2004. ‘The Mirror-Neuron System’. Annual Review of
Neuroscience 27 (1): 169-192. doi:10.1146/annurev.neuro.27.070203.144230.
Rizzolatti, Giacomo, & Corrado Sinigaglia. 2008. Mirrors in the Brain: How Our Minds Share Actions
and Emotions. Translated by Frances Anderson. Oxford/New York: Oxford University Press.
Rodowick, David Norman. 2001. Reading the Figural, Or, Philosophy after the New Media. Durham,
NC: Duke University Press.
—. 2003. Gilles Deleuze’s Time Machine. Durham, NC: Duke University Press.
—. 2007. The Virtual Life of Film. Cambridge, MA: Harvard University Press.
Ropars, Marie-Claire. 2007. ‘A Cinematic Language’. Rouge. Reprint from Esprit 1960. http://
www.rouge.com.au/11/cinematic_language.html.
Roth, Michael S. 1995. ‘Chapter 5: Hiroshima Mon Amour: You Must Remember This’. In Revision-
ing History: Film and the Construction of a New Past, edited by Robert A. Rosenstone, 91-101.
Princeton, NJ: Princeton University Press.
Rushton, Richard. 2009. ‘Deleuzian Spectatorship’. Screen 50 (1): 45-53. doi:10.1093/screen/hjn086.
Searle, John. 1984. Minds, Brains and Science. London: Penguin.

‘Sensory Ethnography Lab: Harvard University’. 2010. http://sel.fas.harvard.edu/.
Shapiro, Lawrence A. 2011. Embodied Cognition. New York: Routledge.
Shaviro, Steven. 1993. The Cinematic Body. Vol. 2. Theory out of Bounds. Minneapolis, MN/
London: University of Minnesota Press.
—. 2010. ‘Post-Cinematic Affect: On Grace Jones, Boarding Gate and Southland Tales’. Film-
Philosophy 14.1: 1-102.
—. 2012. ‘Post-Continuity: Full Text of My Talk’. The Pinocchio Theory. http://www.shaviro.
com/Blog/?p=1034.
—. 2013. ‘Presentation on Spring Breakers: Slide “Comparison” presented at the International
Colloquium on Philosophy and Cinema, Lisbon, 6 December 2013. http://www.shaviro.com/
Presentations/Spring/#/comparison.
Shklovsky, Viktor. 1965. ‘Art as Technique’. In Russian Formalist Criticism: Four Essays, edited by
Lee T. Lemon & Marion J. Reis, 5-24. Lincoln, NE: University of Nebraska Press.
Shotter, John. 1993. Conversational Realities Constructing Life through Language. London/
Thousand Oaks, CA: SAGE Publications Ltd.
Shouse, Eric. 2005. ‘Feeling, Emotion, Affect’. M/C: A Journal of Media and Culture 8 (6). http://
journal.media-culture.org.au/0512/03-shouse.php.
Simons, Jan. 2007. Playing the Waves: Lars von Trier’s Game Cinema. Amsterdam: Amsterdam
University Press.
—. 2008. ‘Complex Narratives’. New Review of Film and Television Studies 6 (2): 111-126.
doi:10.1080/17400300802098263.
Sinnerbrink, Robert. 2008. ‘Time, Affect, and the Brain: Deleuze’s Cinematic Aesthetics’. Film-
Philosophy 12 (1): 85-96.
—. 2011. New Philosophies of Film: Thinking Images. London/New York: Continuum.
Smith, Greg M. 2003. Film Structure and the Emotion System. Cambridge: Cambridge University Press.
Smith, Murray. 1995. Engaging Characters: Fiction, Emotion, and the Cinema. Oxford: Oxford
University Press.
—. 1999. ‘Gangsters, Cannibals, Aesthetes, or Apparently Perverse Allegiances’. In Passionate
Views: Film, Cognition, and Emotion, edited by Carl Plantinga & Greg M. Smith, 217-238.
Baltimore, MD: Johns Hopkins University Press.
Bibliogr aphy 227
—. 2011. ‘Empathy, Expansionism, and the Extended Mind’. In Empathy: Philosophical and
Psychological Perspectives, edited by Peter Goldie & Amy Coplan, 99-117. Oxford/New York:
Oxford University Press.
Smith, Richard. 2001. ‘The Philosopher with Two Brains’ 5 (No. 34: Deleuze Special Issue). http://
www.film-philosophy.com/index.php/f-p/article/view/655/568.
Smith, Tim J., Daniel Levin, & James E. Cutting. 2012. ‘A Window on Reality: Perceiving Edited Moving
Images’. Current Directions in Psychological Science 21 (2): 107-113. doi:10.1177/0963721412437407.
Sobchack, Vivian Carol. 1992. The Address of the Eye: A Phenomenology of Film Experience.
Princeton, NJ: Princeton University Press.
—. 2004. Carnal Thought: Embodiment and Moving Image Culture. Berkeley/Los Angeles/London:
University of California Press.
Speck, Oliver C. 2010. Funny Frames: The Filmic Concepts of Michael Haneke. New York: Continuum.
Speed, Bebe. 1991. ‘Reality Exists OK? An Argument against Constructivism and Social Con-
structionism’. Journal of Family Therapy 13 (4): 395-409.
Stewart, Michael. 2007. ‘Irresistible Death: 21 Grams as Melodrama’. Cinema Journal 47 (1): 49-69.
doi:10.1353/cj.2007.0051.
Stone, Kevin. 2015. ‘BrainBashers: Cambridge University’. Brainbashers.com. http://www.
brainbashers.com/cambridge.asp.
Strogatz, Steven H. 2000. Non-linear Dynamics and Chaos: With Applications to Physics, Biology,
Chemistry, and Engineering. Studies in Non-linearity. Cambridge, MA: Perseus Books.
Tan, Ed S. 2013. Emotion and the Structure of Narrative Film: Film as an Emotion Machine. New
York: Routledge.
Tarnay, László. 2010. ‘On Perception, Ostranenie, and Specificity’. In On ‘Strangeness’ and the
Oever, 1: 141 155. Amsterdam: Amsterdam University Press.
Thanouli, Eleftheria. 2009. ‘“Art Cinema” Narration Today: Breaking down a Wayward Paradigm’.
Scope, no. 14: Articles: 1-14.
Thompson, Evan. 2007. Mind in Life: Biology, Phenomenology, and the Sciences of Mind. Cam-
bridge, MA: Harvard University Press.
Thompson, Kristin. 1988. Breaking the Glass Armor: Neoformalist Film Analysis. Princeton, NJ:
Princeton University Press.
—. 1999. Storytelling in the New Hollywood: Understanding Classical Narrative Technique.
Cambridge, MA: Harvard University Press.
Thoreau, Henry David. 1995. Walden, Or, Life in the Woods. New York: Dover.
Tikka, Pia. 2008. Enactive Cinema: Simulatorium Eisensteinense. Finland: University of Art and
Design Helsinki.
Timberg, Scott. 2001. ‘Indie Angst’. New Times Los Angeles, March 15.
Toffler, Alvin. 1984. ‘Foreword: Science and Change’. In Order Out of Chaos: Man’s New Dialogue
with Nature, by Ilya Prigogine & Isabelle Stengers, xi-xxvi. New York: Bantam.
Toulmin, Stephen. 1992. Cosmopolis: The Hidden Agenda of Modernity. Chicago, IL: University
of Chicago Press.
Tröhler, Margit. 2006. ‘Plurale Figurenkonstellationen: Die Offene Logik Der Wahrnehmbaren
Möglichkeite’. Montage/av 15 (2): 95-114.
Truffaut, Francois, Alfred Hitchcock, & H. G. Scott. 1984. Hitchcock. New York: Simon & Schuster.
Tsoukas, Haridimos, & Mary Jo Hatch. 2001. ‘Complex Thinking, Complex Practice: The Case
for a Narrative Approach to Organizational Complexity’. Human Relations 54 (8): 979-1013.
doi:10.1177/0018726701548001.
Turvey, Malcom. 2014. ‘Evolutionary Film Theory’. In Cognitive Media Theory, edited by Ted
Nannicelli & Paul Taberham, 46-61. New York/London: Routledge.
Vaage, Margrethe Bruun. 2009. ‘Self-Reflection’. Nordicom Review 30 (2): 159-178.

Varela, Francisco Javier, Antonio Coutinho, Bruno Dupire, & Nelson Vaz. 1988. ‘Cognitive
Networks: Immune, Neural and Otherwise’. Theoretical Immunology 2: 359-375.
Varela, Francisco Javier, Evan Thompson, & Eleanor Rosch. 1992. The Embodied Mind: Cognitive
Science and Human Experience. Cambridge, MA: MIT Press.
Voss, Christiane. 2011. ‘Film Experience and the Formation of Illusion: The Spectator as “Sur-
rogate Body” for the Cinema’. Translated by Inga Pollmann. Cinema Journal 50 (4): 136-150.
—. 2014. Der Leihkörper: Erkenntnis Und Ästhetik Der Illusion. Film Denken. München: Wilhelm
Fink Verlag.
Wedel, Michael. 2009. ‘Backbeat and Overlap: Time, Place, and Character Subjectivity in Run
Lola Run’. In Puzzle Films: Complex Storytelling in Contemporary Cinema, edited by Warren
Buckland, 129-150. Malden, MA: Blackwell-Wiley.
Wheeler, Michael. 2005. Reconstructing the Cognitive World: The Next Step. Cambridge, MA/
London: MIT Press.
White, Hayden. 1980. ‘The Value of Narrativity in the Representation of Reality’. Critical Inquiry
7 (1): 5-27.
Wilson, Emma. 2009. Alain Resnais. French Film Directors. Manchester/New York: Manchester
University Press.
Yacavone, Daniel. 2015. Film Worlds: A Philosophical Aesthetics of Cinema. New York: Columbia
University Press.
Young, Kay, & Jeffrey L Saver. 2001. ‘The Neurology of Narrative’. SubStance 30 (1): 72-84. doi:http://
doi.org/10.1353/sub.2001.0020.
Žižek, Slavoj. 2001. The Fright of Real Tears: Krzysztof Kieślowski between Theory and Post-Theory.
London: BFI (British Film Institute) Publishing.
Notes
I. Introduction
1. ‘Complex narratives’ are far from alone in exploring the interstice between
classical and modernist forms. Another favoured example of this is the New
Hollywood (or American New Wave) cinema that ‘had been raised on Old
Hollywood and 1960s art movies’ (Bordwell 2006, 74). In his introduction
to New Hollywood, Geoff King (2002) sums up the two main claims that
has been used to define New Hollywood. The first identifies New Holly-
wood with a particular style of filmmaking that is markedly different from
the classical style that preceded it. Another take has been to identify New
Hollywood with a set of industrial changes and thus place it in a broader
societal, political, ideological, and economic context (1–11; cf. Berliner 2010;
Elsaesser, Horwath, & King 2004; Thompson 1999).
2. It should be mentioned that TV series have also experienced an increased
surge of narrational complexity. The Sopranos (Chase 1999-2007), Lost
(Abrams, Lieber, & Lindelof 2004-2010), Breaking Bad (Giligan 2008-2013),
FlashForward (Braga & Goyer 2009-2010), and True Detective (Pizzol-
latto 2014-present) are but a few examples of this (cf. Mittell 2015).
3. Modern, modernist, or ‘art cinema’ are the most common terms used to
describe the cinematic movement that developed as an alternative to Holly-
wood filmmaking and is associated with European post-war auteurs. However,
there is no agreement as to whether these films constitute an institutional
practice (cf. Neale 1981), a narrational paradigm (cf. Bordwell 1985a), or even
if they are best understood as ‘modern’ in the sense of the most mature,
artistic and refined form of cinema (cf. Deleuze 2005a, 2005b). Depending
on how modern cinema is defined, Orson Welles’ Citizen Kane (1941), for
instance, can be seen as both a classical and a modern(ist) film. Within the
framework of this study, I have decided on the term ‘modern(ist)’. For the
present purpose both the term ‘modern’ and ‘art cinema’ have their disadvan-
tages. Whereas the term ‘modern’ risks confusing contemporary cinema with
modernist cinema, Bordwell’s choice of ‘art cinema’ as contrast to classical
Hollywood cinema comes with the unfortunate and unintended connotation
that classical cinema – indeed anything but ‘art cinema’ – is less ‘artistic’. Ul-
timately, the term ‘modern(ist)’ has been chosen since it denotes the primary
modernist adherence of the time-image, while simultaneously acknowledging
that Deleuze never intended to restrict the time-image to modernist cinema.
4. For more elaborate discussions of the concept of ‘affect’ in the tradition of
Spinoza and Deleuze, see Massumi (2002), Shouse (2005), Clough & Hal-
ley (2007), and Gregg & Seigworth (2010).
5. John Mullarkey (2009) argues that ‘reality itself is processual or divergent.
As such, film’s power is always based on a missed encounter, a convergence
with divergence [...] Commensurately, there is no essential or “Ideal” film
wherein either a particular technology or aesthetic form would render it ab-

solutely Real (“great or true” film art). Rather, at any one time, there is only
a provisional selection of film examples (and film scenes) that converge
on one point – what film “really is” – from a certain frame of reference’ (xv,
emphases in original). In this sense, the embodied fabula should be seen
as a ‘processual’ term that does not claim to capture the essence of the cin-
ematic experience or to answer the question of what ‘cinema really is’, but
enables us to better understand cinema and the cinematic experience from
a certain frame of reference at a particular moment in time.
6. The actual word Aristotle (1996) uses for complex is peplegmenos, which
translates into interwoven. See Buckland (2009a) for a more comprehensive
account of Aristotle’s idea of the complex plot, and its relation to Bordwell’s
conception of the ‘forking-path’ narrative.
7. As encapsulated in the famous passage from A Philosophical Essay on Probabil-
ity (1951) in which Pierre-Simon Laplace entertains the thought experiment of
an intellect (known as Laplace’s demon) that knows the precise location and
momentum of every atom in the universe. Laplace writes: ‘We may regard the
present state of the universe as the effect of its past and the cause of its future.
An intellect which at a certain moment would know all forces that set nature
in motion, and all positions of all items of which nature is composed, if this
intellect were also vast enough to submit these data to analysis, it would em-
brace in a single formula the movements of the greatest bodies of the universe
and those of the tiniest atom; for such an intellect nothing would be uncertain
and the future just like the past would be present before its eyes’ (4).
8. An example of a restricted form of complexity is statistics, in which, follow-
ing Mary Ann Doane (2002), ‘[c]hance was granted its own power, but that
power was ultimately superseded by general laws of order and regularity’ (18).
9. For a critical discussion of the cognitive position, see Mullarkey (2009, 29-57).
10. While the movement-image and the time-image are Deleuze’s two major
categories for cinematic images, it is worth pointing out that these do not
directly correspond to the opposition of ‘classical’ and modern(ist) cinema.
The movement-image is Deleuze’s broad description of a type of cinema
dominant in the pre-war era and defined by an empirical representation
of time as a chronological and linear succession in space. ‘Movement-
images’ come in various shapes such as ‘perception-images’ (e.g. Vertov) and
‘affection-images’ (e.g. Dreyer), however this category is most often associ-
ated with the ‘action-image’. The latter corresponds to the ‘classical’ film as
described in cognitive film theory. In Deleuze’s other major image-regime,
the time-image, the linkages that subordinated time to movement in the
movement-image are loosened or break to reveal a direct image of time (as
opposed to the indirect image of time in the movement-image). Deleuze
(2005b) writes about the ‘crystal-image’, which is the most important
subcategory of the time-image: ‘What constitutes the crystal-image is the
most fundamental operation of time: since the past is constituted not after
Notes 231
the present that it was but at the same time, time has to split itself in two at
each moment as present and past, which differ from each other in nature,
or, what amounts to the same thing, it has to split the present in two hetero-
geneous directions, one of which is launched towards the future while the
other falls into the past. Time has to split at the same time as it sets itself
out or unrolls itself: it splits in two dissymmetrical jets, one of which makes
all the present pass on, while the other preserves all the past. Time consists
of this split, and it is this, it is time that we see in the crystal. The crystal-im-
age was not time, but we see time in the crystal’ (79, emphasis in original).
11. Although this study does not examine the role of sound in cinema, I believe
that the embodied fabula could be a useful tool for understanding the im-
pact of cinematic sound and the way it enfolds or embeds the spectator in
the narrative in a manner that cannot be meaningfully accounted for by the
Bordwellian cognitive-analytical perspective.
12. In relation to this, I believe that it would be a both appealing and pertinent
future task to connect the concept of the embodied fabula to the growing
body of work that connects the logic of contemporary capitalist labour cul-
ture and consumerism with the foregrounding of affect, the accentuation
of non-linearity, and the transformations in spectatorial address witnessed
in contemporary media culture (Shaviro 2010; see also Deleuze and Guattari
1987; Hardt and Negri 2000, 2004).
II. Cinema in the Interstices

1. With a nod to an argument presented in the work of Tom Gunning (1986),
Doane (2002) observes that the ‘overwhelming hegemony of narrative in
the later Hollywood cinema of the classical era led earlier film historians
to construct a teleology that organized silent films and hierarchized them
according to their ability to anticipate the dominant narrative function and
to ‘invent,’ or ‘discover,’ its most salient signifying strategies’ (141).
2. Consider, for instance, Aleksandr Sokurov’s stunning Russian Ark ([Russkiy
kovcheg] 2002), which is shot in one entire take. The film is bathed in the
irreversible time of the camera. A documentary about the making of the film
stresses that it took Sokurov and his team an entire six months to practise and
rehearse to get the shot right on the one day that the Hermitage was reserved
for them. Thus, the film as it unrolls in one take glimmers with the tension
and anxiety stemming from the knowledge that one unpredicted event could
either create a magic moment or ruin the entire take. Consequently, despite
the digital nature of the film – one scene contains digitally crafted snowflakes
– the film incorporates the analogue implications as a reminder and reflec-
tion of the archival nature of images, history, and art.
3. Speaking about ‘post-continuity’ in Spring Breakers (Korine 2012) at a
conference at the Freie Universität Berlin 2013, Shaviro mentioned these
criteria with which ‘post-continuity’ can be differentiated from the cinema of
continuity (cf. Shaviro 2013).
4. ‘SLAB-theories refer to theories within cinema studies that are based on or

frequently refer to Saussure, Lacan, Althusser, and/or Barthes.
5. With the term ‘second nature’ Nadel (2005) refers to ‘the affirming of nature
in the interest of seeing that affirmation return as practice. To second, as a
verb, is to affirm, to add to the consensus, to echo. In this sense, to second
nature is to echo affirmatively; it is more or less the act of mimesis that film
itself has been lauded for performing since its inception, such that many of
the theoretical debates about film in the first half of the twentieth century
focused on which modes of cinematic representation were best suited to
reflect reality. The seconding of nature entailed in responding to cinematic
cues thus can be seen as the affirming of cinematic reality, the seconding of
film’s kinship to nature’ (430).
6. Other insightful introductions to cognitive science are Boden (2006) and
Clark (2000).
7. Wheeler (2005) also points out that this is a kind of cognition with which
most other animals do not engage at all (9).
8. In Reconstructing the Cognitive World (2005) Michael Wheeler draws pri-
marily on Heidegger as philosophical inspiration, Evan Thompson expands
this to include Husserl in his Mind in Life (2007). In Alva Noë’s Action in
Perception (2004) Merleau-Ponty is a frequent reference. Antonio Dama-
sio dismantles the Cartesian assumptions of cognitive (neuro)science in
Descartes’ Error (1995), while turning to Spinoza for an alternative to these
presumptions in Looking for Spinoza (2003). In this light John Protevi (2010)
has suggested that a constructive next step could be found by not only
‘looking for Spinoza’, but also Deleuze.
9. For works examining the core assumptions of cognitive science, see Bick-
hard and Terveen (1996), Dennett (1991), Dreyfus (1991), Fodor (1983), Lem-
men (1997), Gardner (1985), Varela et al. (1992), and Wheeler (2005).
10. According to Wheeler (2005) a ‘creature displays online intelligence, just
when it produces a suite of fluid and flexible real-time adaptive responses
to incoming sensory stimuli’ (12). Online intelligence is thus opposed to
offline intelligence such as ‘wondering what the weather is like in Paris,
or weighing up the pros or cons of moving to a new city’ (12). Whereas the
cognitive-formalist fabula has mostly been perceived from an ‘offline’ per-
spective, the embodied fabula is an attempt to accentuate the importance
of ‘online’ cognition as describing our immediate encounter with the films.
11. At the end of the second cinema book, Deleuze (2005b) suggests that new
powers of the image must go beyond both the movement-image and the
time-image as these are described in his books: ‘An original will to art has al-
ready been defined by us in the change affecting the content of cinema itself:
the substitution of the time-image for the movement-image. So that electron-
ic images [as opposed to analogue, i.e. digital] will have to be based on still
another will to art, or on as yet unknown aspects of the time-image’ (255).
Notes 233
12. In his major work on cinematic modernism Screening Modernism (2007),

András Bálint Kovács points to the film-philosophy of Deleuze as containing
three aspects of the ‘classical-modern’ dichotomy – Deleuze represents one
side of the spectrum, David Bordwell the other. Kovács writes that ‘in the
Deleuzian approach to modern cinema one can find all three aspects of the
classical-modern dichotomy. He sets out a systematic distinction between
classical and modern cinema, whereby modern cinema is seen as a different
utilization of moving images. He also sets up a chronological order whereby
modern cinema appears as an organic development of classical cinema. Final-
ly, he puts modern cinema on a higher level of evolution where cinema fulfills
its potential for expressing abstract thoughts. According to Deleuze, modern
cinema is the most developed structural variation of classical cinema, which
articulates the actual world better and in a deeper sense than classical cin-
ema. “Classical” does not mean for Deleuze an “everlasting”, eternal model of
aesthetic value. Not that he does not respect and admire classical auteurs, but
he considers classical film form to be outmoded, passé, invalid, discredited.
Although Deleuze designates a certain historical moment for the appearance
of modern cinema, he does not treat modernism as an art-historical phenom-
enon in the sense of an art movement, trend, or school. Modern film is the
result of the evolution of cinema’s inherent power of articulating time’ (41).
13. An inspiring analysis that places Adaptation in the vacuum between
Deleuze’s movement-image and time-image as an example of the ‘Kino der
Replikation’ [cinema of replication] can be found in Engell (2005).
14. Richard Smith’s main example is the work of Billy Wilder, who is ‘one example
of a director who worked within the classical system of montage (is there re-
ally such a thing?) but who has “time” as an absolutely central component of
his thought’ (2001). Another obvious example is Alfred Hitchcock, whose films
are widely referred to as classical, yet for Deleuze often fall into the category of
‘relation-images’ that belong in the interstice of movement-image and time-
image. The cinema of Hitchcock thus enables an ideal position to perform a
critical reflection upon the classical film and our comprehension of it.
15. See Fahle & Engell (1997) for a lengthy and multifaceted discussion on the
role of history in the cinema of Deleuze.
16. The fact that the cinema entailed both linear and non-linear elements from
its very birth is consistent with the arguments proposed by Doane (2002)
and Gunning (1986).
17. At times Deleuze writes ‘narration’ to refer to classical linear narration,
and at other times to refer to all conceivable types of narration, including
‘dysnarration’.
18. For Deleuze and the horror film, see Anna Powell’s Deleuze and Horror
Film (2006), for Deleuze and national cinema see David-Martin Jones’
Deleuze, Cinema, and National Identity (2006), and his book on ‘world cin-
ema’, Deleuze and World Cinemas (2011), for Deleuze and feminism see the
collection Deleuze and Feminist Theory (Buchanan & Colebrook 2000), and
Laura Marks (2000, 2002) for an embodied Deleuzian approach. Granted

most of these studies touch upon questions of narration, however, when the
immensity that the topic of narration in the field of cinema is considered,
works that focus primarily on narration appear startlingly modest. Perhaps,
the reason is that Deleuzian scholars often feel that studies of narration
have traditionally been overrepresented. While this may be a fair point, the
rise of complex narratives does call for attempts to develop a Deleuzian or
film-philosophical approach to cinematic narration.
19. The close relation between ‘intensity’ and ‘affect’ in Massumi (2002) is
evident, when he argues ‘emotion and affect – if affect is intensity – follow
different logics and pertain to different orders’ (27).
III. Narrative Ambiguity in the Classical Cinema

1. While Stage Fright is arguably situated within the classical paradigm it si-
multaneously points towards both the time-image and the complex narrative.
Consequently, Deleuze (2005a) holds the cinema of Hitchcock to be represent-
ative of the ‘relation-image’ situated in the post-war gap between movement-
image and time-image (201-209). Furthermore, Elsaesser (2009) perceives the
film as a precursor to the complex storytelling mode found in contemporary
cinema due to its unreliable narration invoked by its ‘lying’ flashback (20).
2. The point being that the history of cinema – if allowed to show – is full of
examples that challenge what cognitivists perceive to be the ‘ordinary’ linear
cinematic perception. Some examples are the ‘cinema of attraction’, avant-
garde cinema, surrealist cinema, various forms of experimental cinema, the
films of D.W. Griffith, ‘Russian montage’, but also, and this is my point, films
that otherwise seem to adhere to a linear onto-epistemology. Yet, cognitive
film theory has demonstrated a predilection for the classical Hollywood film,
presumably because it is most apt in laying bare the relation between innate
cognitive dispositions and how we watch cinema. Malcom Turvey (2014)
has criticized how the universal claims occasionally made by cognitive
scholars such as Murray Smith, David Bordwell, Torben Grodal, and Joseph
Anderson, often rely upon films from the industrialized West. Surely, he
writes, ‘an analysis of a much wider sample of both visual and oral children’s
narrative fiction from Western and non-Western cultures, from the past and
the present (including from contemporary hunter-gatherer populations), is
necessary before one can so confidently state that children’s stories about
empowerment and bonding are universal’ (54). Stage Fright here serves as
an example of how the supposed naturalness of cinematic linearity is often
a product of the bias inherent in our analytical tools, rather than being the
undisputed universal condition or mode of sense-making.
3. The Independent, for instance, wrote in their review, ‘[t]he film’s coup de
grace is as elegant as it is unexpected. The whole movie plays back in your
mind in perfect clarity – and turns out to be a completely different movie to
the one you’ve been watching’ (Q. Curtis 1995).
Notes 235
4. Bordwell develops his theory of cinematic spectatorship on the example

of Hitchcock’s Rear Window (1954). In the film, the wheelchair-bound Jeff
(James Stewart) solves a murder that happened in the opposite building.
Jeff’s attempt to solve the mystery by carefully observing his neighbours
from his window, mirrors the spectator’s cognitive attempts to solve the
puzzle of the film (cf. Bordwell 1985a, 40-47).
5. Making the audience re-aware of internalized cinematic conventions is argu-
ably one of the main defining aspects of contemporary complex narration.
6. Yet, the film does not challenge linear reasoning per se but automatic sche-
mata application. It is thus worth keeping in mind that the theatrical meta-
phors do not ensure cinematic non-linearity in the Deleuzian sense and the
film ultimately expresses a belief in the classical, linear onto-epistemology.
7. See Thompson (1988) for a more thorough examination of how the charac-
ters are at once acting, directing, and auditing in Stage Fright.
8. The correct spelling of the word is actually ostrannenie. However, owing to a
curious spelling mistake made by Shklovsky the word has entered the diction-
aries with the spelling ostranenie, which I here follow (cf. van den Oever 2010).
9. In the following, I use the term ‘cognitive formalism’ to refer to the conjunc-
tion of cognitive psychology and Russian formalism found in the work of
Bordwell, Thompson, and others. It should be noted that Thompson prefers
the term ‘neoformalism’ to stress that her main reference point is Russian
formalism.
IV. Modern(ist) Cinema: Logic of the Encounter

1. It was probably Jacques Rivette who first pointed out the similarities between
the short stories of Jorge Luis Borges and the narrative forks of Hiroshima
mon amour. Rivette observes, ‘Hiroshima [mon amour] is a circular film.
At the end of the last reel you can easily move back to the first, and so on. Hi-
roshima is a parenthesis in time. It is a film about reflection, on the past and
on the present. Now, in reflection, the passage of time is effaced because it is
a parenthesis within duration. And it is within this duration that Hiroshima
is inserted. In this sense Resnais is close to a writer like Borges, who has
always tried to write stories in such a way that on reaching the last line the
reader has to turn back and reread the story right from the first line to under-
stand what it is about – and so it goes on, relentlessly. With Resnais it is the
same notion of the infinitesimal achieved by material means, mirrors face to
face, series of labyrinths. It is an idea of the infinite but contained within a
very short interval, since ultimately the “time” of Hiroshima can just as well
last twenty-four hours as a second’ (Rivette in Domarchi et al. 1985, 69).
2. At least this was the case in 1959. It can be argued that the genre of docu-
mentary in recent years has undergone an exciting development in which
directors have experimented with many innovative approaches to the format
of documentary. These new documentaries have abandoned the classical ‘fly-
on-the-wall’ documentary style, and use acts of intrusion that would be un-
imaginable in traditional documentary. An example is The Act of Killing

(Oppenheimer 2012), a film that has profited from the renewed understand-
ing of ‘representation’ and ‘reality’ promoted in the modern(ist) paradigm.
3. One example is the case of Henry Molaison (also known as ‘Patient HM’),
who, following a brain operation that was intended to cure him of his
epilepsy, lost the ability to form new long-term memories (anterograde
amnesia). Molaison, who died in 2008, is recognized as one of the most im-
portant cases in the history of cognitive neuroscience, and his case counts
among the main sources of inspiration for Memento.
4. In the time-image, Deleuze (2005b) argues that montage has become
‘montrage’ – a term he borrows from Robert Lapoujade. Deleuze writes:
‘[M]ontage has changed its meaning, it takes on a new function: instead of
being concerned with movement-images from which it extracts an indirect
image of time, it is concerned with the time-image, and extracts from it
the relations of time on which aberrant movement must now depend’ (40).
Consequently, the function of montage is no longer to link a series of events
into a horizontal line.
5. Deleuze (2005b) writes: ‘It is only when movement becomes automatic that
the artistic essence of the image is realized: producing a shock to thought, com-
municating vibrations to the cortex, touching the nervous and cerebral system di-
rectly. [...] The spiritual automaton no longer designates – as it does in classical
philosophy – the logical or abstract possibility of formally deducing thoughts
from each other, but the circuit into which they enter with the movement-
image, the shared power of what forces thinking and what thinks under the
shock; a nooshock. [...] It is this capacity, this power, and not the simple logical
possibility, that cinema claims to give us in communicating the shock. It is as if
cinema were telling us: with me, with the movement-image, you can’t escape
the shock which arouses the thinker in you’ (151-152, emphasis in original).
6. I find this example to be illustrative of Bergson’s thesis of the enfolding of
memories and perception: ‘Accondirg to a resecrah at Camirbdge Univsreity,
it dosen’t mettar in waht oredr the lertets in a wrod are. The olny itpormant
thnig is taht the frist and lsat letter be in the rgiht plcae. The rset can be a toatl
mses and you can sitll raed it wihtout plobrem. Tihs is becuase the huamn
mnid deos not raed eevry lteter by ilsetf, but the wrod as a wohle’ (Stone 2015).
7. It should, however, be noted that the classical flashback is not always a
representation of a subjective memory. Stage Fright proves an example
of the borderland, where it is not immediately clear whether it is the film’s
leap back in story time or the film’s representation of a subjective or falsi-
fied ‘memory’ (cf. Chapter III).
8. Deleuze (2005b) explains the cone as follows: ‘The point S is clearly the
actual present; but this is not strictly speaking a point, since it already
includes the past of this present, the virtual image which doubles the actual
image. As for AB, A’ B’... sections of the cone, they are not psychological
circuits to which recollection-images would correspond; they are purely
Notes 237
virtual circuits, each of which contains all our past as this is preserved in
itself (pure recollection). Bergson is unequivocal in this respect. Psychologi-
cal circuits of recollection-images or dream-images are produced only when
we “leap” from S to one of these sections, to actualize some virtuality of it
which must then move down into a new present S’’ (284, n. 22).
9. ‘Pure’ memory is to be understood as the totality of memory (Deleuze 1991,
27), and it is in principle independent of perception (Moulard-Leonard
2008, 65) and thus the opposite of ‘pure’ perception.
10. The very different experience of the atomic event is a constant theme in
the dialogue of the lovers. He, for instance, asks if it was true that it was a
beautiful summer day in Paris, and she recollects that the bombings were
associated with the definite end of the war.
11. When the woman asks ‘why speak of him’ the man answers: 1) ‘Because of
Nevers. I’m only just beginning to know you, and from the many thousands
of things in your life, I chose Nevers’.; 2) ‘I somehow understand that it
was there that you were so young that you didn’t yet belong to anyone in
particular, and I like that’; 3) ‘I somehow understand that it was there that I
almost lost you and ran the risk of never, ever meeting you’; 4) ‘I somehow
understand it was there that you began to be who you are today.’
12. When asked ‘what did Hiroshima mean to you in France?’ the French woman
replies: ‘The end of the war. I mean completely. Astonishment that they dared
to do it, astonishment they succeeded. And the beginning of an unknown
fear for us as well. And then indifference. And fear of indifference as well.’
V. Towards the Embodied Fabula

1. This humorous approach to the separation of body and mind first appeared
in Punch Magazine in 1855.
2. This editing technique – a subcategory of fast cutting – has also been
practised in films such as Boogie Nights (Anderson 1997) and Snatch
(Ritchie 2000) and was coined the ‘hip-hop montage’ following an interview
in which Aronofsky (2001) has explained: ‘I grew up in Brooklyn during the
eighties and the golden age of hip-hop; before Eminem. As a kid I was a
really bad graffiti artist and a really bad breakdancer but I still wanted to
take some hip-hop ideas and apply them to narrative filmmaking. So that’s
where all the fast cutting came from. It just happened to work really well
with the idea of obsession and addiction’.
3. ‘Illusions’ here are not to be understood as ‘unreal’, but rather in the techni-
cal sense as a distortion of the senses capable of revealing the underlying
organizational and perceptual workings of the brain. Understood like this,
illusions are closely related to Deleuze’s concept of the virtual.
4. However, it would be possible to argue that cognitivists tend to understand
the affective (or ‘irrational’) aspects of cinema within the fold of rational
agency. Daniel Frampton (2006), for instance, makes this claim, when he as-
serts that cognitive film theory explains cinema going in terms of ‘problem-
solution’ models according to which the ‘filmgoer’s main emotional engage-

ment with the narrative film is that of “interest”’ (108).
5. Remember that Bordwell’s theory of style is partly naturalistically justified,
which, according to cognitive formalism, explains why certain norms are ex-
perienced as being more natural than others (cf. Chapter II & Chapter III).
6. Richard Allen (2001) makes a similar point when he argues that within cog-
nitive film theory, ‘[p]erception is erroneously conceived as a process that
takes place inside the mind’ (185). Like Noë (2004), Allen (2001) maintains
that perception is not a container of things but an ability manifested in
behaviour (186).
7. The researchers attributed with the discovery of mirror neurons are Vittorio
Gallese, Giacomi Rizzolatti, Giuseppe Pellegrino, and Leonardo Fogassi.
8. It is worth noting that the actual phenomena or experience that would arise
from the firing of mirror neurons is already known to film scholars – one
only needs to think about the audience’s inner arousal when watching a
car about to crash. Therefore, what is central is that these provide scientific
support to those claiming that cinema’s ability to arouse such reactions –
even when the spectator on a purely cognitive level is consciously aware
that the images are fictional (and thus in a sense fictitious) – demonstrate a
deeper underlying bond between the images and the audience.
9. It is, nevertheless, crucial to note the difference between the brain as envi-
sioned by Deleuze and the evolutionarily ‘hardwired’ machinery implied
in embodied neurobiological takes on cinema, narrative, and emotion as
exemplified in the work of Grodal. Whether mirror neuron mechanisms
confirm the evolutionary ‘hardwired’ brain of Grodal or the ‘film-spectator-
world’ (cf. Brown 2013) assemblage of Deleuzian film-philosophy remains a
disputed topic. In any case, the mirroring mechanism in the brain consti-
tutes a further challenge to the cognitive-formalist conception of narrative
comprehension and thus evidences the need for a reconceptualization of
the fabula as a non-representational tool for comprehending the immediate
encounter between spectator and film.
10. Michael Wheeler (2005) explains that this specific mode of causation
involves multiple simultaneous interactions and complex dynamic feedback
loops. This has two consequences: first that the ‘causal contribution of each
systemic component partially determines, and is partially determined by, the
causal contributions of large numbers of other systemic components’ (260).
Secondly, that these contributions are suspect to radical change over time
(cf. Clark 1998; Wheeler 2005). Such a view on cognition means that the clas-
sical cognitive methodology becomes problematic, since it attempts to speci-
fy ‘distinct and robust causal-functional role played by reliably reidentifiable
parts of the system, and to explain interesting system-level behavior in terms
of the properties of a small number of subsystems’ (Wheeler 2005, 260-261).
11. A few examples of complex narratives that can be said to have a non-linear
starting point are the bifurcations of Lola rennt (1998) or The Butterfly
Notes 239
Effect (2004), the fractal structure of Thirteen Conversations about

One Thing (Sprecher 2001) or 21 Grams (2003), the reverse chronologi-
cal storytelling of Memento (2000) or the narrative loops of Lost High-
way (1997). For an exploration of complex system theory and complex
narratives see (Poulaki 2011, 2014a, 2014b).
12. In contrasting ‘grass’ to ‘tree’ Elliott (2010) refers to the rhizomatic (grass)
versus the classical logical (tree) structures of thinking as found in Deleuze
and Guattari (1987).
13. For a more comprehensive account of Lost Highway, see Hven (2010).
14. Breaking Bad (Gilligan 2008-2013), Mad Men (Weiner 2007-2015), and
Dexter (Manos 2006-2013) are but a few examples of how contemporary
television series experiment with simultaneously aligning and misaligning
its spectators from its main characters’ actions.
15. For Deleuze (2000) this immediately becomes a normative issue as ‘[b]ad
cinema always travels through circuits created by the lower brain: violence
and sexuality in what is represented – a mix of gratuitous cruelty and or-
ganized ineptitude. Real cinema achieves another violence, another sexual-
ity, molecular rather than localized’ (367).
16. This is not to deny the possibility that some viewers simply root for Nor-
man, yet in such cases the cognitive dissonances that are of interest here
will not occur.
17. In the concepts of hodological and pre-hodological spaces, one senses the
contours of the movement-image (hodological) and time-image (pre-hodo-
logical). I have chosen to demonstrate this with an example taken from Psy-
cho exactly to avoid this dichotomy, showing that their existence does not
necessarily cancel each other out. Here it is important to note that Psycho is
itself a film that disobeys the dichotomy of classical and modern(ist) cinema.
18. See Holtmeier (2014) for an illuminating exploration of pre-hodological
spaces in the cinema of Jia Zhangke.
VI. The Complexity of Complex Narratives

1. Remember that Wheeler (2005) counts a dynamical system perspective
as the fourth claim for embodied cognition. He states that this amounts
to a rejection of the idea that all cognitive processes are computational,
favouring instead to regard cognition as a matter of state space evolution in
particular kinds of dynamic systems (13-14).
2. Another example of this is Zeno’s paradox with Achilles and the turtle
through which Henri Bergson (1998, 1991) developed his criticism of science.
3. Cilliers (2002) differentiates complex from complicated systems. It is inter-
esting to note that a computer is a complicated system, but not a complex
system. This is the case, since in a complex system ‘the interaction among
constituents of the system, and the interaction between the system and its
environment, are of such a nature that the system as a whole cannot be fully
understood simply by analysing its components. Moreover, these relation-
ships are not fixed, but shift and change, often as a result of self-organisa-
tion [...] The brain, natural language and social systems are complex’ (iix-ix).
4. The unusual circumstances of their love affair are explicitly remarked upon
as Cristina ensures Paul that she knows he ‘has a good heart’.
5. Remember that here, as elsewhere, emotion is understood as being both af-
fectively and cognitively informed responses to the environment (therefore
often the cinematic material). While emotions can be distinguished from
pre-subjective affects and ‘higher-functioning’ cognition they remain deeply
dependent on both. For Newman, however, we are not given sufficient cog-
nitive information to conjure up strong emotions. Instead, the film overloads
our cognitive system with difficult computational tasks. I believe that it is
Newman’s mistake – in fact the mistake of most cognitivists – that he com-
pletely ignores the role of affect (since it is preconscious) in our experience
of the narrative. Similarly, I believe that the new affect theories largely ig-
nore the role of cognition and emotion, when they address cinematic affect.
6. It is quite likely that the popularity of certain non-linear twist films – catego-
rized as ‘psychological puzzle films’ (cf. Panek 2006) – partly depends on the
joy experienced once a narrative order emerges from the initial state of cha-
os. In certain ways, 21 Grams resembles ‘psychological puzzle films’ insofar
as they, too, deal extensively with ambiguity and dispend with the clarity of
classical cinema in favour of sudden narrative fluctuations over brief isolated
fluctuations and clarity of classical narration. Yet, ‘psychological puzzle films’
typically employ the narrative structure of the detective plot to deal with
unreliable narration, surprise or twist endings, narrative ambiguity, and/or
characters suffering from mental illness (cf. Panek 2006). Contrarily, it has
been argued that 21 Grams due to its imploding and dysfunctional families,
suffering women, broken and disoriented men, its intense and expressive
forms of realism together with its emphasis on affect and human desire is
closer related to the genre of the melodrama (cf. Stewart 2007; Azcona 2010).
Unlike the detective film, the melodrama assumes the spectator primarily to
be emotionally and affectively involved with the narrative.
7. The issue for Keating (2006) is exactly that of leaving behind the predomi-
nantly linear understanding of the manner in which Hollywood cinema
elicits emotion: ‘If narrative is seen as a rational system of organization
and containment while the attraction is seen as a momentary appeal to
the senses, then it seems hard to avoid the conclusion that the systems are
locked in struggle. However, if narrative and attractions are both theorized
as systems for strong emotions, then it seems much more reasonable to
conclude that they can coexist peacefully. Indeed, they might even be able
to mutually intensify one another’ (9).
8. Note that this view is resonant with Deleuze’s theory of the sensory-motor
linkages that produce the sense of continuity in the movement-image that
are suspended or ruptured in the time-image (cf. Deleuze 2005a, 2005b).
Notes 241
9. The ‘mosaic’ or ‘network’ narrative is a subcategory of complex narratives

in which the perhaps most prominent factor is the lack of narrative centre
or unification. Yet, the mosaic form is not entirely new to cinema. D.W.
Griffith’s Intolerance (1916) is probably one of the earliest examples of a
‘mosaic’ narrative structure since the film cross-cuts between four parallel
stories separated in both time and place. In recent years, the ‘mosaic’ style
has been associated with the work of the director Robert Altman, and in
particular his film Short Cuts (1993), still it is only within the last dec-
ades that scholars have started to regard the ‘mosaic’ or ‘multiprotagonist’
film in generic terms (cf. Azcona 2010). Short Cuts displays many of the
characteristics of the ‘mosaic’ film since it has a large set of subplots and
characters, whose daily existence we follow in a manner that eschews the
protagonist-centred forward-movement typical of the classical narrative.
The characters and the events are never truly connected, yet the film still
facilitates a sense of interconnectedness since they are all more or less
affected by an earthquake occurring in Los Angeles. For a more comprehen-
sive account of the ‘mosaic’ narrative see Tröhler (2006), Azcona (2010), or
Pisters (2011b).
10. Apart from Lola rennt and 21 Grams the following – by no means exhaus-
tive – list of films has made explicit reference to complexity theory: Butter-
fly Effect (Bress & Gruber 2004), Thirteen Conversations about One
Thing (Sprecher 2001), Chaos Theory (Siega 2008), and Chaos (Giglio 2005).
11. Within the cognitive sciences the term ‘modularity’ refers to Jerry Fo-
dor’s (1983) thesis that the mind operates according to informationally
encapsulated ‘modules’. This influential theory has been criticized by
embodied cognition, as here exemplified by Chemero (2009): ‘Because
perception is informationally encapsulated, theories (not handled by
perceptual modules) do not change perceptual mechanisms or the output
of perceptual modules. Therefore, Fodor argues, perception is not theory
laden. Furthermore, because perceptual modules are innately structured,
they’re the same in all (normal) humans. There is, then, no sense in which
humans who believe different theories perceive a different world. Their
perceptual mechanisms produce the same output given the same input.
Although they may hold different theories about what they perceive, they
perceive the same thing. So, Fodor holds, the modularity of perceptual and
cognitive systems makes perception a neutral basis for theoretical disputes
and this can form the basis for objectivity and scientific realism’ (189).
12. This pun on the traditional altruism that seeing is believing, describes the man-
ner in which our mindset actually affects our perception (cf. Dennett 1991).
13. Borges (2000) writes: ‘In all fictional works, each time a man is confronted
with several alternatives, he chooses one and eliminates the others; in the
fiction of the almost inextricable Ts’ui Pȇn, he chooses – simultaneously
– all of them. He creates, in this way, diverse futures, diverse times which
themselves also proliferate and fork [...] In the work of Ts’ui Pȇn, all pos-
sible outcomes occur; each one is the point of departure for other forkings’
(Borges 2000, 51, emphasis in original).
14. It could be noted that some ‘complex narratives’, such as Timecode (Fig-
gis 2000), have employed a split screen to ensure simultaneity of its storylines.
15. In relation to Lola’s scream, Martin-Jones (2006), too, argues that she in this
manner manipulates the outcome, since the scream can be perceived as a
skill she has acquired over the course of the two first runs, and, therefore,
not a manner of sheer contingency. However, if the successful ending of
Lola’s third run is not completely a matter of sheer contingency, it can also
not be skill alone, and must instead be seen as a mixture of luck and skill.
As Deleuze (2004b) has remarked about chance and fixity in games: ‘In
games with which we are familiar, chance is fixed at certain points. These
are the points at which independent causal series encounter one another
(for example, the rotation of the roulette and the rolling ball). Once the en-
counter is made, the mixed series follow a single track, protected from any
new interference. If a player suddenly bent over and blew with all his might
in order to speed up or to thwart the rolling ball, he would be stopped,
thrown out, and the move would be annulled. What would have been
accomplished, however, other than breathe a little more chance into the
game?’ (71). In relation to this, Lola’s scream does not control or eliminate
the contingency of the game as much as it introduces a new bifurcation to
the game of chance.
16. It should rightly be pointed out that Martin-Jones naturally does not agree
with all aspects of Bordwell’s cognitive formalism. While Martin-Jones finds
Bordwell’s argumentation for the persistency of linearity within ‘forking-
path’ narratives compelling, he finds it problematic that Bordwell operates
at the level of narrative structure exclusively, and thus ignores the context-
specific aims of the various films, which results in a too homogeneous
appreciation of them (cf. Martin-Jones 2006, 87).
17. Here I am especially thinking about the scene in which Tykwer cross-cuts
between Manni and Lola, when the latter realizes that the bag is lost. ‘Die
Tasche!’ says Lola, and Manni replies, ‘die Tasche!’, the repetition of the
word ‘Tasche’ slowly develops into a kind of small musical interlude that
parallels the manner in which a narrative rhythm is introduced by means of
the repetitions of the film’s form.
VII. Memento and the Embodied Fabula

1. The origin of this riddle is unknown.
2. To mention but a few examples, Memento has been studied as an expres-
sion of Deleuzian film-philosophy (cf. Clarke 2002; Gargett 2002; Martin-
Jones 2006), as a ‘complex narrative’ (Simons 2008), a ‘modular narrative’
(Cameron 2008), as an example of the ‘mind-game’ and ‘post-mortem’ film
(Elsaesser 2009, cf. 2004b; Elsaesser & Hagener 2010), as a ‘subjective realist
narration’ (Campora 2014), a ‘psychological puzzle film’ (Panek 2006) as
Notes 243
well as a ‘puzzle film’ (Ghislotti 2009). Claire Molloy (2010) has devoted an
entire book to the study of several aspects of the film including produc-
tion, marketing, and narrative features. Diverse cognitive approaches to the
study of the film include those conducted by Stefano Ghislotti (2009) and
Karen Renner (2006).
3. The reverse motion scene of Leonard killing Teddy can be seen as a play
on the mechanic, analogue, and indexical nature of the cinematic appara-
tus, and its ability to bring images from the past alive. André Bazin (2005)
famously wrote about the cinematic medium: ‘Now, for the first time, the
image of things is likewise the image of their duration, change mummified
as it were’ (15).
4. Yet, as Melissa Clarke (2002) has observed, spectators ‘will be forced to
consider how accurate “normal” is’ (167).
5. It should rightly be noted that Bordwell (2006) does acknowledge the
lingering uncertainty surrounding many of the film’s events. He writes: ‘In
Memento, it seems to me, we can only suspect that the Sammy Jankis story
is Leonard’s projection of his own killing of his wife; the film doesn’t pro-
vide enough redundancy to let us ascertain this. Still, many matters, such as
Leonard shooting Teddy, don’t seem to be in doubt’ (81).
6. The subliminal image can be found here https://mendozalean.files.word-
press.com/2013/07/lenny-and-sammy.jpg.
7. From this perspective, the film resonates with Deleuze’s (2005b) thesis
about indiscernibility in the time-image. Here ‘we no longer know what is
imaginary or real, physical or mental, in the situation, not because they are
confusing but because we do not have to know and there is no longer even a
place from which to ask’ (7).
8. The original poster for the film can be found here: http://news.doddleme.
com/wp-content/uploads/2013/02/memento-movie-poster.jpg.
9. See Redfern (2005) for a comprehensive account of how Bordwell misinter-
prets the radical constructivist position.
10. I would like to thank a reviewer for pointing out that the ‘dizzying of feel-
ings’ described by the spectator of Memento is not necessarily to be taken
literally but metaphorically, i.e. this viewer is not describing dizziness in
the sense of being unable to stand up straight or to focus but metaphori-
cally as a cognitive state of confusion. From this perspective, the use of the
word ‘dizzying’ can be seen as an example of how intricately affective and
cognitive experience link up to form the ‘metaphors we live by’ (Lakoff &
Johnson 2003). According to conceptual metaphor theory (CMT), thought
is embodied insofar as ‘the structures used to put together our conceptual
systems grow out of bodily experience and make sense in terms of it. From
this perspective, ‘the core of our conceptual systems is directly grounded
in perception, body movements, and experience of a physical and social
character’ (Lakoff 1987, xiv; cf. Johnson 1990; Lakoff & Johnson 1999, 2003).
Within the last couple of years, a subfield dedicated to exploring the appli-
cation value of CMT to cinema has emerged within cognitive media studies
(cf. Coëgnarts 2014; Coëgnarts & Kravanja 2012, 2015; Fahlenbrach 2016).
Whether and how the concept of the embodied fabula could gain from, and
possibly contribute to, this cognitive film theoretical subfield remains to be
explored.
11. For interesting theories on cinema’s ability to produce its own subjectivity,
see Vivian Sobchack (2004) on the ‘cinesthetic subject’ and for its ability to
produce its own corporality, see Christiane Voss (2011, 2014) on the concept
of the Leihkörper [‘surrogate body’]. For a more elaborate examination of
the relation of these concepts and the embodied fabula, see Hven (2015).
Acknowledgments
My sincere thanks to all those whose encouragements, assistance, or

financial aid made this book possible. An earlier version of this book was
submitted to the Bauhaus-Universität Weimar under the title Embodying the
Fabula: Cinema between the Lines as my dissertation for obtaining the degree
of Doctor Philosophiae. I would like to thank the Graduiertenförderung des
Freistaates Thüringen for financial support and to express my gratitude to
the Bauhaus-Universität for providing such an exciting and stimulating
research environment. I am especially grateful to Lorenz Engell, whose
continued guidance I could not overestimate, and to Christian Kassung
for his equally superb supervision of this project. A special mention also to
Christiane Voss, whose support and influence especially in the later stage of
this project have been marvellous. My gratitude also goes out to my fellow
students in Weimar, Berlin, Aarhus, and Istanbul for your comments and
critical remarks.
I would also like to express my gratitude to Amsterdam University Press,
and in particular the editors of the series ‘Film Culture in Transition’, Jeroen
Sondervan and Thomas Elsaesser, for their professional, careful, dedicated,
and efficient work with the manuscript. I would like to thank two anony-
mous reviewers for thought-provoking comments and criticisms that have
helped to strengthen the argument of the book. Its weaknesses rest upon
my shoulders alone.
Several other people have contributed to the final shape of this book
with useful comments, critical remarks, and suggestions for improvements
or readings, or for their good cheer. I wish to thank Jeppe Graugaard, Tyler
Parks, Rebecca Sheehan, Kate Rennebohm, Blandine Joret, Bernard Geoghe-
gan, Gert Jan Harkema, Patricia Pisters, Josef Früchtl, Ágnes Pethő, Miklós
Kiss, Steven Willemsen, Brendan Rooney, Maria Poulaki, Carl Plantinga,
William Brown, Vittorio Gallese, David Bordwell, Torben Grodal, Andreas
Gregersen, Birger Langkjær, Ed S. Tan, John Bateman, Maarten Coëgnarts,
and all those people I have not mentioned but who have nevertheless
inspired me tremendously throughout the years.
Small parts of this book have been previously published. I would like to
express my gratitude to the editor Ágnes Pethő for allowing me to reuse
parts of the article ‘Memento and the Embodied Fabula: Narrative Compre-
hension Revisited’, which appeared in Acta Univ. Sapientiae, Journal of Film
and Media Studies, 11 (2015), 93-110. Thanks also to Maria-Theresa Teixeira
and Susana Viegas, the editors of the eProceedings of the International
Conference on Philosophy and Film, for allowing me to republish parts of

the article ‘Narrative is not in us; but it is we who move in a being-narrative,
a world-narrative’ that appeared in Volume 1, 260-277. I would also like to
thank Howard Gardner and Nick Redfern for their kind permission for using
their illustrations, and to Zone Books for allowing me to reprint Bergson’s
cone of memory.
Last but not least, I would like to express my gratitude for the support
of my family. To my parents Poul Erik and Eva for love, support, and so
much more. My gratitude also to my brother Troels, his wife Marianne,
and my sister Lisbeth. To Sigurd and Rolf for great cheer. Thanks to Gefion,
my wife, for the extraordinary attention and passion you have bestowed
upon this project and beyond. Finally, I would not be able to conclude my
acknowledgments without remarking upon my gratitude for the work on
the manuscript done by you, Karin. At times you invested more time and
hard work on the manuscript than I, and for that I am deeply grateful.
Steffen Hven, Berlin 2016

Index
21 Grams 7, 19, 22, 137, 140-60, 207, 209, 239n.11, Bergson, Henri 13, 27, 44, 71-73, 84, 88, 101-04,
240n.6, 241n.10 123, 127, 140, 151, 167-70, 189, 195, 236n.5,
237n.9, 239n.2
A Beautiful Mind 78 Bianco, Jamie Skye 17, 49, 52, 62, 111-12, 160, 170,
À bout de souffle (Breathless) 86 182, 196-97, 202
A History of Violence 132 Blind Chance (Przypadek) 157, 162
Act of Killing, The 236n.2 body 7, 8, 15, 17, 35-37, 48-49, 55, 56, 74, 78,
action-image 13, 26, 52, 54, 56-57, 134, 230n.10 95, 111-12, 114-16, 120-22, 124-26, 133, 137,
Adaptation 42, 233n.13 164, 172, 203, 205, 207, 212, 231n.12, 237n1,
aesthetic 45, 50, 70, 72, 74-75, 88, 89, 91, 97, 243n.10, 244n.11; see also affect; brain-body;
112, 120-21, 126, 146, 167, 169, 197, 202, 204, corporeal; embodiment; visceral
208-09, 211, 230n.5, 233n.12 body snatchers 115-16
dimension of the image 20, 120, 126, 146, Boogie Nights 237n.2
208 Boondock Saints, The 132
affect 9-10, 14-17, 22, 25-26, 32-34, 38-39, 44, Bordwell, David 7, 10-16, 20-21, 28-32, 38, 40-41,
46-49, 52, 61, 64-74, 91, 95, 99, 104, 108- 44-45, 52, 55-57, 59-63, 66, 70-71, 73-75,
28,133-34, 141-50, 154-57, 167, 169-74, 177, 182, 78-80, 83-84, 87-92, 97-99, 114, 116-19, 121-24,
186, 188-200, 204-13, 229n.4, 230n.10, 231n.12, 140, 143, 146-50, 158, 160-72, 177-79, 184-88,
232n.11, 234n.19, 237n.4, 240n.5-6, 241n.9 & 192-95, 198, 206, 210, 229n.1&3, 230n.6,
12, 243n.10; see also body; corporeal; visceral 231n.11, 233n.2&12, 235n.4, 238n.5, 242n.16,
affect theory 9, 48, 52, 207 243n.5&9
affection-image 47, 49, 230n.10 Borges, Jorge Luis 83, 157-58, 161-64, 168, 235n.1,
affective narratology see narratology 241n.13
affective neuroscience 10, 52, 113, 116, brain 8, 15, 35, 56, 71, 73, 91, 98-100, 111-16, 120,
121; see also brain; neurobiology; 122-132, 138, 164, 189, 195-96, 203-04, 208-09,
neuroscience 213, 236n.3, 237n.3, 238 n9., 239n.15, 240n.3;
cognitive-affective 66-68, 174; see also see also affective neuroscience; neurosci-
cognition ence; neurobiology
alignment 129-33, 146, 183, 197-202, 212 brain-body 37, 116, 124; see also body;
Allen, Richard 117, 238n.6, embodiment
ambiguity 55, 62-64, 68, 80, 84, 89-92, 105, 182, brain-spaces 213
186, 188, 240n.6 cinema-brain-world patterns 132
analytical fabula see fabula as screen 99-100, 123, 125, 131-32
Anderson, Joseph 13, 70, 98, 234n.2 Branigan, Edward 10, 12, 15, 55, 114, 119, 129
Andrew, Dudley 23 Brooks, Rodney 35, 187
Aristotle 9-10, 203, 230n.6 Brown, William 7, 13, 15, 17, 19, 22-23, 81, 139,
Aronofsky, Darren 111, 237n.2 193, 201-02, 207, 212-13, 238n.9
arrow of time (physics concept) 139-40, 151, 176; Bruner, Jerome 40-41, 68, 80, 131, 135, 202-03
see also irreversible time Buckland, Warren 7-8, 11, 25, 78-79, 191, 204,
art-cinema 61, 80, 87-88; see also modern(ist) 230n.6
art cinema narration 61, 80, 88 butterfly effect (physics concept) 138-39, 160
assemblage 16, 81, 192, 238n.9 Butterfly Effect, The 195, 241n.10
world-spectator-film assemblage 81, 238n.9
Azcona, Maria del Mar 148, 240n.6, 241 n9 Cahiers du cinema (film magazine) 86
Camargo, Sandy 147-49
background (as formalist concept) 19, 21, 39, Cameron, Alan 7-8, 25, 150-55, 170, 195, 242n.2
69, 76-81, 87, 89, 92, 193, 205 Caracciolo, Marco 113
Barker, Jennifer 15, 189, 211 Carroll, Noël 10, 12-13, 23, 30, 75, 98, 112, 119, 199
Barthes, Roland 153, 183, 232n.4 Cartesian 14, 36-37, 81, 99, 105, 126, 194, 201,
Bazin, André 23, 69, 104, 243n.3 232n.8; see also Descartes, René
becoming (as Deleuzian concept) 72, 102, 106, principles in cognitive science 36-38
175, 201 substance dualism 36
Bellour, Raymond 14, 43, 57, 98, 125 causal 8, 11, 13, 15-17, 19, 28, 30-31, 38, 40, 51,
53-60, 70, 75, 81, 83, 91-92, 96-97, 103-05, 112,
114, 119, 123-24, 133-36, 138-45, 152-55, 162-68,
177, 186, 190, 192-93, 195-96, 200-01, 205, 207, ecological 32, 70
238n.10, 242n.15 embedded 115, 204
causal-linear 8, 16-17, 19, 30-31, 40, 51, 57, 70, embodied 10, 14-15, 20-21, 36, 52, 91, 113-15,
75, 81, 91, 96-97, 103-04, 112, 123, 136, 138, 121, 138, 206, 208, 239n.1, 241n.11
140, 145, 152, 155, 162-68, 177, 190, 192, 196, enactive 16, 20, 38, 113, 122
200, 204-05, 207; see also chronology; in film theory/science 13-15, 20, 25, 30-31,
linear 40-41, 46, 52, 55-56, 79-80, 84, 87, 90-91,
structural multicausality 28, 163 98, 101, 112, 116-20, 124, 126, 146, 150, 158,
Cavell, Stanley 212 186, 197, 230n.10, 234n.2, 237n.4, 238n.6,
Celeste, Reni 107 244n.10
Chaos 241n.10 conceptual metaphor theory (CMT) 243n.10
Chaos Theory 241n.10 Connolly, William 128
chaos theory see complexity theory contingency 26-27, 107, 150, 152, 154, 157, 159,
Chatman, Seymour 40, 68, 70, 114 162, 242n.15
Chion, Michel 169, 171, 173 contingent universals 30, 80
Chomsky, Noam 33, 38 continuity 26-29, 31, 51, 61, 77, 81, 86, 94, 100,
chronology 8, 40-41, 49, 51, 58, 78-79, 91, 99, 102, 104, 171-72, 206, 231n.3, 240n.8
105-06, 112, 124, 129, 135, 142, 144-45, 153, editing 26-28, 61, 77, 86, 94, 206
156, 175-78, 183, 191, 192-93, 196-97, 201, 208, intensified continuity 28
210, 233n.12, 239n.11; see also causal-linear; post-continuity 28-29, 171, 231n.3
linear; temporality; time constructivism 52, 180-86
cinesthetic subject 244n.11 complex
Clark, Andy 15, 34, 113, 125, 175, 232n.6, 238n.10 narratives 7-8, 10-22, 25, 28, 42-43, 46, 49-
Clarke, Melissa 188-89, 242n.2, 243. 4 52, 54, 109, 111, 113, 121,124, 127-30, 136-37,
classical 141-47, 156-63, 155, 170, 175, 180, 197-98,
cinema 11, 13, 20-21, 27-28, 31, 41, 43, 50, 206, 208, 211, 229n.1, 234n.18&1, 235n.5,
53-81, 83, 85-92, 95-97, 104, 112, 147, 238-39n.11, 241n.9, 242n.14, 242-43n.2
160-63, 172, 194, 205, 207, 213, 229n.3, systems 12, 123-24, 138-39, 239n.3, 11
233n.12, 234n.1, 240n.6 complexity 7, 9-12, 16-18, 20, 22, 25, 29, 38-40,
Hollywood see Hollywood 45, 50-52, 54, 58, 137-46, 148-53, 155, 160-62,
narration 30, 57, 76, 87, 240n.6 170, 187, 191-93, 205-10, 241n.10; see also
narratology 11, 13, 15, 22, 26, 29, 46, 97, 139- decomplexification
41, 150-52, 155, 188-90, 196, 205, 207-10 generalized 12, 98
science 10-12, 14, 20, 25, 39-40, 53, 87, 114, restricted 11-12, 39, 53
138-40, 143-44, 151-53, 163, 207, 213 theory 10, 15, 20, 25, 52, 138-41, 150-52, 155,
cognition 9-17, 19-22, 25-26, 29-41, 46-50, 52, 54- 160, 162, 170, 206-08, 241n.10
59, 61, 65-84, 87, 89-92, 95-101, 108-31, 133-35, computational 14-15, 19, 30, 33-37, 48, 75, 91, 114,
137, 142-50, 154, 156-58, 161, 163-67, 169-77, 117-25, 128, 142, 145-46, 149, 193, 208, 239n.1,
180-213, 230n.9-10, 231n.11, 232n.6-10, 234n.2, 240n.5; see also cognition
235n.4&9, 236n.3, 237n.4, 238n.5-6&8-10, corporal 15-16, 115-16, 133-34, 204, 244n.11; see
239n.1&16, 240n.5, 241n.11, 243n.2&10 also affect; body; visceral
appraisal theory 112, 130-31 crystal-image 230-31n.10
cognitive-analytical 112, 147, 183, 200, Culler, Jonathan 134-35
231n.11 Currie, Gregory 62, 119
cognitive-affective see affect
cognitive-emotional see emotion Damasio, Antonio 35-36, 113, 121, 125-26, 131,
cognitive-formalism 11-12, 16, 21, 38, 41, 52, 200, 203, 209, 232n.8
54-55, 71, 73, 76-77, 80-81, 87, 109, 112, 114, Dark City 195
117, 128, 137, 154, 166, 177, 180, 189, 193-94, decomplexification 9, 12, 124, 142, 145, 206; see
206, 210, 213, 232n.10, 238n.9 also complexity
cognitive psychology 29, 235n.9 defamiliarization 18-22, 69-81, 193-209; see also
cognitivism 13-14, 20, 32, 34, 36, 38, 40, 58, ostranenie; refamiliarization
69-70, 75, 77-79, 87, 90, 97-99, 119, 120-21, Deleuze, Gilles 7, 12-14, 21, 29, 41-57, 61, 65-66,
124-26, 145, 149, 177, 195, 234n.2, 237n.4, 69-74, 80, 83, 95-109, 123-28, 131-34, 156-58,
240n.5 166-69, 175, 189-90, 196-97, 204, 211-12,
computational 14-15, 19, 30, 33-37, 48, 75, 91, 229n.3-4, 230n.9-10, 231n.12, 232n.8&10,
114, 117-25, 128, 142, 145-46, 149, 193, 208, 233n.12-15&17-18, 234n.1, 236n.4-6&8,
239n.1, 240n.5 237n.3&9, 238n.9, 239n.12&15
connectionist 37 Descartes, René 36, 232n.8; see also Cartesian
Index 249
determinism 11, 27, 140, 150-55 198, 206-07, 234n.2, 235n.6; see also
predetermined 67, 152-55, 197 linear onto-epistemology
dichotomy realist epistemology 14, 80, 116, 175, 188-92
classical/modern 16-17, 25, 28, 87, 106, 160, Eternal Sunshine of the Spotless Mind 7,
233n.12, 239n.17 195
cognition/affect 112 Euclidean 28, 212-13; see also spatio-temporal;
linear/non-linear 9, 13-19, 25, 28-29, 47-52, temporality; time
87, 138, 155, 158, 166, 171, 174, 207 non-Euclidean 28, 45
subject/object 39 evolution 44-45, 69-70, 76, 98, 116, 209, 233n.12,
disembodied 48-49, 113-14, 145, 173; see also 238n.9, 239n.1
embodiment as scientific approach: 98, 116, 209, 238n.9
disembodied eye 113 extended mind (thesis) 175
disengagement 38-40, 95
double disengagement 38-40 fabula 10-11, 15-22, 25, 29-30, 37-41, 46, 51, 55, 57,
Doane, Mary Ann 26-28, 95, 140, 176, 191, 59, 61-64, 67-70, 74-75, 83-85, 88, 90, 92, 94,
230n.8, 231n.1, 233n.14 97-101, 105-06, 110, 112-57, 173, 177-213, 230n.5,
documentary 18, 84-85, 93-95, 210-11, 231n.2, 231n.11-12&10, 238n.9, 244n.10-11; see also
235-36n.2 syuzhet
Duras, Marguerite 85-86, 93, 96, 207, analytical fabula 16-17, 20, 119, 128, 154, 156,
durée (Bergsonian concept) 27, 84, 167-70 181, 190-93, 201-04, 210-12
dynamic 10, 27, 37, 50, 59, 69, 102, 114-15, 124, embodied fabula 10, 16-22, 110-28, 155, 173,
128, 130, 132, 137-38, 146-47, 171-72, 183-85, 175-77, 181, 188-94, 197, 200-06, 209-13,
196, 203, 206, 209, 213, 238n.10, 239n.1 230n.5, 231n.11-12, 232n.10, 244n.10
dynamical system theory see ‘complexity feedback loops 16, 22, 31, 49, 124, 127, 138, 173,
theory’ 186-87, 200-09, 238n.10
flashback 19, 30, 54-55, 58-70, 79, 84, 100-06,
Elliott, Paul 95, 100, 119, 125, 239n.12 149, 194, 201, 234n.1, 236n.7
Elsaesser, Thomas 7-8, 10, 56, 62, 120, 127, 164- lying flashback 54-55, 58-70
65, 179, 196-98, 203, 229n.1, 234n.1, 242n.2 as psychoanalytical concept 100-06
embodiment 16, 37, 44, 113-16, 121, 125, 128, 164, Flashforward 229n.2
209; see also body; brain-body; cognition; flashforward 8, 160, 163, 173
disembodied; fabula; memory Flaxman, Gregory 36, 57, 189
embodied cognition see cognition Fodor, Jerry 33, 232n.9, 241n.11
embodied fabula see fabula formalism 11-12, 16, 21, 29, 38, 41, 52, 54-55,
embodied simulation 122 71-73, 76-77, 80-81, 87, 109, 112, 114, 117, 128,
emotion 9, 13, 15-19, 22, 26, 38-39, 47-49, 52, 137, 154, 166, 177, 180, 189, 193-94, 206, 210,
73, 93, 96, 104, 112-35, 142, 145-50, 156-57, 213, 232n.10, 235n.9, 238n.9
169, 172, 186, 188, 193-94, 197-200, 204-13, cognitive formalism see cognitive
234n.19, 238n.4&9, 240n.5-7; see also affect; neoformalism 71-73, 76, 235n.9
cognition Russian formalism 29, 71, 235n.9
cognitive appraisal theory of emotion see Fight Club 7, 111
cognition film-philosophy 10, 13-14, 17-21, 25, 41-46, 50,
cognitive-emotional 9, 131, 146, 149-50, 200 52, 70-74, 83-84, 96, 98-99, 105, 109-10, 113,
dissonances 127-28, 131, 147, 200 125-26, 158, 166-69, 188-94, 201, 206-12,
spectator responses 38, 49, 104, 108, 116-17, 233n.12, 234n.18, 238n.9, 242n.2
119, 134-35, 147, 198-99, 204 Frampton, Daniel 79, 120-121, 237n.4
encounter (as film-philosophical concept) 7,
17-18, 21, 40-41, 44, 49, 60, 74, 81, 83-109, 187, Gallagher, Shaun 91, 113, 115
190, 193-94, 204, 207-10, 229n.5, 232n.10, Gallese, Vittorio 113, 121-122, 209, 238n.7
238n.9, 242n.15 Garden of the Forking Paths, The 157
Engell, Lorenz 7, 13, 233n.13-14 Gardner, Howard 32-37, 119, 232n.9
epistemology 12, 14, 20, 29, 31, 40, 46, 51-53, 57, Ghislotti, Stefano 175, 177, 180, 195, 200, 243n.2
77-79, 84, 86-88, 90-92, 95, 97, 108, 136-37, Gillespie, Michael 46, 54-55, 58, 66
167, 169, 172, 175, 179, 180-89, 196, 198, 206-07, Glasersfeld, Ernst von 180, 184
234n.2, 235n.6 grand theory 78
constructivist epistemology 20, 56, 69, 92, Gravity 18, 210-13
164, 175, 180-84, 188-92, 201, 243n.9 Grodal, Torben 13, 113, 116, 119, 125, 234n.2,
onto-epistemological 20, 29, 31, 40, 46, 51, 238n.9
53, 57, 79, 84, 87,92, 97, 136-37, 169, 189,
Groundhog Day 157 Leviathan 18, 205, 210-13

Gunning, Tom 76-77, 231n.1, 233n.16 Lewin, Kurt 132
Leys, Ruth 47-48, 109, 112, 207
hardwired 35, 38, 116, 209, 238n.9 linear 7-31, 40-58, 61-62, 65-71, 75-84, 87,
haptic 15-16, 112, 211 90-92, 96-110, 112. 114, 119, 123-31, 135-75,
Harvard Sensory Ethnography Lab (SEL) 211 177, 179, 181, 183, 186, 189-213, 230n.10,
Hayles, Kathrine 138, 163, 197 231n.12, 233n.16&17, 234n.2, 235n.6, 238n.11,
Hiroshima mon amour 19-21, 83-86, 92-109, 240n.5&7, 242n.16; see also causal-linear;
159-60, 207, 235n.1 chronology; non-linear; temporality; time
Hitchcock, Alfred 17-18, 20, 43, 45, 54, 58-63, interpretative methods 20, 54, 62, 66, 91,
67-68, 132-34, 147, 207, 233n.14, 234n.1, 98, 155, 167, 179, 186
235n.4 linear-nonlinear dichotomy see dichotomy
Hogan, Patrick Colm 114, 116 onto-epistemology 29, 31, 40, 46, 52-53,
Hollywood 7-8, 14, 27-28, 30-31, 42, 45, 53-58, 79, 92, 97, 137, 169, 189, 198, 207, 234n.2,
65, 77-78, 87, 89, 140, 158, 168, 194, 206, 235n.6; see also epistemology
229n.1&3, 231n.1, 234n.2, 240n.7; see also paradigm of thought 84
classical cinema reasoning 54, 56-57, 66-67, 81, 194, 235n.6
classical Hollywood 7, 27, 30-31, 45, 53-56, Lola rennt (Run, Lola Run) 7, 19, 22, 137,
140, 158, 168, 174, 229n.3, 234n.2 157-74, 207, 209, 238n.11, 241n.10
hodological space 133-34, 239n.17 Lorenz, Edward 138-39
prehodological space 133-34, 239n.17 Lost Highway 78-79, 126-27, 239n.11&13
Hurley, Susan 22, 117, 121
hybrid films 7, 50, 166 Marks, Laura 15-16, 113, 189, 211, 234n.18
Martin-Jones, David 7, 50, 158, 166-73,
Iñárritu, Alejandro Gonzales 7, 22, 137, 141, 242n.15-16
148, 156 Massumi, Brian 47-52, 123, 156, 197, 200, 207,
Inception 7, 195, 232n.5 209, 229n.4, 234n.19
information-processing 13, 37, 74, 115, 120, Matrix, The 175
123-24, 145; see also cognition McLuhan, Marshall 174
Ingold, Tim 38-39 Memento 7, 19, 22, 124, 175-204, 207, 209,
Intolerance 241n.9 236n.3, 239n.11, 242n.2, 243n.5,8&10
Irréversible 51 memory 7, 21, 38, 40, 44, 80, 84-86, 90, 93,
irreversible time 27, 51, 139-40, 151-55, 164, 170, 96-109, 118, 127-31, 177-83, 191, 195-201, 236n.7,
231n.2; see also arrow of time 237n.9; see also recollection
reversible time 139-40, 151-55, 164, 170, 190 Bergson’s theory of memory 44, 99, 101-04,
It’s a Wonderful Life 162 195, 237n.9
container theory of memory 195-96
Johnson, Mark 113, 116, 243n.10 embodied theory of memory 195-201; see
Jullier, Laurent 69-70 also embodiment
narrative memory 108
Kantian 83 Proustian memory 127
Keating, Patrick 55, 147, 240n.7 ‘pure’ memory 44, 195, 237n.9
Kessler, Frank 69, 77, 87, 89 mental 8, 13-20, 25, 32-33, 38, 54, 57, 66, 71,
Kinder, Marsha 25 73-74, 78, 81, 83-84, 90, 99, 104-06, 108, 111-12,
Kiss, Miklós 79 118-19, 122, 127, 141, 177, 179, 180-81, 185-86,
Klein, Andy 176-77, 181, 191 195-96, 200, 204, 209, 243n.7
Kovács, András Bálint 7, 27, 42, 45-46, 84, 86, mental operation 112, 118
165, 206, 233n.12 mental representation 13, 16, 25, 32, 81, 118,
127, 180; see also representation
La guerre est finie (The War is Over) 87, mental schema 108, 180, 195
90-92 Merleau-Ponty, Maurice 198, 232n.8
Lakoff, George 113, 116, 243n.10 mind 12-15, 17, 32-36, 41, 47-51, 57-59, 63, 66,
L’Année dernière à Marienbad (Last Year 91, 99, 109, 111-17, 120, 122, 125-27, 130-32, 137,
at Marienbad) 85, 167 141, 149, 181-85, 192, 199-200, 203, 205, 209,
Laplace, Pierre Simon 11, 153, 230n.7 234n.3, 237n.1, 238n.6, 241n.11-12
L’avventura 86 closed state of mind (closed-minded) 57-58,
Lazarus, Richard 112 65-68, 119, 139, 151, 154, 185-86, 199
Lefebvre, Henri 170-74 computational model of mind 48, 114, 125,
Leihkörper see surrogate body 149
Index 251
extended mind thesis see extended mind neuroscience 10, 15, 21, 33-35, 52, 91, 110,
mind-game films 7-8, 127, 198, 242n.2 113-14, 116, 125, 130, 207, 209, 236n.3; see also
mind-reading theories 122-23 affective neuroscience; brain; neurobiology
mirror neuron 121-22, 125, 131, 238n.7-9 Newtonian 81, 88, 138, 150
modern(ist) 8, 10, 19, 21, 25-28, 42-45, 54, 80-81, Noë, Alva 15, 113, 121, 232n.8, 238n.6
83-109, 127, 147, 159-60, 167-68, 180, 197, Nolan, Christopher 7, 22, 175, 183, 191-95
229n.3, 230n.10, 236n.2, 239n.17; see also non-linear 7-9, 12-16, 19-22, 25-27, 29, 41-52,
art-cinema 54-55, 61, 66-67, 79-84, 87, 91, 96-101, 104-06,
modularity 117, 130, 151, 241n.11 109-10, 124, 127, 129, 135, 137-44, 149-74,
modular narratives see narrative 189-92, 205-09, 231n.12, 233n.16, 235n.6,
movement-image 8, 13, 21, 29, 42-44, 50-52, 54, 238n.10, 240n.6; see also linear
56, 72, 83, 99, 104, 111, 158, 166-67, 188-90, linear-non-linear dichotomy see
205, 230n.10, 232n.11, 233n.13-14, 234n.1, dichotomy
236n.4-5, 239n.17, 240n.8 non-linear dynamics see complexity theory
Möbius strip 126-28, 132, 204, 209 non-linear temporality 8, 150-51; see also
Mulhall, Stephen 10 temporality
Mullarkey, John 10, 13, 43-44, 80, 120, 167-68, Nuit et brouillard (Night and Fog) 85
230n.9
music 19, 48, 169, 171-73, 199, 242n.17; see also Oever, Anna van den 235n.8
sound; techno music onto-epistemological see epistemology
optional thinking 57-58, 65-66, 153-54
Nadel, Alan 29, 31, 41, 80, 232n.5 ostranenie 69, 71, 235n.8; see also defamiliari-
narrative 7-23, 25-81, 84-213, 242-43n.2, 229n.1, zation; refamiliarization
230n.6, 231n.1&11, 233-34n.18, 234n.1-2,
237n.2, 237-38n.4, 238n.9&11, 240n.5-7, Passion Fish 143-44
241n.9, 242n.14&16, 242-43n.2. perception 7 ,9, 16-19, 22, 26-32, 37-38, 44,
complex narratives see complex 54-56, 61, 68-74, 77, 79-81, 95, 100-04, 115,
forking path narratives 7, 22, 137, 157-58, 117, 119-24, 128, 140, 156, 167, 169, 175-77, 180,
161-69, 172, 230n.6, 242n.13 182, 191, 193, 196-204, 208-13, 230n.10, 234n.2,
modular narratives 7-8, 150-51, 155, 241n.11 236n.6, 237n.7, 238n.6, 241n.11, 243n.10
mosaic narratives 22, 137, 141-42, 147-49, classical sandwich of perception 22, 35, 117,
154, 241n.9 121-28, 150
narrative complexity 9-10, 22, 39, 58, embodied perception 7, 38, 149, 198
142-45, 208 linear cinematic perception 17, 55, 68,
network narratives 140-41, 241n.9 70-71, 79, 81, 119, 128, 156, 175, 193-94, 197,
narratology 22, 26, 46, 55, 97, 116, 139-41, 150-55, 200-01, 210-11, 234n.2
167 ‘natural’ perception 27-28, 72, 77
affective narratology 116 ‘pure’ perception (Bergsonian concept) 44,
classical narratology see classical 237n.9
cognitive-formalist narratology 41, 55 snapshot logic of perception 140, 163
database narratology 153 phenomenology 15, 69, 110, 113, 189
narration 8, 11, 16-17, 29, 30, 38-41, 46-47, 57, Pierrot le fou 79
61-63, 68, 70, 76, 80, 84, 87-92, 97, 104-08, Pisters, Patricia 7-8, 13, 17, 19, 21, 44, 83, 99, 11,
114, 118, 128, 130, 137, 143-45, 151, 157, 160, 113, 125, 148, 192, 205, 207-08, 241n.9
168-69, 175, 179, 181, 191-93, 197, 203, 206, 212, Plantinga, Carl 9, 13, 31, 112,119, 128, 130, 146
229n.2-3, 233n.17, 234n.1&18, 235n.5, 240n.6, post-mortem film 164, 196, 242n.2
242n.2 post-theory 13, 175
art-cinema narration see art-cinema Poulaki, Maria 8-9, 139-41, 180, 239n.11
classical narration 30, 57, 76, 87, 240n.6 Powell, Anna 12, 47, 49, 156, 189, 233n.18
contemporary narration 137 problem/solution model 74-75, 90, 92, 97, 181,
dysnarration 105-06, 130, 233n.17 213; see also question/answer model
falsifying narration 105 Protevi, John 19, 21, 123-26, 192, 208, 232n.8
unreliable narration 63, 68-69, 88, 105, 178, Psycho 132-34, 239n.17
181, 234n.1, 240n.6 psychoanalysis 30, 107, 131
naturalizing 29, 41, 54, 70, 77 Pulp Fiction 7, 129
denaturalizing 28, 59, 62, 65, 69-70, 77, 81 puzzle 7, 8, 11, 60, 75,78-79, 89-90, 118, 146, 186,
neurobiology 114, 116, 131, 238n.9; see also 188, 191, 204, 235n.4, 240n.6
affective neuroscience; brain; neuroscience puzzle films 7, 8, 11, 240n.6
neuro-image 7-8, 21, 111, 113, 209 puzzle solving 146, 186
question/answer model 75, 181; see also Short Cuts 241n.9

problem/solution model Simons, Jan 7-8, 11, 62, 137, 150, 153-56, 183,
242n.2
Rancière, Jacques 43-44, 109, 190 Sinnerbrink, Robert 75, 120-21, 146, 175-77, 186,
Rashômon 67, 175 189
Rear Window 43, 235n.4 Sixth Sense, The 129
recollection 101-07, 151, 194-95, 236-37n.8, Sliding Doors 162, 166
237n.10; see also memory Smith, Greg 13, 112, 119, 146, 199
Redfern, Nick 92, 184-87, 243n.9 Smith, Murray 13, 15, 112, 119, 122, 129-32,
refamiliarization 21, 79, 194; see also 146-50, 198, 234n.2
defamilizariation; ostranenie Snatch 237n.2
relation-image 233n.14, 234n.1 Sobchack, Vivian Carol 15-16, 113, 189, 244n.11
Renner, Karen 198-99, 243n.2 sound 86, 158, 170-74, 209, 231n.11; see also
Requiem for a Dream 111, 124 music; techno music
representation 13-16, 20, 25, 30-33, 37-39, 44, soundtrack 171-73
48, 54, 73, 81, 83-91, 94-99, 104-08, 111, 114-15, spatio-temporal 7, 20, 31, 45, 173, 201-02, 208,
117-18, 120-28, 135, 145, 154, 160, 167-69, 212-13; see also Euclidean; temporality; time
177, 180-81, 187, 203, 206, 208-10, 230n.10, Stage Fright 18, 20, 54-55, 58-81, 105, 194, 201,
232n.5, 236n.2&7, 238n.9; see also mental 207, 234n.1, 235n.7, 236n.7
representation suppressed gaps 143
anti-representational 14, 84, 106 surrogate body 244n.11
non-representational 14-15, 91, 96, 115, synchresis 173-74
238n.9 syuzhet 9, 11, 22 ,29-31, 38, 40, 79, 88-89, 105,
Resnais, Alain 17-19, 83-87, 91-99, 102, 105, 159, 124, 135, 140, 142-46, 152; see also fabula
167, 207, 235n.1
reversible see irreversible time Tan, Ed 13, 119-20, 130
rhythm 111, 120, 159, 166, 169-75, 197, 204, 209, techno 49, 158, 171-74, 197
242n.17 music 158, 171-74; see also music; sound
rhythmanalysis 170-71, 174 techno-cinematic event 49, 197
Ricœur, Paul 151 temporality 7-8, 10, 12, 15, 20, 25-27, 29-31, 37,
Rodowick, David Norman 12, 17, 23, 27, 42, 44, 40-41, 45, 50-51, 55, 61, 81, 83-86, 99-104,
50, 54, 167 107, 111, 114, 127, 134, 138-40, 142, 144-45,
Ropars, Marie-Claire 103 151-53, 155, 159, 163, 167-75, 189-93, 195,
Rushton, Richard 99 200-06, 208-10, 212-13; see also Euclidean;
Russian Ark 231n.2 spatio-temporal; time
Thirteen Conversations about One
screen 16, 21, 94, 99-100, 106, 113, 119, 123, 125, Thing 239n.11, 241n.10
131, 145, 168, 173, 176, 188 Thompson, Evan 15, 111, 114-15, 232n.8-9
brain as screen see brain Thompson, Kristin 12, 53, 59, 62-76, 202,
off-screen 94, 106 229n.1, 235n.7, 9
screen culture 21, 113, 131 time 7-8, 13-14, 21, 26-29, 31-37, 42-43, 45,
split screen 159, 168, 173, 242n.14 50-63, 67, 69, 72, 74-75, 81, 83, 86, 94,
Searle, John 112 98, 100-06, 109, 111, 114, 121, 126, 128, 132,
sensory-motor (also: sensorimotor) 13, 29, 138-42, 148, 150-53, 157-76, 181-201, 205, 212,
35,38, 43, 46, 53, 56-57, 72, 83, 91, 98-101, 229-30n.3,230-31n.10, 231n.2, 233n.12-14 &
104-05, 108-09, 115, 122, 124, 133-35, 147, 194, 17, 234n.1, 236n.4&7, 238n.10, 241n.9&13
197, 240n.8 see also chronology; Euclidean; linear;
sensory-motor capacities 13, 115 spatio-temporal, temporality
sensory-motor contingencies 91, 115 Borgesian labyrinth of time 83, 158, 161-62,
sensory-motor linkages 13, 29, 72, 83, 168
99-100, 109, 133-35, 147, 240n.8 clock time 170-73
sensory-motor schema 57, 101, 104, 108 in-between time 83, 103, 159
sensory-motor system 43, 46, 56, 104-05, Timecode 242n.14
108, 194, 197 time-image 8, 13, 21, 26, 28-29, 43-45, 50-52, 57,
sensuous 112, 120, 126 72, 83, 98, 102, 104-05, 109, 111, 128, 158, 166-
Shapiro, Lawrence 15, 113-14 67, 188-90, 194, 197, 205, 229n.3, 230n.9-10,
Shaviro, Steven 8, 15, 17, 28-29, 50, 81, 113, 156, 232n.11, 233n.13-14, 234n.1, 236n.4, 239n.17,
163, 169-71, 204, 207, 231n.12&3 240n.8, 243n.7
sheets of the past 101-03, 106-07 trauma 7, 93-94, 96, 102, 106-09, 129, 196
Index 253
Tröhler, Margit 148, 241n.9 visceral 49, 112, 117, 128-29, 189, 204; see also
Truffaut, François 58-59 affect; body; corporal
Voss, Christiane 244n.11
Unanswered Question, The 171
unreliable narration see narration Wedel, Michael 159, 163, 170-74
What a Difference a Day Makes 171
Varela, Francisco Javier 15, 36, 111, 114-15, 232n.9 Wheeler, Michael 15, 32-37, 113-4, 121, 232n.7-10,
virtual 19, 21, 27, 47, 49, 51, 67, 83, 95, 109, 117, 238n.10, 239n.1
123, 126, 153-57, 160, 162, 170-71, 188, 190, 195, White, Hayden 107
207, 213, 236n.8, 237n.3
virtual history 155-56 Yacavone, Daniel 120-21, 207
virtual narrative pathways 49, 83, 154-56,
162 Žižek, Slavoj 169
Film Culture in Transition
General Editor: Thomas Elsaesser
Thomas Elsaesser, Robert Kievit and Jan Simons (eds.)

Double Trouble: Chiem van Houweninge on Writing and Filming, 1994
isbn paperback 978 90 5356 025 9
Thomas Elsaesser, Jan Simons and Lucette Bronk (eds.)

Writing for the Medium: Television in Transition, 1994
Karel Dibbets and Bert Hogenkamp (eds.)

Film and the First World War, 1994
Warren Buckland (ed.)

The Film Spectator: From Sign to Mind, 1995
isbn paperback 978 90 5356 131 7; isbn hardcover 978 90 5356 170 6
Egil Törnqvist
Between Stage and Screen: Ingmar Bergman Directs, 1996
Thomas Elsaesser (ed.)

A Second Life: German Cinema’s First Decades, 1996
Thomas Elsaesser
Fassbinder’s Germany: History Identity Subject, 1996
Thomas Elsaesser and Kay Hoffmann (eds.)

Cinema Futures: Cain, Abel or Cable? The Screen Arts in the Digital Age, 1998
Siegfried Zielinski
Audiovisions: Cinema and Television as Entr’Actes in History, 1999
Kees Bakker (ed.)
Joris Ivens and the Documentary Context, 1999
Egil Törnqvist
Ibsen, Strindberg and the Intimate Theatre: Studies in TV Presentation, 1999
Michael Temple and James S. Williams (eds.)

The Cinema Alone: Essays on the Work of Jean-Luc Godard 1985-2000, 2000
Patricia Pisters and Catherine M. Lord (eds.)

Micropolitics of Media Culture: Reading the Rhizomes of Deleuze and
Guattari, 2001
William van der Heide

Malaysian Cinema, Asian Film: Border Crossings and National Cultures, 2002
Bernadette Kester
Film Front Weimar: Representations of the First World War in German
Films of the Weimar Period (1919-1933), 2002
Richard Allen and Malcolm Turvey (eds.)

Camera Obscura, Camera Lucida: Essays in Honor of Annette Michelson, 2003
Ivo Blom
Jean Desmet and the Early Dutch Film Trade, 2003
Alastair Phillips
City of Darkness, City of Light: Émigré Filmmakers in Paris 1929-1939, 2003
Thomas Elsaesser, Alexander Horwath and Noel King (eds.)
The Last Great American Picture Show: New Hollywood Cinema in the
1970s, 2004
isbn paperback 978 90 5356 631 2; isbn hardcover 978 905356 493 6
Thomas Elsaesser (ed.)

Harun Farocki: Working on the Sight-Lines, 2004
Kristin Thompson
Herr Lubitsch Goes to Hollywood: German and American Film after World
War I, 2005
Marijke de Valck and Malte Hagener (eds.)

Cinephilia: Movies, Love and Memory, 2005
Thomas Elsaesser
European Cinema: Face to Face with Hollywood, 2005
Michael Walker
Hitchcock’s Motifs, 2005
Nanna Verhoeff
The West in Early Cinema: After the Beginning, 2006
Anat Zanger
Film Remakes as Ritual and Disguise: From Carmen to Ripley, 2006
Wanda Strauven
The Cinema of Attractions Reloaded, 2006
Malte Hagener
Moving Forward, Looking Back: The European Avant-garde and the
Invention of Film Culture, 1919-1939, 2007
Tim Bergfelder, Sue Harris and Sarah Street

Film Architecture and the Transnational Imagination: Set Design in 1930s
European Cinema, 2007
Jan Simons
Playing the Waves: Lars von Trier’s Game Cinema, 2007
Marijke de Valck
Film Festivals: From European Geopolitics to Global Cinephilia, 2007
Asbjørn Grønstad
Transfigurations: Violence, Death, and Masculinity in American Cinema,
2008
Vinzenz Hediger and Patrick Vonderau (eds.)

Films that Work: Industrial Film and the Productivity of Media, 2009
François Albera and Maria Tortajada (eds.)

Cinema beyond Film: Media Epistemology in the Modern Era, 2010
Pasi Väliaho
Mapping the Moving Image: Gesture, Thought and Cinema circa 1900, 2010
Pietsie Feenstra
New Mythological Figures in Spanish Cinema: Dissident Bodies under
Franco, 2011
Eivind Røssaak (ed.)
Between Stillness and Motion: Film, Photography, Algorithms, 2011
Tara Forrest
Alexander Kluge: Raw Materials for the Imagination, 2011
Belén Vidal
Figuring the Past: Period Film and the Mannerist Aesthetic, 2012
isbn 978 90 8964 282 0
Bo Florin
Transition and Transformation: Victor Sjöström in Hollywood 1923-1930, 2012
isbn 978 90 8964 504 3
Erika Balsom
Exhibiting Cinema in Contemporary Art, 2013
isbn 978 90 8964 471 8
Gilles Mouëllic
Improvising Cinema, 2013
isbn 978 90 8964 4551 7
Christian Jungen
Hollywood in Canne$: The History of a Love-Hate Relationship, 2014
isbn 978 90 8964 566 1
Michael Cowan
Walter Ruttmann and the Cinema of Multiplicity: Avant-Garde Film ‒
Advertising ‒ Modernity, 2014
isbn 978 90 8964 585 2
Temenuga Trifonova
Warped Minds: Cinema and Psychopathology, 2014
isbn 978 90 8964 632 3
Christine N. Brinckmann
Color and Empathy: Essays on Two Aspects of Film, 2014
isbn 978 90 8964 656 9
François Albera and Maria Tortajada (eds.)
Cine-Dispositives: Essays in Epistemology Across Media, 2015
isbn 978 90 8964 666 8
Volker Pantenburg
Farocki/Godard: Film as Theory, 2015
isbn 978 90 8964 891 4
Paul Cuff
A Revolution for the Screen: Abel Gance’s Napoléon, 2015
isbn 978 90 8964 734 4
Scott Loren and Jörg Metelmann (eds.)

Melodrama After the Tears: New Perspectives on the Politics of Victimhood,
2015
isbn 978 90 8964 673 6
Steve Choe
Sovereign Violence: Ethics and South Korean Cinema in the New
Millennium, 2016
isbn 978 90 8964 638 5
Melis Behlil
Hollywood is Everywhere: Global Directors in the Blockbuster Era, 2016
isbn 978 90 8964 739 9
Thomas Elsaesser
Film History as Media Archaeology: Tracking Digital Cinema, 2016
isbn 978 94 6298 057 0
Michael Walker
Modern Ghost Melodramas: ‘What Lies Beneath’, 2017
isbn 978 94 6298 016 7

(Film Culture in Transition) Steffen Hven - Cinema and Narrative Complexity - Embodying The Fabula-Amsterdam University Press (2017)

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Film Culture in Transition) Steffen Hven - Cinema and Narrative Complexity - Embodying The Fabula-Amsterdam University Press (2017)

Uploaded by

Copyright:

Available Formats

FILM

Amsterdam University Press

Cover design: Kok Korpershoek, Amsterdam

isbn 978 94 6298 077 8

© S. Hven / Amsterdam University Press B.V., Amsterdam 2017

II. Cinema in the Interstices 25

III. Narrative Ambiguity in the Classical Cinema 53

IV. Modern(ist) Cinema: Logic of the Encounter 83

V. Towards the Embodied Fabula 111

VI. The Complexity of Complex Narratives 137

VII. Memento and the Embodied Fabula 175

The brilliance of Christopher Nolan’s Memento (2000) lies in its complex

(Pisters 2012) film-historically. The most detailed definitions of what shall

When it comes to narrative complexity, Poulaki states: ‘It is no longer

distinction between, say, affective and non-affective behaviours, or between

into a single, continuous action with a clearly defined beginning, middle,

psychoanalytical and semiotic studies that dominated film studies at that

traditional film-theoretically founded cognitive formalism are interested

Despite the reservations between cognitivists and Deleuzian film phi-

Thus, we should not expect the computational assumptions of cognitive

narrative cinematic domains have thus up to now coexisted peacefully

and representational roots of cognitive film studies, it becomes possible to

have produced an overtly linear understanding of classical cinema. It is

on viewing Deleuzian philosophy as working across – rather than playing

often seen on multiple platforms – such as on a laptop or a smartphone –

Since a paradigm of simplification controls classical science, by imposing a

components that in the tradition of classical narratology are thought to be

Cinema and its two Temporal Modes

storytelling of classical Hollywood serves a similar function; insofar as it,

The Cognitive Fabula and Linear Storytelling

In Narration in the Fiction Film (1985a), Bordwell analyses the processes

the task of narrative cinema, in other words, is to naturalize a counterin-

Bordwell uses the terminology of Russian formalism to describe how specta-

In order to explain the prevalence of classical cinematic norms, Bordwell

These three factors [use of informative technique, spatio-temporal

[t]he historical conditions that created the conventions of Hollywood-

Cognitive formalism is founded on a similar linear onto-epistemology that

in works that understand themselves as operating from within an ecological

representational understanding were basically inseparable insofar as ‘all

Figure 1. Cognitive Science the First Decades (Gardner 1935, 37)

illustrates precisely the opposite of what I believe to be true about the

Cognitivism may not produce allegories as obvious as those of its theoreti-

according to Wheeler (2005), is in fact dismissed by practically all cognitive

embodied perspective the fabula can no longer be seen as a representation

claim of perceptual relativism – that people from different cultural

In a similar line of reasoning, the cognitive theory of narrative comprehen-

‘play a central role in narrative comprehension’ (Bordwell 2002a, 90). This

interpret conventions by “naturalizing” them’ (49). In the process of ‘natural-

Film-Philosophy and the Challenge of Contemporary Cinema

In this book I draw upon film-philosophical insights to challenge the Bord-

itself a practice, just as much as its object. It is no more abstract than

The perhaps most important film-philosophical contribution consists in

film-philosophy reaches out to grasp the digital cinema yet-to-come, it

it is deceptively simple to read the Cinema volumes along the axes of

Smith supports his argument by referring to a series of directors and films

the ‘paralysis’ of each of these characters [Jeff in Rear Window and

According to Rancière (2006), Deleuze is unable to ‘find a visible incarnation

‘the most powerful embodiment of the time-image through-out [Cinema 2]

The time-image is an image of memory. Or rather, for Deleuze it is

This, nevertheless, does by no means discredit the ability of the concept

guilty of what he, following E.H. Gombrich, calls a ‘Hegelianism without

Fright (1950) is an example of how cinematic transgression and the rupture

II. Cinema in the Interstices 25

III. Narrative Ambiguity in the Classical Cinema 53

IV. Modern(ist) Cinema: Logic of the Encounter 83

V. Towards the Embodied Fabula 111

VI. The Complexity of Complex Narratives 137

VII. Memento and the Embodied Fabula 175

Figure 1. Cognitive Science the First Decades (Gardner 1935, 37)