University of Groningen Faculty of Arts
Shifting Complexity
Complex Storytelling in Historical Film Noir and Contemporary Mainstream Cinema
Vincent Ros
Master thesis, June 2014
1
Shifting Complexity: Complex Storytelling in Historical Film Noir and Contemporary Mainstream Cinema
Vincent Ros
s1395319
vincentros84@gmail.com
2 Table of Contents
Introduction 4
1 Defining the Complex Plot: The Viewing Experience 13
1.1 Torben Grodal's PECMA-Flow Model 14
1.2 Complications of the PECMA-Flow 18
1.3 Conventions, Cultural Training and the PECMA flow 23 1.4 Problematic Terms in the PECMA Flow Model: Disembodiment, 26
Abstraction and Immersion in the Cinematic Subject
1.5 Experience versus Cognitive Processing 28
2 Defining the Complex Plot: Structures, Techniques and Devices 30 2.1 Representational Complications versus Diegetic Complications 30
2.2 Marked versus Unmarked Complexity 32
3 Complicating Narrative Techniques in Cinema 36
3.1 Temporal Fragmentation 38
3.2 Embedding & Metalepsis 42
3.3 Alternate Timelines 45
3.4 Withholding of Plot Information 51
3.5 Narrative Unreliability 54
3
4 Narrative Complexity in Film Noir 63
The Maltese Falcon (1941) 64
The Big Sleep (1946) 70
Out of the Past (1947) 81
Stage Fright (1950) 86
5 Narrative Complexity in Contemporary Mainstream Cinema 95
Pulp Fiction (1994) 96
The Usual Suspects (1995) 107
Fight Club (1999) 122
Memento (2001) 133
Eternal Sunshine of the Spotless Mind (2004) 140
Conclusion 156
References 166
4 Introduction
Since the early 1990s, mainstream cinema has seen the proliferation of the use of
non-classical, disruptive narrative devices such as non-linear storytelling, unreliable narration and embedding, devices that were previously thought to be practically exclusive to the realms of (post)modernist literature and art cinema. The unexpected cult hit Pulp Fiction (1994) pioneered the use of non-linear storytelling in mainstream cinema, leading to countless film productions adopting this device. The Usual Suspects (1995) pioneered the use of an
unreliable narrator, combining this device with an unexpected twist ending that was
experienced as radically novel at the time. Following this film, the twist ending seems to have gained immense popularity, causing filmmakers to experiment with ever more complicated narrative structures to achieve such endings. The Sixth Sense and Fight Club, both released in 1999, capitalized on the hunger for twist narratives that had apparently been awakened in mainstream audiences by using innovative, novel strategies to mask their unreliable narration. The wave of unreliable and twist narratives further continued into the new millennium with hit films like Memento (2000), Identity (2003) and Eternal Sunshine of the Spotless Mind (2004), each of which utilized new innovations (or borrowings from art cinema) in order to deliver their unexpected endings.
5 view these new storytelling strategies as a radical departure from classical narration
(Buckland 2009: 5). Bordwell's claim that the rise of complex narrative structures in
contemporary films is merely indicative of an “intensified continuity”, an “amped up” version of what is essentially still a classical style of narration (2002a, 2002b) is relativized by two responses published in the same year. Branigan (2002) emphasizes the cognitive ambiguities that may arise from complex narrative structures, pointing out how Bordwell's assertion that forking-path narratives still follow the same rules as classical narratives fails to take these into account. Young criticizes Bordwell's oversimplification of the viewer, whose reliance on “folk-psychology” makes him “by nature quite shaky in [...] encounters with ambiguity, resistant to understanding things over time and form multiple, perhaps conflicting,
perspectives, and desirous of solving problems by arriving at a happy ending” (2002: 115-6), and points out that complex narratives may be designed to challenge and expand, rather than confirm, our cognitive limits.
6 While the rise of complicating narrative devices in mainstream cinema certainly demands this kind of scientific attention, the common problem with most contributions to this field is that they are attempts at classification, primarily aimed at placing complex, “puzzle” (Buckland 2009), “mind-game” (Elsaesser 2009) or “forking-path” (Bordwell 2002) films in a separate category from classical film narratives. While suggesting the need for a separate category of film narratives, they supply little theoretical foundation to justify the need or use of a separate category for complex film narratives. These contributions “avoid the difficulty and trouble of delivering a clear-cut explanation for their argument, treating their subject safely by
explaining the trend’s media archaeological, that is industrial and technical, context, or by simply providing extensive taxonomies of complex storytelling techniques” (Kiss 2012: 237-256). While these taxonomies may be initially helpful in establishing and defending the topic of complex narratives in mainstream cinema as a relevant field of study, they are careful not to explore difficult and fundamental questions such as how the structural properties of a film can manage to produce the effect of complexity.
7 would try to make use of these universally shared processes; films are, after all, intended for broad audiences. Film narratives may possess certain structural qualities that are likely to elicit an experience of complexity in a viewer because they problematize universally shared processes of cognition. One of the central aims of this thesis will be to construct a working definition of narrative complexity based on this cognitive, experiential point-of-view. First, we will use Torben Grodal’s concept of the PECMA-flow in order to gain an understanding of how certain elements of a film text can block the cognition and comprehension of a viewer. Then, building on Grodal’s theory, we will argue how PECMA-flow blocking moments may contribute to an experience of complexity. Finally, we will attempt to identify the disruptive narrative strategies that may cause PECMA-flow blocks to arise and can therefore be said to be indicative of complex narration. We will, in other words, attempt to bridge the gap between the cognitivist and structuralist perspectives on complex narration by attempting to identify the cognitive reasons why certain structural elements of a film text elicit an
experience of complexity.
Another problem with previous contributions to the study of contemporary complex cinema is that they generally fail to take into account the historical dimensions of complex storytelling in cinema. Art cinema has always problematized issues of representation and interpretation, and the fact that the new wave of postwar European art cinema developed in close relation to the modernist novel led to the development of some extremely complicated, even paradoxical, plot structures in films like Hiroshima, mon Amour (1953), Last Year at
Marienbad (1961) and La Jetée (1962). To disregard the influence of these films on
contemporary mainstream cinema is to think of mainstream film production as operating in a separate universe from independent art cinema, which is obviously not the case. Apart from the fact that filmmakers frequently look for inspiration in classic art cinema (the director of
8 (Wells 1995)), independent filmmakers occasionally find their way to mainstream popularity, bringing new innovations into mainstream cinema. Filmmakers like Quentin Tarantino and Christopher Nolan are both examples of this development; starting out as independent
filmmakers, both found acceptance into the realm of mainstream cinema, where their unusual narrative styles became widely imitated. Even within the realm of mainstream cinema, specific genres have always dealt with fairly complex diegetic situations, which gradually led to the development of ever more complex plots. One such genre is the film noir, where the viewer, tasked with solving a complicated murder mystery, is frequently confounded by the extensive withholding of essential plot information and an immoral, possibly unreliable, protagonist-narrator. Noirs like The Big Sleep (1946), Stage Fright (1950) and Where the
Sidewalk Ends (1950) stretch classical Hollywood narration to (or perhaps beyond) its very
limits with their intricate plot structures and extensive use of internal focalization.
If we wish to claim that the narrative complexity seen in contemporary mainstream cinema is more than an 'intensified' form of the classical continuity system1, we have to
explain what makes contemporary complex mainstream films fundamentally different from their historical predecessors. Historical film noir films provide an excellent basis for such a comparison. Like most contemporary puzzle films, these films were intended to appeal to broad audiences; at the same time, they provide us with examples of some of the most
intricate narratives ever produced by the studio system. If any group of films is to demonstrate
1 The classical continuity system was the dominant form of narration in Hollywood film productions
9 how far narrative complexity can be stretched while still obeying the rules of the classical continuity system, this is it.
A comparison between historical film noir and contemporary complex mainstream films could also clarify the possible historical connection between the two groups of films. As Eva Laass notes, contemporary complex mainstream films are frequently labeled as
“neo-noirs” (201), and indeed many of them exhibit a number of key characteristics of the film
noir, such as the dark urban setting, the use of low-key lighting, the crime plot, and the lonesome, frequently unreliable protagonist. Although historical film noir is generally seen as a style of filmmaking that reached its peak between the early 1940s until the beginning of the 1950s (most accounts start with either The Maltese Falcon (1941) or Citizen Kane (1940) and end with Touch of Evil (1950)), several scholars of film noir have argued that since noir is a style, rather than a genre, it is unrestricted by historical boundaries (Schwarz 2005: ix, Röwekamp 2003: 26). These scholars suggest that the noir style was reinvented and imitated by filmmakers from the early 1960s onwards, pointing at films like Psycho (1960), Chinatown (1974), Taxi Driver (1976) and The Usual Suspects (1995) as examples of neo-noir (Schwarz 2005: xii). If it is true that contemporary complex mainstream films must be seen as neo-noirs, this would grant a great deal of validity to Bordwell's claim that their supposed
10
Aims & Methods
This thesis seeks to contribute to the study of complex narratives in contemporary mainstream cinema by performing such an investigation. It departs from the question of whether the narrative complexity of historical film noir films produced between 1940 and 1950 can be seen as the historical predecessor of the narrative complexity that currently abounds in mainstream cinema. We will attempt to answer this question by performing a comparative analysis of a number of historical style-defining film noirs and a number of contemporary complex mainstream films. With each film we discuss, we will attempt to define what,
precisely, contributes to the complexity of its narrative and to what extent it remains within or breaks the boundaries of classical film narration (as defined by Bordwell, 1985: 156-66). In order to perform a fair comparison, the first step that must be taken is to construct a working definition of narrative complexity in film. Most descriptions of narrative complexity provided by film scholars working in this field suffer from being either vague descriptions of the experience of complexity (words like 'mind-boggling', 'puzzling' or the 'mind-game' are frequently used) or provide lists of structural devices that may be used to achieve complexity in narratives, without diving deep into the question why these devices cause a sense of complexity.
11 the use of the same complicating device may act as a complicating factor in one film narrative while having little effect on another by pointing at several conventional strategies films use to mark devices and reduce their impact. It is worth noting at this point that these discussions will be based purely on an interpretation of Torben Grodal's PECMA-flow theory, rather than on empirical data; in other words, they are suggestions of what I believe goes on in the brain of a viewer who is confronted with these devices. Further empirical research beyond the scope of this thesis would be necessary to confirm these hypothetical suggestions.
Goals & Preliminary Hypothesis
This thesis aims to provide a functional, working definition of narrative complexity in cinema based on the combined perspectives of cognitive film studies and narratology, and to apply this definition to a comparative study of historical film noir and contemporary complex mainstream films. While this research aims to provide a definition of narrative complexity that is functional for the purpose of its comparative study, it by no means claims to provide a complete or final explanation of this intricate and endlessly stimulating phenomenon.
Furthermore, it should be noted that the definition provided here builds extensively on the work of cognitive film scholar Torben Grodal, whose notion of PECMA-flow
blocking (2009) has provided the fundamental basis for this thesis. This thesis merely seeks to show how this theory can be applied to the hotly debated topic of narrative complexity in contemporary mainstream cinema.
12 between these two groups of films and establish whether the apparent complexity of
contemporary complex mainstream films is merely an 'intensified' form of the complexity already present in film noir of the 1940s and '50s and must therefore be seen as a direct continuation of the conventions of classical narration or if this latter set of films actually exceeds the boundaries of classical film narration (as defined by David Bordwell). I will start by addressing narrative complexity as a cognitive effect, seeking to explain through Torben Grodal's cognitive film theory how narrative devices can block a viewer's cognitive flow and cause a sense of complexity to arise. I will then move on to a discussion of narrative devices that are frequently used to achieve this effect. With each device I discuss, I will hypothesize how it could be used to cause an effect of complexity to arise as well as the methods that could be used to prevent this from happening. Next, I will formulate a working definition of the complex film plot that combines within it the cognitive experience of complexity and the structural, film-textual properties that may trigger this experience. I will then use this
13 1. Defining the Complex Plot: The Viewing Experience
Narrative complexity should, first and foremost, be understood as a label we put on a specific type of viewing experience. A film may have certain structural properties that elicit this experience, but it is only from the actual act of viewing that the sense of complexity arises. It is therefore important that a definition of the complex film plot explains how certain structural properties of film plots may elicit this experience of complexity. Torben Grodal's cognitive approach to film (2009) may offer a good basis for this project. While his psychological approach may not be the only or the best way to get a grip on the complicated matter of the viewer's experience, it offers a way of understanding that takes into account both the
subjectivity of the individual viewer and the aspects of the viewing experience that are shared collectively. Furthermore, it takes the notion of the intentionality of experience – a notion that was at the heart of the phenomenology of Husserl and Merleau-Ponty - as its starting point, while supplying strengthening evidence for it from research done in the field of empirical cognitive psychology.
Although Torben Grodal’s model may help us formulate a definition of narrative complexity in film that is based on clearly delineated properties of a film plot, it suffers from various theoretical problems which I have tried to address at the best of my abilities by nuancing and adapting Grodal’s claims. Grodal’s model clearly runs the risk of being
14 and his lack of attention for the ways in which cultural conventions do not only influence how a viewer deals with a narrative disruption, but also determine to a large extent whether a narrative device is experienced as disruptive in the first place. By adapting and nuancing Grodal’s theory on these points, I have tried to shift it from a rigid, operational model, to a model that shows awareness of the immense influence of subjectivity and cultural
conventions on the interpretative process.
1.1 Torben Grodal's PECMA-Flow Model
Grodal's theory is based on the premise that there are biologically hardwired structures in the human brain which shape, to a great extent, the way in which we watch films. He suggests that the viewing experience must be understood as a cognitive flow he calls the PECMA flow: Perception, Emotion, Cognition, Motor Action (Grodal 2009: 145-157). This model describes how the experience and comprehension of a film "relies on a flow that follows the brain's general architecture, a flow from perception (ear and eye), via visual and acoustic brain structures, association areas and frontal brain structures to motor action (150). It is the same flow we use in our day-to-day perception, comprehension and interaction with the real world. According to Grodal, we experience films in much the same way, the only vital difference being the fact that our frontal cortex is capable of suppressing actual motor action in situations where the brain is aware of the fact that it is experiencing a representation: "In mental
simulations such as film viewing, the modeling of actions in the premotor areas, of course, does not lead to actual motor actions; the motor centers are only resonating, not executing" (150).
15 Perception stage). The initial filtering process takes place almost simultaneously with the movement of our eyes, for it is done in a part of the brain that is directly connected to the eyes (the visual cortex). Parts of this process are unaffected by cultural training, as is evidenced by our culturally independent susceptibility to visual illusions. When it comes to the
identification and emotional labelling of visual information, however, personal experience and cultural training play a significant role (147). The brain attempts to match incoming visual data to known schema and images that are stored in adjacent association areas located in the temporal and parietal lobes (148). Each schema and image is stored with an emotional ‘tag’ attached to it, triggering a positive emotional response when incoming visual data can be matched to it (149). These tags trigger basic, deeply seated emotions (such as fear, anger, joy or erotic desire) and determine the initial emotional valence the brain attributes to the
perceptual data (ibid.). Emotions originate from the oldest, deepest parts of our brain, the limbic system (or the old “mammalian” brain), several nearby areas in the brain stem (the “reptilian” part of the brain) and cortical areas in front of the limbic system (Grodal referring to Damasio 1994, 1999; Panksepp 1998; Ledoux 1998; Kringelbach and Rolls 2004, 147) and affect every stage of the PECMA flow.
16 individual's cultural background comes strongly into play. While emotions tend to have a core that is instinctive and unmediated by culture, the way and extent to which we access or
suppress our emotions, as well as the ways in which we manage our associations, our ethical choices and our planning of complex actions are all heavily influenced by our individual cultural backgrounds and individual dispositions.
17 out the goal or sub-goal it considers urgent, intentionally guiding its perception to seek out the relevant information to construct the next plan of action, thereby effectively restarting the entire PECMA flow (ibid., 150-151).
Films that feature active human agents, follow a more or less chronological
representation of events and adhere to a seemingly realistic representation of space allow for an unproblematic PECMA flow. Our brains have evolved to their current state through a process of natural selection. Consequently, they are specialized in pursuing goals that have direct relevance on the survival of our species, ranging from simple goals such as the search for nutrition and the evasion of predators to more complex goals such as the selection and courting of a sexual partner or the safeguarding of our offspring. When a film narrative features human agents that are engaged in the active pursuit of concrete goals, it allows for unproblematic embodiment and closely approaches the practical, day-to-day problem solving and our brains are built for. For example, a horror film like Aliens (1986) focuses on the very concrete main goal of survival of the protagonists, a goal which can be subdivided in
countless sub-goals (such as evade the aliens, seek shelter, protect the child, etc.). Its protagonists are humans we can easily emphasize with and its representation of time and space are both highly conventional (events are presented chronologically, space is presented as continuous). The PECMA flow we experience when watching such a film is
unproblematic: from the visual information we receive, we immediately filter out those details that could be relevant to the main goal of the protagonists' survival (a door, a gun), from this, we identify sub-goals that must be achieved in order to ensure the main goal of the
protagonist's survival, we plan out motor actions and may even subconsciously tense our muscles or increase our heart rate as we get ready to perform these actions ourselves. If the representation is convincing enough, it is only the awareness that we are watching a
18 predator jumps at our protagonist. When a sub-goal is met, this results in momentary
relaxation of our motor systems, only to start up the next PECMA cycle; we immediately reassess the new situation and begin to seek out new visual details that may matter to the protagonist's survival.
1.2 Complications of the PECMA-Flow
Narrative strategies such as the withholding of plot information or the fragmented presentation of time complicate and obstruct the PECMA flow. When something seems relevant to a plot, but cannot be immediately found or understood, cognition is not immediate and, as a result, we remain indecisive on the motor action we (or the characters) should take until the complication is resolved. For example, when information that is withheld is made to seem particularly relevant to the plot, it may cue us to seek out the missing information among all the new information we receive. Likewise, a fragmented representation of story time may force us to keep in mind events that were presented earlier and test them against newly received information in order to puzzle together a consistent timeline, thereby straining both our memory and cognition. Every such complication has the same effect of further complicating the cognition stage of the PECMA flow.
Grodal distinguishes between transient, embodied meaning and permanent,
19 that must be undertaken to achieve them (Grodal 2009: 205, 208). It is, however, also
transient; a viewer does not need to construct additional layers of symbolic, abstract or metaphoric meaning in order to construct the concrete relevance of the narrative situation. Although the viewer is still free to construct such additional layers of meaning2, the fact that
the concrete relevance of the narrative situation appears to be evident does not clearly encourage such behavior. By contrast, art films tend to purposefully complicate or block the PECMA flow, practically forcing their viewers look for associations that transcend the level of concrete goals and actions. Presenting viewers with events, objects, characters or narrative structures that cannot be readily understood, they arrest the PECMA flow at the point of cognition. When faced with ambiguities, inconsistencies or paradoxes, the mind cannot readily construct concrete meaning, but has to rely on association instead. In these cases, the mind calls upon abstract, higher level categories in its attempt to construct the relevance of the narrative. These categories are, by Grodal's definition, not embodied3 (ibid., 209) and lead to the construction of levels of meaning that are more permanent (in the sense that they transcend the transient, immediate relevance of concrete goals and actions) (210-16).
When we encounter paradoxes or counter-intuitive elements that cannot be resolved, we become aware of the possibility that the information a film is communicating to us may transcend the concrete relevance of goals, actions and spaces and immediately begin hypothesizing how it may deal with different levels of meaning, such as abstract, thematic, symbolic, or metaphorical meanings. According to Grodal, this may cause an over-activation of association areas that is similar in nature to a spiritual experience: “paradoxes and
counterintuitive events […] arrest the PECMA flow and over-activate the association areas,”
2 Of all the viewers that watch an action blockbuster, only a select minority of trained specialists tend to
engage in sociological, cultural or thematic readings; for most viewers, the concrete level of meaning provides sufficient relevance.
20 which may trigger “experiences of deep significance” (Grodal 2009, 149-150). At the same time, we may become aware of the constructed nature of the art work and begin to form hypotheses and expectations about its structure. We look for ways to establish the relevance of what we are witnessing through associations and abstractions. How we do this exactly is highly dependent on the contextual knowledge that is readily available to us; personal experience and cultural training determine how this network of possible associations is structured.
Grodal attempts to link the blocking of the PECMA flow to a qualitative definition of art cinema (211-14) by suggesting that mainstream films tend to avoid PECMA blocks and consequently revolve around concrete, transient meaning while art films tend to utilize PECMA blocks to refer to more permanent, abstract levels of meaning. Against this view, I would argue that even the most straightforward crime plot relies on occasional blocking of the PECMA flow to create viewer involvement. As Meir Sternberg has argued, narrative gaps form the driving force behind each narrative (2001: 117)4. He distinguishes among three
narrative ‘universals’: suspense, curiosity and surprise, each of which relies on a different way of presenting informational gaps. Suspense relies on clearly marked missing information about the future; for example, when we see a character racing to disarm a ticking bomb, but we do not know if he will succeed in time. Curiosity relies on clearly marked missing
information in the past; for example, when we see how someone is murdered at the beginning of a film, but the identity of the killer is hidden. Surprise, finally, relies on unmarked missing information (usually also located in the past) which unexpectedly comes to light and
4 It must be noted that Grodal focuses on disruptions that cannot be resolved using information provided
21 drastically alters our understanding of the entire situation (ibid.). Many contemporary
complex films rely on a combination of all three effects; for example, the narrative of The
Machinist (2004) creates curiosity by clearly marking missing information about a possible
conspiracy against the protagonist, Trevor, it creates suspense by conjuring up such questions as “will Trevor succeed in solving the mystery?” and finally creates surprise by unexpectedly revealing the unmarked missing information that Trevor is guilty of killing a child and his mother in a car accident. This missing information pulls viewers into the narrative; the marked gaps invite the viewer to engage in a game of problem solving and hypothesis formation, while the unmarked gap allows the narrative to deliver its surprise twist ending.
22 has come to claim a correlation between the use of PECMA blocks in a film and its
qualification as art or mainstream cinema.
To summarize, Grodal's PECMA flow offers a way of understanding the viewing experience as a cognitive flow that is biologically hardwired yet at the same time influenced by the individual viewer's cultural training and disposition. Grodal uses the PECMA flow to distinguish between films that allow for an unproblematic embodied experience and
23 complexity to an experiential-phenomenological definition. Before we proceed, however, we must first look in more detail at some of the problematic elements of Grodal’s PECMA flow theory that may require revision or elaboration. Following this, we will look in more detail at the structural techniques and devices that are used to cause obstructions of the PECMA flow cycle, in an attempt to construct a complimentary definition of the complex film plot from a structural perspective.
1.3 Conventions, Cultural Training and the PECMA flow
At first glance, the PECMA flow model may seem to suggest a high degree of universality in the way we experience films. It is important to keep in mind, however, that conventions and cultural training play a large role, especially in the Cognition phase of the flow. We have hypothesized that PECMA flow blocks may be caused by disruptive techniques such as the withholding of plot information, the fragmented presentation of time and space, the
embedding of story layers, or the suggestion of unreliability on the part of the narrator or protagonist. It is likely, however, that the degree to which such techniques are experienced as disruptive is highly dependent on the viewer's familiarity with cinematic conventions. We know that disruptive techniques may, when used frequently by filmmakers, gradually become conventional. A technique like the marked flashback which may have been experienced as disruptive when it was first used is now considered conventional. The degree to which a technique is experienced as disruptive appears to vary depending on a viewer's familiarity with it, which, in turn, is heavily influenced by cultural training. Rather than following Grodal's suggestion that an uninterrupted PECMA flow is dependent on a film's close
24 representation of the real world. Cinematic techniques do not necessarily cause PECMA blocks when they are part of a conventional form of cinematic language that a viewer is familiar with. Grodal leaves room for this in his discussion of the PECMA flow. In his discussion of the Perception phase, he mentions how previous experience may influence both the identification and the value that is attached to perceptual information (147-8):
When it comes to the identification and emotional labeling of visual information and the development of associational networks, culture and personal experiences play a much larger part. The identification of course depends on previous experience, but so too, to a great extent, does the emotional valence attached to the information [...]
In his subsequent discussion of the "top-down" cognitive processes that direct a viewer's perception and attention, cultural training plays an even greater role (152):
The procedure of matching input to stored memories and schemata may be described as a top-down procedure, insofar as matching presupposes previous learning, as does the evaluation of possible actions and the evaluation of a given protagonist's coping potential. The bottom-up [PECMA] flow is therefore supported and constrained both by the architecture of the brain and by previous learning.
Cultural training does, in other words, have a significant impact on the PECMA-flow
process5. While the flow of perception to cognition and motor action depends on a number of hard-wired brain functions, it can be directed and constrained by previously acquired
knowledge and emotion in the Perception and Cognition phases. Emotion is always mixed
25 with cognition and knowledge. Our instinctive emotional reaction to run at the sight of a predator may be inhibited by our knowledge that the cage that holds it is strong enough to keep it from harming us. Likewise, our brain's ability to learn and adapt allows us to acquire new cognitive schemata that help us to quickly make sense of structurally complex situations; for example, we may find it easy to navigate through a busy city district because we have traversed it before. This remarkable adaptability of the human brain depends on a principle that known as neuroplasticity. Neuroscientific research has shown that the synaptic
connections in our brains can change within the space of milliseconds and entirely new networks may be constructed within the space of hours (Shaw et al. 2001: 6-10). Both in narrative and in real life, situations that are initially experienced as complex, frightening or perplexing may quickly cease to be experienced as such as a result of our familiarity with them.
In order to demonstrate how the disruptive quality we ascribe to certain cinematic techniques is more dependent on our familiarity with them than on the extent to which they disrupt an accurate representation of our everyday experience of reality we need only look at the unnatural ways in which classical film narratives represent reality. In classical Hollywood cinema time is highly condensed, characters represent certain stereotypes or archetypes, good and evil are clearly delineated, and characters are rarely shown performing repetitive or uninteresting tasks. Classical Hollywood narration cannot be seen as an accurate or realistic representation of reality, but rather, must be seen as a highly conventionalized style of representation (Bordwell 1985: 156-66). Competent film viewers have been so trained so extensively in its conventions that the style has taken on the quality of a 'natural'
26 1.4 Problematic Terms in the PECMA Flow Model: Disembodiment, Abstraction and Immersion in the Cinematic Subject
A potential problem with Grodal's theory lies in his sharp distinction between, on the one hand, the concrete, embodied experience of films that support the PECMA flow cycle and, on the other hand, films that block the viewer's embodied experience by calling up abstract concepts which, in Grodal's view can be called "disembodied" (209). To link abstraction to disembodiment is problematic, for, as both phenomenologists such as Merleau-Ponty (Moran 2000: 402-403) and cognitivists such as Raymond Gibbs (1999: 83-86) have convincingly demonstrated, embodied categories form the foundations upon which many of our more abstract categories are built. To draw a sharp distinction between, on the one hand, concrete, embodied categories and, on the other hand, abstract, disembodied categories is to ignore the fundamental way in which the first are dependent on the latter. Rather than stating that a PECMA block triggers a sense of "disembodiment", it is therefore more accurate to suggest that a PECMA block disrupts the viewer's embodied immersion in the immediate narrative moment or action of the film and forces the viewer to consider the film narrative on a more abstract, structural level. The embodied, immersed viewing position is thus severed and substituted, if only momentarily, with a position of more distanced and abstract consideration. Therefore, when we use the term 'disembodiment' in the context of the PECMA flow, what we mean to indicate is a momentary disruption of the embodied connection between the viewer and the cinematic subject. We must be careful not to discount the way in which human experience and cognition are, at their most fundamental levels, always embodied.
27 connection with the cinematic subject and force the viewer to momentarily distance himself from the diegetic world. It is important to note, however, that such distancing does not necessarily have to be experienced as immersion-breaking. In films, just as in real life, characters are frequently confronted with puzzling or perplexing situations which, either literally or figuratively, block the flow of their action. The subject through whom we
experience the diegetic world of a film is generally not only an actor in that world, but also a thinker about that world. In our day-to-day interaction with the real world we, too, are frequently confronted with perplexing, confusing or complicated situations which force us to take a step back from our active engagement in the world in order to consider the world from a more abstract or distanced position. The way in which we interact with the world when we drive a car is fundamentally different from the way in which we engage with the world when we analyze the political situation of the Middle East. Likewise, cinematic subjects may engage with their diegetic worlds in a very active, engaged, goal-oriented manner as Grodal describes, but may also be forced to consider their world from a more distanced position. In narrative cinema, moments where the viewer is forced to fundamentally (re)consider the structure or presentation of a plot frequently coincide with moments where the protagonist is forced to (re)consider his beliefs / hypotheses about the structure of the diegetic world he inhabits or the way in which information is presented to him. Viewers often hypothesize alongside cinematic characters about the possible significance of withheld plot information or the trustworthiness of homodiegetic narrators. When a PECMA flow block in the viewer thus coincides with a moment of profound (re)consideration in the cinematic subject, the
28 subject is temporarily severed and we may speak of a disruption of the viewer’s embodied immersion in the diegetic world.
1.5 Experience versus Cognitive Processing
Described from a phenomenological perspective, narrative complexity is a label we put on a unique type of experience that arises when our consciousness is confronted with a narrative that features an arrangement of plot elements that is somehow perplexing, confusing, or puzzling. From a cognitive viewpoint such as that of Grodal, narrative complexity is a specific frame we assign to narratives that complicate the viewer's embodiment in the
immediate narrative situation and require a great(er) extent of cognitive investment on the part of the viewer to process. These two definitions may overlap, but they are certainly not
identical. The main problem with the cognitive approach that we have adopted is that it assumes that an experience of complexity arises out of complications in the cognitive process. We may ask ourselves, however, to what extent a viewer is aware of the complications that take place in his mental processing of a film narrative. While Grodal's PECMA flow model convincingly shows how complications in a film narrative may lead to blocks in the flow of our cognition, it disregards the question of how much of this flow, or the blocking thereof, is consciously experienced by the viewer. This is problematic because before a viewer can decide that a narrative is complex, the viewer must first become consciously aware of the cognitive strain it puts on him.
Although it is difficult (perhaps even impossible) to determine when the processing of narrative information becomes conscious, we may advance the following hypothesis: when a PECMA flow block occurs, the viewer is strongly encouraged to rely on processes of
29 and relevance of the narrative information presented before him. While many of these
30 2. Defining the Complex Plot: Structures, Techniques and Devices
Now that we have established how narrative complexity arises out of a viewing experience that is frustrated by complications that block the PECMA flow cycle, thereby complicating the viewer's embodiment in the immediate action of the narrative and cueing the mind to explore subjective associations, our next task is to discuss the various narrative techniques and devices that film plots use to achieve this effect and how they achieve it. Once we are able to link the structural properties of film narratives to the experience of complexity, we can begin to compare the use of these structural properties in contemporary puzzle films with their use in historical film noir and discover to what extent the use of complicating narrative
techniques has changed over time. This will allow us to test the possibility of a direct family relation between film noir and contemporary puzzle films.
2.1 Representational Complications versus Diegetic Complications
Before we move on to our discussion of each individual complicating device, we must first make the important distinction between the two layers of a narrative at which complicating devices may operate. These two layers are what Bordwell (following the Russian formalists) distinguishes as the fabula (story) and the syuzhet (plot) (Bordwell, 1985: 49-53). Simply speaking, the fabula, or story, is a sequence of events that take place within a storyworld (diegesis), while the syuzhet, or plot, is the manner in which these events are presented to the viewer. While this distinction is so widely accepted by literary scholars and film scholars that it may seem too obvious to mention, it is still common source of confusion. Take, for
31 claiming this, appears to disregard the important difference between complications at the level of the storyworld (characters are involved in very complicated love dramas and family feuds, suffer memory loss or deceive each other) and the level of presentation (which is very
straightforward and chronological, albeit serial). To draw on an example closer to the focus of this thesis, we could name the 1946 Hawks and Faulkner film The Big Sleep; a film which is intricately complicated on the diegetic level while following an almost completely classical style of narration. This is markedly different from a neo-noir like The Machinist (2004), which presents a fairly straightforward story of guilt and trauma in a plot that makes extensive use of the technique of unreliable, subjective narration. The frequent confusion is
understandable; even in daily talk about films, we constantly confuse the two layers. When asked about a difficult to follow murder mystery, we may say that we found the plot too complex, when in fact the events within the diegetic world were highly complex while the plot was a very straightforward chronological account of a detective unraveling a mystery. If we wish to make meaningful statements about the possible family relation between
historical film noir and contemporary complex mainstream films, it is imperative that we keep the distinction between these two layers very clear. Therefore, I propose to adhere to the following definitions:
Representational complications are techniques that are used to complicate the presentation of the narrative. For example, the fragmented presentation of time is a purely representational technique that can potentially lead to an experience of complexity even in telling a very simple story.
32 delicate game of love and betrayal. These complications arise from narratives that feature multiple protagonists, an abundance of characters, difficult personal, moral or political dilemmas, logical puzzles, multi-layered conspiracies or intricate webs of interpersonal relations.
Now that we have made this distinction clear, we must discuss one more important factor in the use of complicating narrative techniques: marking.
2.2 Marked versus Unmarked Complexity
Before we embark on our discussion of the various complicating narrative techniques and devices that may be used to render a film more complex, we need to take into account the fact that the extent to which the use of a potentially complicating technique is openly
acknowledged by a film text can greatly strengthen or soften its potentially confusing effect. In David Bordwell's seminal article Film Futures (2002a), he argues against the idea that contemporary forking path or “parallel worlds” films (films that explore multiple timelines) are highly complex narratives and identifies seven conventions that are frequently used to make such potentially confusing experimental narrative structures “cognitively manageable” (91). Each of the conventions he names are, in fact, conventions of classical Hollywood style, ranging from chronologically ordered narration (“all forks are [in themselves, ed.] linear” (92)) to the use of traditional “cohesion devices” such as “appointments and deadlines” (95) and the archetypical resolution of action at the end of the narrative (“all forks are not equal, the last one taken, or completed, is the least hypothetical one” (100)). Yet even more
33 still looming over contemporary mainstream cinema is the second convention he mentions: signposting (2).
Signposting is the overt marking of potentially confusing transitions from one diegetic layer to another or from one diegetic time to another. The overt signposting of transitions is the standard rule of classical Hollywood narration. While Bordwell applies the term only to the overt marking of transitions between alternate timelines or layers, I suggest that every complicating narrative device may be implemented in either a marked or an unmarked way, and that in nine out of ten cases, the unmarked use is the more confusing.
34 a boring day at work. This final version clearly has the least marking of all three versions I suggested. It is also by far the most confusing; even though the second scene returns us to everyday normality, the relative place of the surreal sequence within the possibly multi-layered diegetic world of the film remains a complete mystery to us. This last example is one of a kind seldom seen in mainstream cinema, but frequently seen in art cinema such as the films of David Lynch6. The first version I presented used to be the standard way of handling dream sequences in classical Hollywood style. The second version I presented, which temporarily maximizes the confusing effect of a complicating narrative device such as
embedding by withholding important information (in this case, information about the relative place of the surreal sequence with regard to other diegetic layers), only to revert the confusing effect via a plot revelation later on, is highly typical of contemporary complex mainstream cinema.
An important thing to note about the example we've just discussed is that the marking relies on a convention. The blurring of the screen and the non-diegetic sound of a harp are both conventional codings for a dream sequence. In effect, they rely on their viewer to be familiar with these conventions, which, in turn, may account for significant differences in the viewing experiences of individuals when confronted with marked complexity. Some of the most frequently used conventions do, however, seem to refer to real-life perceptual
experiences. For example, a blurred screen may have become such a pervasive convention simply because it corresponds to the perceptual experience of blurring vision, which, in our real-life experience, would generally precede the loss of consciousness and the transition to a dream or hallucination.
6 Although the films of David Lynch utilize many elements of the classical continuity system, the way these
35 Furthermore, not all marking relies on conventions. A good example of the use of unconventional marking to counter-balance confusion is found in Christopher Nolan's
Inception (2010). It was one of the first films targeted at a mainstream audience to experiment
with multiple layers of embedded storylines that metaleptically affect one another. To counter-balance this potentially confusing structure, filmmaker Nolan used two marking techniques. First, he chose to mark each diegetic layer very clearly by giving each layer a completely different setting: the top layer (or what we take for the top layer) takes place in an airplane, one layer below this they are in a car chase through a rain-soaked city, one layer below this they are in an upper-class hotel, one layer further down they are at an army base in the snow, one layer below this, the hotel room where the protagonist lost his wife to suicide, one layer lower they are in a ruined city, and finally, on the bottom layer, they are in a post-apocalyptic wasteland. Although the transitions to the bottom-most layers may become somewhat confusing, especially near the end of the film, the color coding of the top three dream layers is highly effective. As the scenes have us constantly switching from the car chase to the hotel, to the snow-base and back, only the fact that each of these environments looks completely different allows us to easily keep track of the action. The second marking device Nolan uses is the machine that the dreamers use to enter a deeper layer of their
collective dream. Every time they activate this device, the viewer is alerted to the fact that we are entering another layer. This furthermore naturalizes the embedding structure from within the diegesis, which, in turn, saves the viewer from having to activate subjective interpretative processes in order to construct the possible significance of the embedding technique.
36 consider when we discuss the complexity of films and may well be one of the characteristics by which we can distinguish contemporary complex cinema from its historical predecessors.
3. Complicating Narrative Techniques in Cinema
Now that we have clarified the distinction between complications at the level of diegesis and complications at the level of presentation and discussed the possible implications of the overt marking of complications, we can finally move on to our discussion of the narrative
techniques that may be used to achieve these complications. Each of the techniques that I present here can be shown to generate ambiguities in the significance of characters’ actions, the relative position of narrative layers or the order of story events. These ambiguities frustrate the viewer's PECMA flow, prevent immediate comprehension and activate the flow of subjective abstract associations. The viewer is thus engaged in a process of subjective hypothesis-formation and testing that may take the form of a playful “mind-game” (Elsaesser 2009) or may serve to communicate a more serious, hidden message about ethics, human relations, the role and function of art or the problems of representation.
It is at this point that we depart from Torben Grodal's strict distinction between mainstream cinema and art cinema based on complications on the level of presentation. Although the messages communicated by complications in art cinema narratives may differ greatly from the messages that are communicated by the complications in mainstream cinema, the strategies and techniques used to achieve complications may still be the same. As our previous example of Inception (2010) showed, mainstream films may adopt complicating narrative techniques of presentation traditionally associated with art cinema and still be highly successful, as long as these complications are balanced by a sufficient degree of overt
37 diegesis. At the same time, our example showed that the use of balancing mechanisms may lessen the impact of such complications and prevent a complete blocking of the PECMA flow. I suggest, therefore, that we must accept the possibility that the same complicating technique can have various degrees of impact on the viewer's PECMA flow cycle depending to the extent to which balancing mechanisms are employed and that rather than making a sharp, black-and-white distinction between films that employ such techniques and films that do not, we must keep an open mind towards the possibility that mainstream film narratives have recently begun to adopt complex strategies of presentation previously thought to be exclusive to art cinema narration.
With our minds open to this possibility, let us now embark on an in-depth discussion of each of the techniques that are most frequently used to complicate film narratives. With each complicating technique we discuss, we will pay attention to three aspects: 1) How can this complicating technique cause a blocking in the viewer's PECMA flow? 2) Does this complicating technique function on the level of the diegesis, on the level of presentation, or may it function on both levels? 3) How can the technique be marked and to what extent can marking be argued to possibly lessen its cognitive impact? The complicating techniques we will discuss are temporal fragmentation, embedding, alternate timelines, the withholding of plot information, character unreliability and narrative unreliability. Complicating narrative techniques that are practically never used in even the most complex of contemporary film narratives, such as denarration7, are left undiscussed for the purpose of brevity.
38 3.1 Temporal Fragmentation
One of the most frequently employed complicating narrative techniques in cinema is the fragmented presentation of time. This may either take place entirely on the level of
presentation, or may happen within the diegetic world itself in films about time travel, such as Terry Gilliam's Twelve Monkeys (1995). The flashback technique is almost as old as narrative cinema itself, while the technique of flashing forward to possible future events is far less common. As Bordwell suggests, events that are presented as having taken place in the past have a far greater credibility and certainty to them than events that are presented as taking place in the future, for the simple reason that it is hard for us to imagine the future as
39
Memento (2000). Here, scenes that follow each other seem to belong to the same time line, for
they deal with the same story of a man suffering from short-term memory loss who is set on avenging the violent death of his wife, but a temporal leap occurs in-between each subsequent scene. Nor are these temporal leaps flashbacks of the conventional kind; we do not jump back and forth between current time and the memory of past events, but rather, two sets of scenes are intermixed. In the color scenes, the story is told in reverse chronology, from its end to its beginning. In the black-and-white scenes, the story is told chronologically, but the viewer requires information from both the color and the black-and-white scenes to puzzle the story together. Only at the end of the plot do the two currents meet: here, the protagonist finally confronts the violent truth of his wife's death and the ultimate consequence of his endless quest for revenge. This film takes the complicating technique of temporal fragmentation to the extreme, yet again, it balances the potentially confusing effects of this technique by marking it. Thanks to the fact that the scenes are color coded, viewers can make sense of the confusing narrative much easier, although it is still a puzzle film that appears to be intended for multiple viewings.
We can explain the confusing effect of temporal leaps with the help of Torben
40 they have both proven essential to our survival as a species. As a result, cause-and-effect chains have become the most habitual way in which we make sense of the world. The fragmented presentation of time frustrates our embodiment in a diegetic world because it shows events outside of the chronological sequence that we are used to. Moreover, a temporal leap creates ambiguity: we cannot be certain where to place the events of the following scene in chronological relation to the events of earlier scenes. This ambiguity makes it more
difficult to decide on the significance of the events we are now shown, and we are
consequently forced to activate subjective, abstract associations in order to make sense of them and hypothesize how they fit in the chronological order of the narrative. Our drive to construct chronological consistency is so great, that even when faced with inconsistencies we try to force events into an order that makes causal sense. For example, viewers of films with purposefully inconsistent chronologies, such as Last Year at Marienbad (1961) or Triangle (2009) frequently attest to trying to puzzle together a consistent timeline from the
contradictory information that the film gives them.
Temporal fragmentation is a technique that operates primarily on the level of presentation, although it is frequently diegetically motivated, and almost without exception rhetorically motivated. The temporal fragmentation of Memento is diegetically motivated, because it allows the viewer to experience the diegetic world from the subjective perspective of a man suffering from short-term memory loss. It is also rhetorically motivated, for two reasons. First of all, knowing very well that is impossible to have his viewers experience literally what it feels like to constantly forget everything that happened more than a few minutes ago, Nolan has chosen to recreate the effect as closely as possible by cutting up the narrative in so many little pieces that viewers will have almost the same difficulty
41 allows the film to delay its tragic revelation to the very end – a powerful rhetorical effect that is designed to leave the viewer deeply impressed with the filmmaker's ability as a storyteller. Temporal fragmentation can be marked in several ways. The most common way is to have a voice, emanating from either an off-screen narrator or a character in the film suggest that he or she is about to narrate events that took place in the past. The screen then blurs, or another clear transition effect is used, and the flashback is shown. At the end of the flashback, the screen goes through the same transition and we find the characters where they were, as if they've just told each other this story. It is not surprising that this is the most common form of temporal fragmentation seen in mainstream cinema, for it is the one that most closely
resembles our daily experiences of conversational storytelling. Someone may tell a story of events that took place at an earlier time, but the events are almost always narrated in
chronological order and clearly indicated as having taken place in a previous time. Color coding, such as is used by Memento, is another frequently used method. A common practice is to use the association with past times that black-and-white film stock calls up. Dead Again (1991) shows us the world of Los Angeles in the 1990s in color, but all of the film's flashbacks to the world of 1949 are shot in black-and-white. Even more common is the use of natural seasons or widely known holidays to indicate different times. When a woman waits for her lover to return, a montage sequence indicating the passage of seasons is a universal way of saying she is waiting for a very long time. The passage of seasons is something viewers understand on a very basic, evolutionary/experiential level. Also
frequently used are more technological markers of time, such as clocks and radios (think of
Groundhog Day). In The Machinist (2004), clocks are used subversively to disrupt the sense
42 The marking of temporal fragmentation greatly lessens its potentially confusing
impact by removing some or all of the ambiguity about where the events we witness fit in the chronology of the narrative. This allows us to remain immersed in the narrative without having to activate higher cognitive processes such as association, abstraction and hypothesis formation. Furthermore, if the way new information is fit into the larger chronology of a narrative does not contradict information we have been given earlier, no ambiguities arise from the temporal leap. When, on the other hand, an entirely different set of events is clearly suggested to have taken place in the same place at the same time, the effect is the opposite and an ambiguity arises which necessitates subjective judgment.
3.2 Embedding and Metalepsis
Embedding is a narrative technique whereby one narrative layer is nested in, or framed by, another narrative layer. In its least radical incarnation, this technique consists of a frame narrative, in which a character introduces a story, followed by a leap into the embedded narrative which consists of the story told by that character. The most well-known example from literature are the stories of the One Thousand and One Nights, which are framed by a narrative of Scheherazade, a slave girl who narrates fantastic tales to her king in order to delay her execution. Some of the tales have additional embedded layers, as narrators within a tale tell another tale. This may be part of Scheherazade's scheme, for by embedding narratives within narratives a tale can potentially grow infinitely long.
43 characters each tell their own, subjective versions of the events leading up to a murder, but this scene at the in the court room is itself embedded in a frame narrative of two men talking about the murder case while taking shelter from the rain under the Rashomon gate. Again, embedding is combined with temporal fragmentation; the embedded sequences are
flashbacks. Equally common is the embedding of alternative or virtual realities, as is done in films like Tron (1982), Labyrinth (1986), Lawnmower Man (1992), eXistenZ (1999), The
Matrix (1999), The Cell (2000), Pan's Labyrinth (2006) and Inception (2010) – although the
fact that all of these films appeared from the 80's onwards may suggest that this type of embedding is a relatively novel trend in mainstream cinema that could very well be related to the emergence of videogames and virtual realities.
A jump from one diegetic layer to another can have an immensely disorienting effect and may create a large number of metaphysical ambiguities. When we are given insufficient information to determine the relative place of a sequence in relation to what we experience as the top diegetic layer or the film, or what we experience as the 'reality' of the film world, we cannot be sure whether the sequence takes place in a different time, a different place, or in an alternative, virtual or dream reality. As a result, we cannot determine the significance of the information in the sequence for the main plot, and cannot be certain whether the new
44 entire experience of the narrative between brackets – precisely as happens at the end of both
eXistenZ and Inception.
Because jumps between diegetic layers can be so unsettling, viewers are dependent on the clear marking of diegetic layers in order to understand embedding. A film like Inception achieves this by situating each of its layers in a very different setting and by marking the metaleptic jumps that the characters make with the use of a dream machine, which
simultaneously has the effect of naturalizing the complex plot structure using the internal laws of the diegetic universe. ExistenZ (1999) does the same by introducing organic “game pods” that the characters can plug into their spines in order to access alternative realities, but this film fails to clearly mark the differences between diegetic layers in any other way, which quickly leads to extensive confusion for both the characters and the viewer. At each layer, the characters meet more game-junkies like themselves who are eager to plug themselves in and experience another, deeper layer of virtual reality. As is common in many films of the cyberpunk genre of the 1990's, eXistenZ thus seems designed to communicate a moral
warning about the 'dangers of virtual reality'; go in too deep and you might end up losing sight of reality altogether. While the moral message of The Matrix (1999) may be very different (the film's story can be summarized as a cyberpunk version of Plato's allegory of the cave), its use of an embedded virtual reality is equally extensive. Here, however, the distinctions
45 Films like Inception, eXistenZ and The Matrix all supply explanations for their use of embedding from within the diegesis. This motivates the embedding mechanism by making it seem like a normal part of the fictional world, just like a series Star Trek presents space travel at light speed as a normal occurrence. This effect is usually achieved by presenting an
unnatural device or occurrence as the product of an aspect of the fictional world that escapes explanation because it is either magical or the product of scientific advancements that go far beyond our current ability to understand. While each narrative layer of Inception is
technically another layer of reality, the film reassures the viewer that each layer is still linked to the same, fully consistent, greater reality via the dream devices. Moreover, events at higher diegetic levels affect the lower levels, such as when a van drops off a bridge and the
protagonists that are dreaming in the back of the van notice a loss of gravity inside their dream. This suggests that, even though the protagonists have descended to a deeper layer of the dream world, this layer is still fully consistent with the higher layers of the dream world. As a result, the diegetic world manages to maintain a high degree of continuity of space and time, which, as Grodal suggests, is the most vital requirement for the viewer's unproblematic embodiment.
3.3 Alternate Timelines
This narrative technique is the primary subject of Bordwell's famous article Film Futures, and as such it makes sense to begin our discussion of it with a few of Bordwell's observations. He opens his article with a citation from Jorge Luis Borges' short story “The Garden of Forking Paths”, in which the protagonist meets a man who explains to him the structure of a
46 He believed in an infinite series of times, a growing, dizzying web of divergent,
convergent, and parallel times. That fabric of times that approach one another, fork, are snipped off, or are simply unknown for centuries, contains all possibilities. In most of those times, we do not exist; in some, you exist but I do not; in others, I do and you do not; in others still, we both do. (Borges 1941: 127)
This novel, of course, is a fantasy of Borges, for if it were ever to be written, it would be infinite in length. In actual fiction, Bordwell remarks, “alternative futures seem rather limited affairs” (89). Most film plots limit themselves to only two or three alternative timelines: “None of these films hints at the radical possibilities opened up by Borges or the physicists.
Blind Chance and Run Lola Run present only three alternative worlds, while Sliding Doors
and Too Many Ways to Be Number One offer the minimum of two” (89). He points out that most fictional stories make use of no more than three alternative timelines, while those that do, like Groundhog Day, adhere to very clearly marked and unchanging points of departure and present the events that take place within each alternate timeline in neatly chronological order, and suggests that such limitations may suggest an intentional choice on the part of the filmmakers to keep the narratives “cognitively manageable” (90). In the same article, he furthermore suggests that by clearly “signposting” (marking) the fork in time, these films significantly limit the disturbing impact of the technique.
Since the publication of Bordwell's article in 2002, however, several mainstream films have pushed the alternate timeline plot beyond its conventional boundaries. The Japanese animated feature Ghost in the Shell 2: Innocence (2004) features an extensive time looping sequence that presents three alternative versions of two detectives entering a hacker's
47 the number of alternate timelines is still limited, the moment at which time splits comes as surprise due a lack of clear marking. In the middle of a conversation with the hacker, one of the detectives looks at a diorama of the mansion. Zooming in on it, we notice the two detectives standing near the mansion's entrance and the scene repeats with only slight alterations. The transition even allows for the other detective to finish the last line he was speaking, except it now seems as if he was addressing his partner rather than the hacker. This sequence is a clear example of how the lack of overt marking greatly increases a complicating technique's disturbing impact.
Mr. Nobody (2009) features perhaps the most advanced use of alternate timelines seen
in mainstream cinema to date; this film attempts to narrate many different possible versions of a man's life from his birth to his death, and in doing so manages to stay as close to the
multiple-worlds concept of reality as any film ever has. Contrary to most forking path plots that seem to prefer either blind chance (Sliding Doors, Blind Chance) or conscious choice (Groundhog Day, Too Many Ways to be Number One) as the catalyst for temporal forks, forks in this film can be actually be caused by both. Furthermore, forks frequently split off into three or more alternate paths. In my analysis of the film I counted a total of 4 forks leading to an impressive number of 7 different endings.
Although the use of alternate timelines is a relatively novel trend in mainstream cinema, art cinema, in particular surrealist cinema, has pioneered the technique from as early as the 1920's. The Seashell and the Clergyman/La Coquille et le Clergyman (1928) presents multiple conflicting and seemingly simultaneous erotic fantasies of a priest lusting after a general's wife.
Meshes of the Afternoon (1943) shows a woman entering a house and falling asleep in
