Flashbacks in Film

A Cognitive and Multimodal Analysis

Adriana Gordejuela

• Publisher: Routledge
• Year: 2021
• Print ISBN: 9780367721312
• eBook ISBN: 9781000379419
• Edition: 1
• Topic: Media & Performing Arts
• Subtopic: Film & Video

Flashbacks in Film examines film flashback as a rich multimodal narrative device, analyzing the cognitive underpinnings of film flashbacks and the mechanisms that lead viewers to successfully comprehend them.

Combining a cognitive film theory approach with the theoretical framework proposed by blending theory, which claims that human beings' general ability for conceptual integration underlies most of our daily activities, this book argues that flashbacks make sense to the viewer, as they are specifically designed for the viewer's cognitive understanding. Through a mixture of analysis and dozens of case studies, this book demonstrates that successful film flashbacks appeal to the spectator's natural perceptual and cognitive abilities, which spectators exercise daily.

This book will serve as a valuable resource for scholars interested in film studies, media studies, and cognitive linguistics.

1 Introduction

A cognitive approach to film

1.1 Cinema and cognition

Throughout the history of cinema (a relatively young medium with little more than a century of life), the approaches adopted for its analysis have been diverse. Since the early years, the exploration of the possibilities of cinematic technique came hand in hand with a theoretical interest in the new medium. Intellectuals from various fields and filmmakers themselves began to theorize about different aspects of film: prominent examples of this are, for instance, psychologist Hugo Münsterberg’s book The Photoplay: A Psychological Study (1916) (one of the first cognitive considerations of cinema) and the Soviet montage theory propounded by authors such as Eisenstein and Pudovkin (cf. Corrigan & White, 2004, pp. 437, 441–443). Over time, the so-called classical film theory was developed by figures like Balázs, Arnheim, Bazin, or Kracauer, who among other things discussed cinema in terms of formalism and realism (cf. Corrigan & White, 2004, pp. 443–447). At the same time, up to the 1970s, “auteur theory” was the predominant conceptual framework, which gained ground, among other reasons, thanks to the Cahiers du Cinéma critics, European postwar “art cinema”, and the appearance of major directors in Hollywood in the 1950s (Bordwell, 1996, pp. 4–5).

It was not until the 1960s that film studies were born as an academic discipline. However, being a new field, it was in need of accreditation, and in order to gain academic status film studies resorted to trendy theories of the moment, some of them already well established: first to structuralism and later on to semiotics, Marxism, psychoanalysis, and others (cf. Bordwell, 1996). In the light of structuralism and semiotics, one of the ideas that rapidly gained popularity was that of approaching film as a kind of “language”: that is, as some sort of code composed of signifiers and their corresponding signifieds and governed by a set of rules that would work as “film grammar”. This approach sought to account for how film creates and conveys meaning and how, as a consequence, the viewer has access to it (cf., for instance, Carroll, 2003, pp. 14–25; Bordwell, 2010, pp. 3–5, 2011b).

Broadly speaking, what has come to be called (by its proponents) “contemporary film theory” took shape from the 1970s on, emerging from a combination of the aforementioned theories. But, from a critical position, it has also been named SLAB theory, since it is mostly based on Saussurian semiotics, Lacanian psychoanalysis, Althusserian Marxism, and Barthesian textual theory (Bordwell, 1989b, p. 385), and it has also been called grand theory due to its all-encompassing, totalizing aim and doctrine-driven functioning (cf. Bordwell, 1996; Carroll, 1996a). In this tradition, “[t]heorizing becomes the routine application of some larger, unified theory to questions of cinema, which procedure churns out roughly the same answers, or remarkably similar answers, in every case” (Carroll, 1996a, p. 41).

During the 1980s, a current critical with this grand theory emerged in the light of the “cognitive turn” that was taking place in different fields of knowledge. An alternative framework began to develop, that of cognitive film theory, which opposed the reigning paradigm in a number of ways:¹ first, it did (and still does) not claim to be the theory of film, but rather, a research program with a multidisciplinary spirit that draws from a variety of theories from different disciplines (e.g., anthropology, neuroscience, psychology, linguistics) in order to answer specific questions (Carroll, 1996a, pp. 38–41). In this sense, cognitive film theory also advocates bottom-up inquiry: that is, middle-level or piecemeal research that proceeds by asking particular questions about films (Bordwell, 1996, pp. 26–30) and seeks answers by being “committed to clarity of exposition and argument and to the relevance of empirical evidence and the standards of science (where appropriate)” (Plantinga, 2002, p. 20). Furthermore, the cognitive program focuses on explaining films (i.e., accounting for how they function, particularly in relation to the viewer) rather than on interpreting them (i.e., offering new “readings” of specific films), and in this way it also sets itself apart from grand theory (Bordwell, 1996, pp. 24–26; Carroll, 1996a, pp. 41–44) (cf. also Chapter 2, Section 2.2.2, and Chapter 3, Section 3.1.1).

1Summaries of the cognitive approach to film can be found, for instance, in Bordwell (1989a, 2009a) and Plantinga (2002).

This cognitive approach to cinema has also been named “naturalistic” because many of its advocates have underlined the centrality of viewers’ natural cognitive capacities in the activity of watching films (e.g., Anderson, 1996; Carroll, 2003, pp. 10–58; Bordwell, 2010; Carroll & Seeley, 2013). This view directly challenges that of cinema as language (i.e., a system of codes), since it argues that we do not make sense of films because we learn their particular codes or conventions (at least not like we learn a given natural language), but because movies are designed to appeal to the same natural abilities with which we perceive and understand the world around us.

This study is placed within this cognitive, naturalistic trend in film analysis. Specifically, it addresses the question of how film viewers successfully comprehend movie flashbacks: how is it that a device that is jarring in principle (it breaks a narrative’s temporal and spatial continuity) is usually understood without effort by spectators? What are the cognitive principles at work behind this device? Certainly, narrative retrospections are not exclusive to cinematic discourse: they are indeed present in all kinds of narratives, from the most elaborated ones to the spontaneous everyday stories that come up in conversations. Retrospections are even part of our own daily reasoning. Considering all this, a hypothesis could be raised at a broader level: it may be that human beings are naturally “narrative”, and thus concepts such as those of sequentiality or cause/effect are fundamental tools of our way of thinking (cf., for instance, Anderson, 1996, pp. 144–149, for some “ecological” reflections on narrative). One of the questions that ensue is whether this “narrative quality” of our minds predisposes us to naturally understand film narratives (and, therefore, movie flashbacks, these being a particular subversion of this narrative sequentiality), which would rely on basic daily narrative mechanisms and other natural cognitive processes.

In recent years, there has been a growing research interest in explaining how viewers comprehend film narratives. The paths taken in order to find answers have been many and varied: there is, for instance, the socio-semiotic model, which considers cinema as a particular type of multimodal discourse and analyzes it from a discursive and textual perspective (cf. Bateman & Schmidt, 2012). Within the cognitive field, conceptual metaphor theory (summarized in Kövecses, 2017), which considers metaphors as a tool of our minds for the creation and communication of meaning, has also been applied to cinematic analyses of different kinds (e.g., the metaphorical representation of character perception in film) (cf. Coëgnarts & Kravanja, 2015). The cognitive analysis of flashbacks proposed here relies primarily on a particular theory of cognition, blending theory or conceptual integration theory (Fauconnier & Turner, 2002), which has already helped to shed light on the workings of multimodal artefacts (e.g., Vandelanotte & Dancygier, 2017), including film (e.g., Oakley & Tobin, 2012), and on the ways readers understand literary narratives (e.g., Dancygier, 2012b). This interdisciplinary spirit also inspires this study of film flashbacks, which draws from concepts and theories set forth by different disciplines, mostly by cognitive linguistics, narratology, and film studies. It is from the points of convergence between these disciplines that answers to the questions raised will be put forward.

1.2 Conceptual integration theory

1.2.1 Blending theory: an overview

Developed by Gilles Fauconnier and Mark Turner (2002), blending or conceptual integration theory is a cognitivist framework that accounts for how human beings construct and comprehend meanings in a broad sense. Essentially, Fauconnier and Turner argue that conceptual blending is a basic (but complex) mental operation characteristic of human beings’ way of thinking and that “is as indispensable for basic everyday thought as it is for artistic and scientific abilities” (Fauconnier & Turner, 2002, p. vi). This cognitive operation is composed of nonconscious processes that are prompted by “language expressions (but also visual images, sounds, gestures, and all other meaningful forms of human expression) … [that] the human mind uses in an act of meaning construction and comprehension” (Dancygier, 2006, p. 6). Thus, it is argued that we perceive manifestations of conceptual integration, but those are only the tip of the iceberg: the complex mental processes motivated by those forms take place beneath the surface, and they are so natural to us that they go unnoticed.

This theory, largely based on Fauconnier’s (1994 [1985], 1997) mental spaces theory, proposes a model of dynamic construction of meaning that involves the integration of information from different input mental spaces (defined as temporary conceptual constructs set up as we think and talk), which are prompted by linguistic and nonlinguistic (images, gestures, etc.) units. Counterpart elements from the different input spaces are connected via cross-space mappings, and those spaces and their connections shape a conceptual integration network. In the network, different elements from the input spaces are selectively projected into a blended space, where new meanings that were not previously available in any of the inputs appear (i.e., emergent structure). Furthermore, those features shared by all the input spaces in the network make up a generic space that also takes part in the blend. Also, structured, long-term knowledge may be projected into the blend in the shape of cognitive frames. A further point is that processes of compression and decompression, which are central to conceptual blending, operate at different levels in a given integration network. Basically, compression involves tightening up the vital relations (e.g., time, space, identity, change) that exist between elements in the input spaces, and decompression is the opposite process. As a result, the blend can be manipulated as a unit, and it gives access to an otherwise complex set of conceptual structures in a clear and simple manner.

These and other concepts related to blending theory will be explained in depth as they appear in the flashback analyses in Chapters 3, 4, and 5. However, a simple case will serve now as a preliminary illustration of these concepts and operations. Fauconnier and Turner (2002, pp. 59–62) propose an example of a conceptual integration network that has now become a classic one in blending literature: “the debate with Kant”. Imagine that a contemporary philosopher who studies Kant’s writings gives a lecture, and in the course of the session, he says things like “Kant disagrees with me on this point”. Or, after finding no answers to his own questions in Kant’s texts, he states: “And he [Kant] gives no answer”. This scenario would be activating at least two main input spaces: one for the contemporary philosopher (who studies Kant, but has his own ideas) and another one for Kant himself (with his own thinking and his writings). There would be cross-space mappings between both inputs, and the counterpart elements include Kant and the professor, their respective languages (German and English), their times and places of activity, their respective claims, and so on. Elements from both inputs would be projected into a blended space, since only in such a space can a philosophical discussion between the men in both input spaces take place. Furthermore, “the frame of ‘debate’ has been recruited to frame Kant and the modern philosopher as engaged in simultaneous debate, mutually aware, using a single language to treat a recognized topic” (Fauconnier & Turner, 2002, p. 60). But projection of elements from the inputs to the blend is selective, since, for instance, only one of the two languages of the philosophers would be projected (English in this case, and not Kant’s German). Also, in the blend, the temporal and spatial distance between the two philosophers would be compressed, so that the two of them are face to face having a debate. Finally, there would be also emergent structure in “running the blend”: the debate could go on, new questions and answers could be elaborated, attitudes like defensiveness or overconfidence could appear, and so on. In short, the discussion between the two philosophers could only be conceived through the activation of a conceptual integration network. In this case (as in many others), “we do not even notice the blending … since the general blending template it deploys is conventional for engaging the ideas of a previous thinker” (Fauconnier & Turner, 2002, p. 61). But this only speaks in favor of the aforementioned nonconscious, natural quality of conceptual integration processes, whose existence has been supported so far by several kinds of experimental research (cf. among many others Coulson, 2001; Eppe et al., 2018).

1.2.2 Film flashbacks and blending theory: a cognitive analysis

Conceptual integration theory seems like a suitable framework for the analysis of film flashbacks for a number of reasons. First, blending theory seeks to account for our basic general capacity for conceptual integration, a kind of mental operation that would underlie our simplest everyday thoughts and activities but also our understanding of sophisticated products like literature and art. Thus, conceptual integration could also play an essential part in our comprehension of films. Furthermore, blending has already been successfully applied to the study of different fields such as literature (e.g., Dancygier, 2012b) and multimodal communication (e.g., Dancygier & Vandelanotte, 2017; Vandelanotte & Dancygier, 2017). Cinema, being a particular kind of multimodal narrative discourse, could also be approached from the theoretical framework that blending proposes. In fact, there is already some research on conceptual integration and film (e.g., Oakley & Tobin, 2012; Tobin, 2017), and the prospects look promising. Finally, the cognitive, naturalistic approach to film explained earlier, in its multidisciplinary spirit, could profit from conceptual blending and find in this framework a fruitful theoretical ground to describe how viewers make sense of film flashbacks. Thus, another important objective of this book is to test the validity of blending theory for the analysis of cinema and, more specifically, of movie flashbacks.

At this point, it should be emphasized that this study does not intend to offer new “readings” of a series of flashbacks. Finding new interpretations is not an aim of blending theory. Rather, it seeks to offer an explanation of some of the backstage cognitive processes at work when we make sense of products like a novel, a painting, or a film. Thus, the main goal of this study is to cast light on how viewers successfully and effortlessly understand movie flashbacks as such (i.e., as narrative retrospections) and not as something else. Certainly, the line between comprehension and interpretation is very thin: Persson (2003), for instance, distinguishes six levels (from zero to five) in the process of understanding cinematic meaning. Levels zero to three involve basic perception, object and character recognition and categorization, and identifying abstract situations (which entails relations between events, character psychology, etc.) (Persson, 2003, pp. 27–32). At level four, however, more abstract meanings emerge (e.g., thematic inferences, symbolical understandings), and “we enter the twilight zone between comprehension and interpretation” (Persson, 2003, p. 32). Finally, Persson describes level five as that of interpretation in a broad sense, which involves, among other things, aesthetic judgments and critical evaluations of a film’s message. Nevertheless, Persson (2003, pp. 36–37) also points out that the viewer does not simply move up along this continuum from the lower to upper...

Table of contents

• Cover
• Half Title
• Series
• Title
• Copyright
• Contents
• Acknowledgments
• 1 Introduction: a cognitive approach to film
• 2 Flashbacks in film
• 3 Blended joint attention
• 4 Viewpoint compression
• 5 Time compression
• 6 The whole picture
• 7 Conclusions
• References
• List of films
• List of flashbacks
• List of figures
• Index