Cohesion in Film

Tracking Film Elements

C. Tseng

• Publisher: Palgrave Macmillan
• Year: 2013
• Print ISBN: 9781137290335
• eBook ISBN: 9781137290342
• Topic: Media & Performing Arts
• Subtopic: Film & Video

This book presents a new approach to film analysis. It provides methods for analysing meaning making in film through tracking concrete details of film images such as characters, objects, settings and character action. It also represents new ground for investigating empirical issues in film.

1

Introduction

This book proposes an analytical framework for systematically exploring how film elements interact and are signaled coherently to viewers. Meanings in film are created by the complex and coherent combination of modalities such as words, images, sounds, actions, technical features, and so on, and in this book I investigate just how this creation of meaning takes place. Particularly, this study focuses on the four types of film elements: characters, objects, settings and characters’ actions, and examines how these elements are cohesively tied together as films unfold.

Recent cognitive experiments such as eye tracking have told us that these four elements in film are among the most salient elements viewers attend to in their viewing processes (cf. Smith, 2012). Moreover, these four types of elements are also the main perceptual leads for viewers to construct narratives as films unfold. For instance, according to Smith (2012), the process in which viewers attend to these film elements on the screen also triggers their perceptual inquiry (Hochberg and Brooks, 1978). Namely, they ask questions and construct expectations about ‘Who is speaking?’, ‘What are they looking at?’, ‘What was that movement?’, and so on. Other research disciplines associated with film analysis also imply that the elements of characters, their actions and objects associated with them are the main resources for viewers to make predictions and inference about higher-level understanding of characters’ traits, emotion, entire narrative, genre and so on (cf. Smith, 1995; Anderson, 1996; Eder, 2010; Visch, 2007). Drawing on the significant role these elements play in viewers’ meaning making processes, this book will propose analytical methods to systematically unravel how these elements within and across shots and scenes are coherently signaled to viewers and will elucidate how the patterns constructed through tracking these four elements further shed light on issues in broader social contexts such as genre comparison.

Methodologically, this study extends the linguistic notion of cohesion in recent developments in functional linguistics (Martin, 1992; Halliday and Matthiessen, 2004) because its analytical categories can be used to unravel effectively how cohesive ties are established between characters, objects, settings and characters’ actions in film.

The research emphasis on viewers’ coherent construction of meaning in film has been discussed substantially in different disciplines such as media studies, semiotics, cognitive theory, and so on. For instance, in recent decades scientists associated with cognitive film theory have interconnected the understanding of narratives and investigations at lower levels to explain how audio-visual properties trigger processes of cognition (cf. Zacks and Magliano, 2011; Smith, 2012). Following these findings, the concomitant issue facing empirical research is how to further link them to the higher-level social context (cf. Coco and Keller, 2009). This book will endeavor to take precisely this necessary step. It will propose a framework capable of exploring how empirical findings can be scientifically explained as reflecting the more abstract level of broader social meanings.

Moreover, many studies conducted in social and cultural disciplines tend to relegate to the background the workings of the forms and structures of the artifacts being analysed. That is, the correlation between the presence of broad content elements, emotional impact on spectators and genre issues tends to be investigated without consideration of the detailed internal forms and structures. However, recent research on media communication indicates that the ‘context’ constructed by the textual unfolding of film narrative needs to be considered in order to unravel filmic meaning (Hartmann and Vorderer, 2010). From this perspective we consider it necessary to build a bridge across the interpretation of social meaning and systematic, empirical support from textual and formal analysis.

To frame our investigation, this chapter will first address three questions which bring the objects of the study to the fore more concretely: 1. What is the analysis of cohesion? 2. Why use a linguistic approach (again)? 3. What does cohesion in film have to do with film genre?

1.1 Tracking elements in language and in film

Cohesion analysis in language displays significant aspects of how a text unfolds. It is concerned with the linguistic devices which are mobilised to meaningfully tie together bits and pieces and ‘refers to the range of possibility that exists for linking something with what has gone before’ (Halliday and Hasan, 1976, p10). Cohesive devices such as cohesive reference tracks people, places and things throughout a text; and devices of lexical cohesion can be used to track similar words describing types of actions in a text. In linguistic analysis, the tracking of these elements throughout a text constructs patterns which represent the mobilisation of these cohesive devices as a text unfolds. The theory and practice of the analytical framework will be elucidated in detail in the following chapters; nevertheless, to give a general picture of how the application of linguistic cohesion to film is possible and how the analysis in film is conducted, I will give some brief examples of tracking elements in language and in moving images.

First, let us consider the following short text:

(1) The opposition has called for the president to step down, (2) and he needs his party’s backing to survive a move to oust him.

Linguistic analysis shows how devices of cohesive reference are mobilised here to tie together the identity of ‘the president’. Following the presentation of ‘the president’ in clause (1), the president is tracked by the re-presentations ‘he’, ‘him’ and ‘his’ in clause (2). These cohesive links allow the construction of a cohesive pattern, or cohesive chain, as displayed in Figure 1.1. This kind of pattern shows how cohesive ties are signaled to the reader throughout the short text.

Cohesive devices and cohesive patterns can also be applied to moving images. This, and the empirical issues filmic cohesion can approach, will be the main tasks taken up in this book. In the remainder of this section, therefore, I briefly exemplify the application of cohesion analysis to moving images to set the scene for what will follow. Consider the filmic extract portrayed in Figure 1.2. Here we see the first six shots of a TV commercial that is attempting to communicate to the viewer how trustworthy the service of the company is. Upon seeing the image and the dialogue, the viewer readily grasps the story event as ‘a man talks to a car about some thing/symptom it has’.

The process for constructing cohesive patterns begins with the tracking of how the same identities of salient characters, objects and settings are presented and re-presented throughout a film sequence. We also examine whether these salient characters, objects and settings are presented and reappear mono-modally as visual elements or in the spoken/written language, or cross-modally in visual, verbal and audio modes.

Figure 1.1 Example of cohesive pattern constructed by tracking a character. An arrow shows how elements are tracked in order to signal this dependence

Figure 1.2 Extract from a Carfax TV commercial (Full transcription of this TV commercial is included in Chapter 3)

Figure 1.3 Cohesive patterns of the extract in Figure 1.2 constructed through cohesive linking of the identities of salient characters, objects and settings (numbers = the images in which the identities are present, ‘’ = elements in spoken text)

For instance, the content of the above extract concerns the man, the car, the things the car has (odometer, accident) and the overall setting. Like the tracking of ‘the president’ in Figure 1.1, how the four elements are tied together throughout the extract can be set out usually in cohesive chains as is displayed in Figure 1.3. The cluster of ‘man’ in the figure indicates that he is represented cross-modally in most of the images. Apart from being represented in the visual track, after his voice is associated with his identity the audio device of voice also tracks his identity over the extract. In image 2 his identity is realised in three modes (with visual, audio and the verbal text ‘us’, when he says ‘Let’s see . . . ’). So is the car, which is presented cross-modally both in the visual mode and in the verbal text ‘us’ (also in the man’s spoken text ‘Let’s see . . . ’) in image 2, as ‘you’ in image 3 and mono-modally referred to as ‘you, your’ in the man’s spoken text in image 5.

This book will investigate how analyses of this kind can help to reveal how films construct their meanings. Certain filmmaking strategies function precisely to signal filmic coherence by establishing cohesive patterns. For instance, the major devices used for signaling identities of filmic elements in the present example belong to strategies of continuity editing, which by the 1920s had already become a standardised style that filmmakers used to create coherent spatial and temporal relations within narratives (Bordwell and Thompson, 1993, p405). In this extract, image 2 establishes the spatial relation between the man and the car, and thus, in image 4, the identity of the man is signaled by an over-the-shoulder-shot despite the fact that his face is not seen. Many books on film theory and practice have devoted large sections to overviews and discussions of continuity editing (cf. Reisz and Millar, 1953; Salt, 1983; Dmytryk, 1986; Katz, 1991) and these established, standardised filmmaking strategies will be systematised in this book as one main feature of cohesive devices in film.

Figure 1.4 Cohesive patterns of the extract in Figure 1.2 constructed through cohesive links of the semantic relations between the salient characters, objects and setting (numbers = the images in which the identities are present, ‘’= relations spoken in verbal text)

Returning to the TV commercial, apart from character (man), object (car and things the car has) and setting (room), it is also useful to track salient actions or movements associated with these. In this example, the most salient action interlinking the man and the car is ‘talking’ (by the man to the car), while the relation between the car and ‘the things the car has’ is the semantic tie of ‘having’, spoken by the man in his verbal text in image 3 and image 5. These two salient semantic relations can also be cohesively tracked and patterned as displayed in Figure 1.4.

Mapping out the interaction of these cohesive patterns highlights the further significant dimension of story events in film. This dimension is constructed as another kind of pattern displayed in the upper part of Figure 1.5. The figure shows precisely how this pattern interlinks ‘man’ and ‘car’ through the action of ‘talking’ (by the man to the car), connects ‘car’ to ‘the things the car has’ through the semantic ties of ‘having’, and links the location ‘room’ to the overall event. This book will use the term action patterns to refer to such bottom-up constructions of content-based narrative patterns. In the subsequent images of this TV commercial, the man actually continues to inquire about the condition of other ‘things’ the car has. That is to say, the action patterns of the whole TV commercial would indicate dominantly that the man talks to the car, which has several explicit attributes that are made salient and talked about.

Figure 1.5 Event patterns of the extract in Figure 1.2 constructed by interlinking the cohesive identity patterns (Figure 1.3) and the cohesive action patterns (Figure 1.4) of the salient characters, objects and setting

This level of patterning is precisely the starting point where aspects of social meaning can be highlighted. If we apply the method for forming action patterns to a larger corpus, we can examine broadly whether analogous structures can be seen as typical in certain contexts. For example, in the present case the action pattern is clearly typical in a very different scenario – that of doctor-patient interaction. This allows us to explore how the viewer’s prediction of this story event is triggered according to the conventional action patterns existing in her culture.

This example analysis reveals at once that our account is dominantly bottom-up and stratified.¹ The basic assumption here is that filmic meaning can be seen as constructed through concrete filmic devices at the bottom level which are deployed coherently to realise meaning patterns at the higher level. The particular aspect of this higher level focused on in this book is the action patterns of film story events. These more complex meaning patterns can then be compar...

Table of contents

• Cover
• Title
• 1 Introduction
• 2 The Application of Functional Linguistics to Film
• 3 Cohesion in Film
• 4 Analysing Action Patterns in Film
• 5 Conclusion
• Appendix A: Filmography
• Notes
• Bibliography
• Author index
• Index