Open Your Eyes, and Watch Closely

(1)

Open Your Eyes, and Watch Closely

The role identification, empathy and sympathy with fictional characters play in the

viewer’s experience of films that use unreliable narration and narrative twists.

Master thesis

Arts, Culture and Media

Film Studies

University of Groningen

Rianne Pras

1880489

Supervisors

Dr. M. Kiss

Groningen

(2)

2

3 Introduction

Since the beginning of narrative film people have been experimenting with the possibilities of

narrative form, and the effects these might have on their audience. Over the past twenty years a new trend in cinema can be found, which is the use of complex narrational strategies. Complex narratives have been steadily making their way into mainstream cinema since the beginning of the 1990s, and nowadays it is not uncommon to encounter complex narratives in Hollywood blockbusters. There are several forms of complex narration but this thesis focuses on one in particular: complex narratives that make use of unreliable narration. This phenomenon is not new to film and can be found throughout film history, with examples dating as far back as Robert Wiene’s Das Cabinett des Dr. Caligari (1920).

Unreliable narration can come in different forms, and can be summarized as the mis- or under evaluating, interpreting or reporting of events by characters or the narrative itself.1_{It is often}

(but not always) revealed through a narrative twist, which most of the time is placed near the end of the narrative, overthrowing all the assumptions the audience has built up over the course of the film. This thesis focuses on the role that the relation between the viewer and the fictional characters plays in the viewer’s overall experience of the unreliable film. The main question that this thesis will try to answer is:

How do the emotions that are evoked by identification, empathy and sympathy with fictional characters relate to the experience of narrative unreliability?

It is my hypothesis that this relation between the viewer and the characters of an unreliable film is of key importance for the viewer’s overall experience of unreliability. I will aim to show that this relation influences the experience of the unreliable filmic narrative in several ways, before, during, and after the occurrence of the narrative twist. Since unreliability is narrative tool that has been used in cinema for a long time (and its use is increasing during the contemporary trend of complex cinema), it is important that we understand how these narratives work and how viewers experience them. I have chosen to look at this narrative form from the perspective of the viewer’s relation with the fictional characters because this relation is often described by both viewers and scholars as an important part of the film viewing experience in general.2

1_{Shen, 2013.}

(4)

4

(5)

5

Chapter 1 Unreliable narration and narrative twists in film

Unreliable narration has been a much debated topic in the field of narratology. It has been primarily studied in the context of literature, where the starting point has been Wayne Booth’s seminal definition that a narrator is “reliable when he speaks for or acts in accordance with the norms of the work (which is to say the implied author’s norms), unreliable when he does not.”3_{Further research on}

the subject that has been done since the 1960’s can be divided into two different but related approaches, the text based rhetorical approach (represented by, amongst others, Wayne Booth, James Phelan and Seymour Chatman) and the reader oriented cognitivist/constructivist approach (amongst others, Monika Fludernik, Ansgar Nünning and Tamar Yacobi).

As mentioned in the introduction, unreliable narration can be found in the cinema since the beginning of narrative fiction film. However, since the rise of complex narration in the 1990’s it has become more and more frequent, and can now be found in many films, ranging from mainstream Hollywood productions to world cinema, and different genres, from thrillers to comedies. It is within the context of this current trend of complex narrative films that we will be looking at unreliable narration, although a great number of findings in this thesis will apply just as well to films that use unreliable narration that do not fall within this context. When examining unreliable narration I will be looking at it within the context of film studies, leaving aside the discussions that take place in the field of literature because, despite the fact that there will be many overlaps, a lot of the theory and concepts from literature do not translate well directly to film. I will look at different perspectives, which in the end will hopefully provide a taxonomy of sorts, which lists the different forms of unreliable narration that can be found in film. Before I do this however, I will put it all into context. Therefore I will first take a brief look at the relation between film viewers and film narrative in general. After that I look at the specific form of narration this thesis focuses on, namely complex narration. At the end of this chapter I will also take a look at the emotional impact films that use unreliable

narration can have on their viewers, before delving deeper into the specific emotional experiences of identification, empathy and sympathy in relation to fictional characters in the second chapter.

(6)

6 David Bordwell’s constructivist theory of psychological activity

What follows here is a brief summary of the activity of the film viewer described by David Bordwell within his theory of narration, from his seminal work Narration in the Fiction Film. It is important to have a basic understanding of the way narrative comprehension works to have a clearer view on theories that treat ‘deviant’ forms of narration such as complex narration, unreliable narration and narrative twists. This is needed because films that use unconventional narrative forms, such as the unreliable narrative, often confront the viewer with the way they comprehend filmic narratives. While Bordwell’s theory is, of course, extensive, I will be focusing on the specific aspects of Bordwell’s theory that are the most important when it comes to studying unreliable narratives, and how the viewer responds to these narratives. These notions, such as schemata-driven thinking, hypothesis forming, the workings of assumptions and inferences and the notion of the primacy effect will provide relevant background information for the theories I will address and the analyses I will make in this thesis, since they form the basis of many of the tools that are exploited by film makers to mislead their viewers.

The viewer uses the concepts explained below to make sense of the narrative as a whole. This mostly happens unconsciously. Yet, when the unreliability comes to light, the viewer is directly confronted with how these processes work, and the fact that they have led him/her to draw the wrong conclusions, based on what turn out to be false hypotheses. This forces to viewer to actively consider how these hypotheses were formed in the first place. Thus, to understand complex narrative structures, such as the unreliable narrative, it is of great importance to understand the basic processes that the viewer needs to be able to watch films that not only require the viewer uses these processes, but who actually actively plays with them to achieve certain effects. Therefore I will briefly explain the concepts within Bordwell’s theory that are crucial for understanding the processes the viewer uses to comprehend film narratives and that the filmmaker can exploit to mislead the viewer in the case of unreliable narratives and narrative twists.

Bordwell’s theory is based on the cognitivist and perceptual aspects of film viewing. He builds his theory on an active viewer whose “central cognitive goal [is] the construction of a more or less intelligible story.”4_{Thus film is seen as a “dynamic psychological process” which combines a number of}

factors, namely perceptual capacities, prior knowledge and experience and the material and structure of the film itself.5_{What makes this process dynamic is the fact that the viewer is constantly revising} his/her hypotheses about the outcome of the story. This process is driven by the constant

confirmation or disconfirmation of these hypotheses. The viewer picks up information (perception)

(7)

7

and processes this information through the use of his/her available schemata, which Bordwell describes as organized clusters of knowledge, which can come in different forms, such as prototypes, templates or procedural patterns. These schemata, which provide knowledge about people, objects and situations are derived from the viewer’s everyday life experience and previous interactions with other films and art in general. When confronted with a a specific character, situation or object in a film, the viewer reaches back to the this stored knowledge and applies this to the film, ranging from recognizing specific objects or character types, to filling in certain actions where information is missing. The viewer then uses this information to either confirm or adjust their hypotheses. Thus schemata “coax us to anticipate and extrapolate.”6_{Schemata also guide the construction of memory}

of the viewer, which plays an important role in the construction of the overall (outcome of the) story. This can be of importance when looking at films that employ a narrative twist, because, as we will see, for a narrative twist remain hidden until the ideal moment of revelation the viewer must only

remember certain details of a scene. Throughout the viewing of a film the viewer is constantly testing his/her hypotheses about the story, based on the new information the story supplies. Nonetheless, as Bordwell states, “[n]arratives are composed in order to reward, modify, frustrate or defeat the perceiver’s search for coherence.”7

The use of schemata results in (nonconscious) inferences and assumptions from the viewer, concerning both the outcome of the events in the film and the fictional characters. The viewer assumes certain things to be true about both diegetic and nondiegetic elements. This can concern for example the ontological status of characters and objects in the diegesis, but also the fact that the same character will be played by the same actor in every scene. The viewer also makes inferences, for example that when a character is crying he/she must be sad. These conclusions can be adjusted, for example when it turns out the character is crying because she’s happy. The adjustments of these assumptions and inferences are made on the basis of schemata. The viewer assesses the overall situation that is presented, thereby taking into account all his/her previous knowledge of the film, his her hypotheses about how the character might be feeling in this specific situation, and knowledge about situations such as these in general. Based on this information the viewer can conclude what the reason behind the reaction of the character is. The viewer does this automatically and therefore these processes are usually only noticed when they are disturbed, for example in the case of films that use unreliable narration.8

An important factor in hypothesis making, both on a level of anticipation and the forming of assumptions and inferences is the primacy effect, a term that originally comes from cognitive

(8)

8

psychology and that Bordwell borrows from Meir Sternberg. The primacy effect describes how the order in which information is presented about a certain character or situation to the viewer, i.e. which information is given first, forms a frame of reference in which all consecutive information is evaluated. It can help create a positive or negative first impression of a character. The primacy effect influences the initial hypothesis the viewer has concerning the characters in the film, since it forms the basis for the schemata the viewer uses to assess the further actions of this character. Thus filmmakers can exploit this effect to mislead their viewers. Further on in this chapter I will show how the primacy effect plays an important role in films that employ narrative twists when I discuss Daniel Barratt’s theory on twist blindness. Bordwell and Sternberg already mention the potential and consequences of the primacy effect in relation to particular narrative forms: “[i]n a limit case, our primacy effect might be completely undermined by the suppression of key fabula information; we would then be forced to revise our assumptions and hypothesis when the data comes to light. Sternberg calls this the ‘rise and fall of impressions.’”9

As I already stated above, I focused this brief summary of Bordwell’s constructivist theory on the film viewer’s psychological activity on the specific elements that seem most relevant to the overall goal of this thesis, namely to see how identification, empathy and sympathy with fictional characters influence the viewer’s experience of films that use unreliable narration and a narrative twist. A I will show, unreliable narration and narrative twists derive their success from deliberately exploiting the way these processes work to mislead their viewer. While the basis of these processes lie in their use in everyday life, they are pivotal when it comes to both experiencing film and art in general. In these cases our schemata, and related to them the other processes described above, are used in a specific manner. As Bordwell concludes:

The artwork sets limits on what the spectator does. Salient perceptual features and the overall form of the artwork function as both triggers and constraints. The artwork is made so as to encourage the application of certain schemata, even if those must eventually be discarded in the course of the perceiver’s activity. […] In experiencing art, instead of focusing on the pragmatic results of perception, we turn our attention to the very process itself. What is

nonconscious in everyday life becomes consciously attended to. Our schemata get shaped, stretched and transgressed: a delay in hypothesis-conformation can be prolonged for its own sake. And like all psychological activities, artistic

(9)

9

activity has long ranging effects. Art may reinforce, or modify , or even assault

our normal perceptual-cognitive repertoire.10

Complex narration

Bordwell’s quote, which ended the previous section of this chapter, can most certainly be applied to the workings of complex narratives, which seem to have an inherent goal of making the viewer question the way they construct a narrative. Since the 1990s there has been a steady rise of films (and television series) that make use of complex narrative strategies to tell their story. Different terms have been given to this phenomenon such as puzzle films (Buckland) or mind-game films (Elsaesser).11

Whatever one chooses to call them, the idea is the same; namely that these films use a number of different techniques to create a complex plot structure, that presents a challenge to the viewer. In his edited volume Puzzle Films, Complex Storytelling in Contemporary Cinema, Warren Buckland, together with a number of scholars, focuses on the different aspects of complex narration in film, amongst them unreliable narration and the use of narrative twists. In his introduction Buckland sets off the puzzle plot against Aristotle’s concepts of the unified classical mimetic plot and the complex plot, arguing that the puzzle plot can be seen as a third type of plot. This plot is “intricate in the sense that the arrangement of events is not just complex, but complicated and perplexing; the events are not simply interwoven, but entangled.”12

According to Thomas Elsaesser, the mind game film, as he calls the phenomenon, overlaps both different types of cinema such as the Hollywood blockbuster and the European art-house film, and different genres, ranging from horror and science fiction to the film noir and even comedies. The overall common feature of these films is “a delight in disorienting or misleading spectators.”13_{They do}

this, according to Elsaesser, by playing games, which can happen at two different levels. First there are films in which a character is being played games with, without being aware of it, and/or who is doing this to him/her. Secondly, there are films in which the audience is being played games with, through the amount, quality, and way information is presented. Information can be withheld from just the audience, but it is also possible that it is withheld from the characters in the film as well. Sometimes, but not always the ‘master(s)’ of the game are revealed, and sometimes they get caught up in the

10_{Ibid., 32.}

11_{Buckland, Puzzle Films, 1 and Elsaesser, The Mind-Game Film, 13.} 12_{Ibid., 3.}

(10)

10

game as well. We will see later on that both of these levels correspond with different forms of unreliable narration.14

The viewers of these films, according to Elsaesser, do not mind the fact that they are being tricked. They accept the challenge the film sets them. What we see in these films is a change in the relation between the film and its audience. Instead of the passive, voyeuristic spectator position required from the viewer by the classical cinema, films that use complex narration demand an active, problem-solving attitude from their viewers. This allows the spectator to keep up with the frequent tricks, twists and other narratological devices used by these films to mislead them. This change in the relationship with the viewer has also been detected by Jason Mittell, who has seen similar tendencies in contemporary television series and serials. He notices that a number of television shows use complex narrative strategies and by doing so they “convert many viewers into amateur

narratologists.”15_{Mittell sees a “need for gaining competency in decoding and comprehending}

diegetic worlds [that] is particularly salient across a number of contemporary media.”16_{In all these}

media, including the complex narrative film, he observes a game aesthetic, which invites the audience to play along with the narrative. What is key however, for these films in relation to their audience, is that “the goal of these films is not to solve the mysteries ahead of time; rather we want to be competent enough to follow their narrative strategies but still relish in the pleasure of being manipulated successfully.”17

Unreliable narration in film

Now I will take a look at how unreliable narration can be defined when it comes to film. Dan Shen begins her article on narrational unreliability with a very clear basic definition of the concept which seems to encompass all the different definitions which have been given to unreliable narration within both the fields of literature and film studies. She states that “if a narrator misreports, -interprets or -evaluates, or if she/he underreports, -interprets or -evaluates, this narrator is unreliable or

untrustworthy.” This definition is based on the work of James Phelan, who considers reporting, interpreting and evaluating the main roles of a narrator. His work focuses mostly on literature. In film however, there is not always a clearly identifiable narrator. Most films do tend to focus on only one or a small number of characters which has as a result that the viewer is dependent on the information

14_{Ibid., 14.}

15_{Mittell, Complex TV, Complexity in Context.} 16_Ibid.

(11)

11

these characters receive and/or give to the viewer. Phelan further defines unreliability based on three axes, namely those of facts, values or ethics and knowledge and perception, and into two larger categories. The first contains misreporting, -interpreting or -evaluating, which are all based on being wrong. The second category consists of underreporting, -interpreting or evaluating, which are based on being insufficient. These forms can interact with each other. Nonetheless, within this very broad definition there is still plenty of room for more specific definitions of different kinds of unreliability that can be found in film.

In his article Fight Clubs, American Psychos and Mementos, The scope of unreliable narration in film Volker Ferenz starts by claiming that, unlike in the field of literary studies, in film studies there is still a lot of confusion about the notion of unreliable narration. Within literary studies most scholars agree that an unreliable narrator is “a character-narrator ‘whose rendering of the story and/or commentary on it the reader has reason to suspect”18_{which makes the reader sense a “discrepancy between the}

character-narrator’s version of the story and the events as they ‘truly’ are or have been” after which the reader judges this character-narrator as unreliable. 19_{Ferenz proposes a similar construct in the}

field of film studies, where he limits the notion of unreliable narration to the pseudo-diegetic

character-narrator. According to Ferenz “it does not make sense to attribute [unreliable narration] to a theoretical construct like the cinematic narrator or impersonal processes of narration.”20_{He claims}

that in the case of unreliable narration there always needs to be a clearly identifiable (fictional) character who has enough authority over the text for the viewer to blame for any inconsistencies. Ferenz refers to the perspectival principle proposed by Tamar Yacobi, which according to her, is one of the five recuperation strategies we use when we encounter inconsistencies in narrative texts.

According to Ferenz this is the only principle that allows for unreliable narration, since it is the only principle that attributes inconsistencies to a character through which the narrative is mediated.

Ferenz thus presents a very narrow definition of unreliable narration, which limits itself to pseudo-diegetic narratives, which are defined as “a second degree narrative brought up to the first level of the primary narrative and taken in charge by its narrator.”21_{This results in more than one}

narrative level in a film (pseudo-diegesis). In these films the pseudo-diegetic character-narrator is a creation of the primary cinematic narrative and the narrator of the second degree narrative, which is embedded in the primary cinematic narrative. It is a character in the film who narrates a story, hence the term ‘diegetic narrator’. If this diegetic narrator is convincing enough, the viewer will accept

18_{Ferenz, 2006, 134.} 19_{Ibid., 134.}

(12)

12

him/her as the creator of everything in this second degree narrative, such as the images and sounds, but also the mise-en-scène and the way these diegetic elements are shown to the viewer through, for example, framing and editing. This makes it possible for the viewer to blame this narrator, which is a fictional character in the story, for any inconsistencies in the second degree narrative that come to light over the course of the film. Inconsistencies can become clear because these pseudo-diegetic character-narrators can always be overruled by the cinematic, impersonal narrator of the primary cinematic narrative, the overruling principle that produces the fictional truth and is always reliable.22

Ferenz describes three ways in which the effect of pseudo-diegesis can be achieved. The first is to have a character that tells a story to another character. Over the course of the film there is a constant switching between the first and second narrative level, so that is always clear what the primary level is. This is the most commonly used form of pseudo-diegesis, which can be seen for example in The Usual Suspects (Singer, 1995). A variation on this form can be the use of multiple embedded levels which is done in The Prestige, a film that makes use of three narrative levels (and also features two pseudo-diegetic character-narrators). The second way to achieve the effect of pseudo-diegesis is to open the film in medias res and to return to the primary level at a later point (usually at the end) of the film. This can be seen in amongst others Atonement (Wright, 2007) and Fight Club (Fincher, 1999). The third form that Ferenz describes is one where the different narrative levels seem to flow into one another and the narrative switches are not marked clearly. An example of this form is the switches between the impersonal and personal mode in Spider (Cronenberg, 2002).23

Ferenz’s take on unreliable narration in film is limited. He excludes unreliability based on supernatural, horror or science-fiction elements, having the argument that all the textual incongruities in these films can be attributed to the genre of the film, by holding on to Yacobi’s generic principle. However, at this point I have to disagree with Ferenz, because narration is not the same as genre. Different genres can use different forms of narration. Just because a film happens to fall in the horror or science-fiction genre doesn’t mean that this film cannot also make use of narrational strategies that can be deemed as unreliable. There are a number of films that make use of both unreliable narration and

supernatural, horror, and/or science-fiction elements, for example The Sixth Sense (Shyamalan, 1999),The Prestige and The Others (Amenábar, 2001). Likewise, by clinging to the notion that

unreliability is limited to the use of an unreliable pseudo-diegetic character-narrator, Ferenz excludes a number of films that many would consider to be unreliable, because these film do not have diegetic narrators who are intentionally telling a false second degree narrative. There are plenty of films however, that do not so much make use of unreliable narrators but of unreliable narration.

(13)

13

Gregory Currie argues that it is indeed possible to have unreliable narration in a film without having a narrator. In the case of some films, he states, it is indeed possible to ascribe unreliability to an embedded narrator, as Ferenz does. However, there are films, which are considered to be unreliable, where this does not apply. In these cases, Currie argues, it is possible to ascribe unreliability to some complex intention.24_{He explains complex intention as follows:}

An agent can do something with an intention of the following complex kind: she creates or presents something which she intends to be taken as evidence of her intentions, and she intends that a superficial grasp of that evidence will suggest that her intention was Φ, whereas a better, more reflexive grasp of the evidence will suggest that her intention was Ψ.25

Currie calls this form of unreliability complex unreliability and states that we can speak of this form of unreliability when the narrative “has a tendency to mislead an attentive viewer with some experience of the relevant genre.”26_{In this case it is the narration itself that misleads the viewer by}

under-/misreporting. One of the ways this can be achieved is by staying close to one or only a few characters over the course of the film. Such narration can offer a limited perspective of the events in the film to the viewer. Even though this character is not a pseudo-diegetic character-narrator, the way he or she (mis-/under-) interprets, evaluates and reports things to other characters can be a significant influence on the viewer’s interpretation of the film.

Thus far I have made a broad distinction between two different forms of narration, namely between an unreliable narrator (the embedded pseudo-diegetic character-narrator, and unreliable narration (complex unreliability). However, unreliability can be divided even further, because there can be different reasons behind the unreliability in both cases. While reflecting on films that make use of complex narration Thomas Elsaesser concludes there are six frequently recurring situations, that could result in unreliable narration. These situations can lie at both the basis of unreliable narration through a pseudo-diegetic character narrator and complex unreliability, especially when they try to mislead the viewer by limiting the focus of the entire film to a limited amount of characters.

24_{Currie ascribes this complex intention to the implied author. However, whether or not it is possible, let alone}

necessary, to attribute an implied author to film, it is still possible to ascribe complex intention to the narration, whether that is ascribed to an implied author or the ‘real’ biographical author/filmmaker(s).

(14)

14

The first frequently occurring situation is a protagonist who is part of or witnesses events of which he/she does not get the meaning or oversee the consequences. The main question of the film (and the protagonist) is what exactly has happened. This can be achieved by a suspension of cause and effect or a reversal of linear progression. The second situation is a protagonist who cannot tell the difference between his/her imagination and reality. The protagonist is made to question what reality is, since there is no clear marking of the difference between the real and the imagined. Thirdly a protagonist can have a friend, mentor or companion who turns out to be a figment of that

protagonist’s imagination. The fourth situation has the protagonist asking who he/she is, what his/her reality is, and whether or not he/she is still alive. The fifth situation Elsaesser describes is when both the protagonist and the viewer are made aware of the existence of parallel universes over the course of the film, and that the plot has been based on a false cognitive or perceptual premise. Both the audience and the protagonist experience this as a (narrative) twist. Finally, the last situation occurs when a protagonist is persuaded by others that he/she is mistaken about the existence and/or disappearance of a person, while the protagonist keeps claiming that what he/she believes is true. Usually, a conspiracy against the protagonist is then discovered over the course of the film. A number of these situations can be combined within a single film, Fight Club, for example, combines the third and the fifth situation and The Sixth Sense the fourth and the fifth.27

Narrative twists

A narrative feature that is often seen in combination with an unreliable narrative is the narrative twist. This is the moment when the narrative reveals a piece of information which makes the viewer, and in some cases also the characters, re-evaluate everything they have seen before, which often results in drastic changes in the way the narrative is interpreted. In the introduction of their edited volume Was stimmt den jetzt? Unzuverlässiges Erzählen in Literatur und Film Fabienne Liptay and Yvonne Wolf relate the narrative twist to the combination of staging and editing and the limited perception of (a) character(s) in the film. They state that:

[d]urch subtile Mechanismen der Inszenierung und gezielte, lückenhafte Informationsvergabe wird die verschobene oder eingeschränkte

Wahrnehmung einer Figur zunächst mit einem Deckmantel scheinbarer Objektivität getarnt. Lange Zeit sind wir als Zuschauer gefangen im begrenzten

(15)

15

Blickwinkel der Figur, ohne uns dessen Gewahr zu werden. Zumeist übersehen

wir dabei den entscheidenden Schlüssel zum richtigen Verständnis […], der uns absichtsvoll vorenthalten wird.28

Once the narrative twist is revealed, Erlend Lavik explains, the viewer becomes aware of another fabula, which has been running parallel with the one the viewer was inferring up to the point, that makes everything in the film take on a new meaning.29_{These twists are often placed near the end}

of the narrative. This gives the viewer enough time to build up hypotheses about the outcome of the story during the film. With the revelation of the narrative twist the viewer is forced to adjust his/her hypotheses and go over the entire narrative again to construct the newly emerged ‘correct’ fabula. Lavik describes what it takes to create a successful twist film:

The success of twist movies similarly depend on the careful manipulation of syuzhet information—there is a fine line between revealing too much and too little. But what has to be kept just outside the audience’s grasp is no longer the solution to the mystery, but its very existence […]. In general, other factors being equal, the more clues the syuzhet contains without the audience being able to put two and two together, the more successful the movie.30

A successful narrative twist is built upon several preconditions and tactics. First of all it is crucial to establish certain rules about what is and isn’t possible within the story world. These rules are especially important when it comes to supernatural events. The clearer the rules, the more credible and consistent the revealing fabula. It is important that, even after the revelation of the narrative twist, all the previously presented syuzhet information should make sense in light of the newly revealed fabula. The timing of the revelation of the twist is also very important. Lavik compares this timing to the deliverance of the punch line of a joke. It is, of course, always possible that some

members of the audience predict a coming twist, but if the entire audience sees the twist coming, the effect is ruined thus the unreliable narrative is flawed.31

28_{Liptay and Wolf, Wass stimmt den jetzt?, 15. “}_{Through subtle mechanisms of staging and the arrangement of}

explicitly chosen, incomplete information, the delayed or limited perception of a character is disguised with apparent objectivity. For a long time we as viewers are trapped within the limited viewpoint of the character without becoming aware of it. Usually we are unable to see clues which could lead to a full comprehension [of the story] which are right in front of us.” [my translation]

(16)

16

Daniel Barratt describes how the viewer can remain blind to unexpected twists in the narrative. He examines the role of cognitive notions in what he describes as twist blindness. These notions, primacy, priming and schema, are all based on the psychological faculties of memory and attention. The scope and limitations of these faculties are often exploited by filmmakers in various ways to achieve certain effects, like narrative twists. To remember something properly it is important that there is as less distraction as possible and that something is repeated a number of times. This means filmmakers can manipulate how the viewer remembers certain (important) information by playing with these two factors. In the case of twist narratives the filmmaker can, for example, show an important piece of information once and only briefly. At the same time, or in the scene directly thereafter, the filmmaker can distract the viewer with other, less important pieces of information, placing the viewer’s attention elsewhere. Attention also tends to decrease over time, both on the level of one single scene and the course of the entire film. Therefore we tend to pick up less information at the end of a scene or at the end of a film. With regards to memory it is important to remark that the process of cognitive rehearsal is important to transfer information from the short term memory to the long term memory.32

Barratt’s theory draws on a number of principles described by David Bordwell in his constructivist theory on the film viewer’s psychological activity, which I summarized earlier in this chapter. One of the most important facts that a filmmaker can take into account when he/she wants to mislead the viewer about certain events in the film is, as Barratt explains, that we construct the fabula through the use of or available schemata. This applies to both individual scenes and the entire film and is based on how our memory works. Since our memory is not photographic we tend to remember only specific details of scene. We use our schemata to fill in the gaps of our memories and of the selective (and sometimes unreliable) information provided by the syuzhet. This phenomenon is called schematic elaboration. A phenomenon that is related to this is schematic distortion. This means that when a certain schema is applied based on a number of specific details, we also assume a number of other events to have taken place based on that schema, even though we have not actually seen them take place on screen.33_{Filmmakers can use these phenomena to make viewers believe certain}

events have taken place in a certain way, while in fact they haven´t.34

An important way the filmmaker can play with the viewer’s use of schema is to address the fundamental assumptions the viewer has about the nature of the diegesis, as was already mentioned by Lavik. Barratt states that “[i]n terms of perception, assumptions concern the basic properties of physical objects [and that i]n terms of cognition, assumptions form the default settings of various

32_{Barratt, Twist Blindness, 62-64, 68, 73.} 33_{Ibid., 65}

34_{Barratt uses the restaurant script as an example and explains that when we see a character paying at a}

(17)

17

schemata.”35_{For example, the viewer has certain assumptions about the basic properties of certain}

objects (e.g. they are solid and persists in space off screen), the ontological status of beings (e.g. whether or not a character is alive or dead, human or not) and whether or not the diegesis contains sci-fi or fantasy elements. Barratt states that these assumptions are fundamental and taken for granted, and also cites Bordwell, who states that these assumptions are usually only noticed when they are violated.36_{The filmmaker can play with these assumptions by violating them without the}

viewer noticing it. This happens for example in The Sixth Sense. Here the viewer assumes throughout the entire film that the main protagonist of the film, Malcom, survived the shooting at the beginning of the film, while in fact, as is revealed in the end, he did not (which is also the narrative twist in the film).

Another important method the filmmaker can use to mislead or distract viewers is to take advantage of the primacy effect, which, as we saw, has to do with the order in which certain information about characters or situations is presented to the viewer. The information that is given first determines how the viewer processes all following information. This can help the filmmaker present characters or situations in a positive or negative way. For example, when it comes to

presenting the personality of a character, the filmmaker can present a range of positive traits from a character before presenting some negative ones. This will lead the viewer to assume the character as a positive, ‘good’ person, albeit with some minor flaws, opposed to an essentially bad person. These processes work on several levels in a film, especially when it comes to fictional characters. Not only do they affect the ‘status’ of a character (e.g. alive/dead), but also the judgment of a characters

personality, which is undoubtedly important when it comes to identification with, and empathy and sympathy for, a fictional character. What is also important to note is that our cognition tends be conventional and therefore always tends to trust our first impressions, and goes for the most obvious explanation (on the basis of the schemata we automatically employ).37

In addition to this, the filmmaker can also both distract the attention of the viewer and interrupt the process of memorizing by obstructing the generation of inferences and/or hypotheses through presenting the viewer with either an informational load or an emotional load. By distracting the viewer and not allowing him/her to memorize certain events properly the filmmaker can hide important elements that must be in the film to keep the story consistent, but might give away the narrative twist if they are discovered to soon. Reviewing the film will prove these elements are indeed consistent in keeping with the narrative twist and the overall story, although the viewer did not see them the first time around. In the case of an informational load the viewer is distracted by the

35_{Barratt, Twist Blindess, 66.}

(18)

18

presentation of additional information he/she has to process, thus dividing the viewer’s attention and focus, thereby distracting them from what is really important. In the case of an emotional load the filmmaker can evoke a certain emotion in the viewer which can direct the viewer’s processing and distract the viewer from inferring what the consequences of the presented (elements in a) situation are.38_{These emotions also tend to linger for a while, so they influence both the processing of the}

scene they are evoked in, but also the processing of the following scenes.

A filmmaker can use the processes and methods described above in a number of ways. For example, one scene can provide an emotional load which already limits the forming of hypotheses in and of itself, the filmmaker can then follow this emotional load with an informational load which halts the processing of information in the viewer even more. It is also possible to use informational or emotional loads to halt or interrupt the process of cognitive rehearsal. The filmmaker can use the film’s style, which includes editing, cinematography and mise-en-scene, as a tool when making use of the primacy effect, for example by making it appear that two people are at the same place at the same time. This is made possible by the viewer’s inability to attend to both the form and content of a scene in the same amount, which makes the viewer attribute perceptions that are influenced by the film’s form to the content of a scene.39

The emotional effects of unreliable narration

According to Volker Ferenz, films that use unreliable narration can be perceived as emotional roller coasters which can elicit some unusual emotional responses in their audience. This is because these films create psychologically rich situations. They do this by creating an intimate connection between the viewer and a character, which then turns out to be unreliable. The stronger the connection between the viewer and the character, the bigger is the emotional impact when the unreliability is revealed and the more effective is the narrative twist. This, as we will see, can bring about both positive and negative feelings in the viewer, which makes that the emotional responses to films that use unreliable narration can be experienced as more intense then films that use more classic forms of narration. Before delving deeper into the specific emotions of identification, empathy and sympathy, and how these emotions relate to unreliable narration, I will end this chapter with a brief look at the emotional effects of this narrative form.

38_{Emotions are, of course, subjective. Therefore some viewers may not experience the emotions that were}

intended by the filmmaker. However, as long as the viewer is pre-occupied with emotions of any kind the emotional load is still a successful tool in distracting the viewer.

(19)

19

While watching a film that employs a protagonist who turns out to be an unreliable one, the viewer, from the moment of realizations onwards, constantly has to check his/her feelings towards that character. This demands an active attitude from the viewer with regards to his/her emotions. This can lead to both positive and negative feelings, as Ferenz explains. He quotes Ed Tan, who states that “understanding a character is in itself pleasurable, not least when the complexity of the narrative film is rooted in the development of a character.”40_{According to Ferenz this process can become a source}

of frustration in films that use unreliable narrators because of a lack in character coherence. This can lead to feelings of disappointment, frustration and the feeling of being cheated. Yet, when the viewer finally ‘figures out’ the character this can be a source of pleasure and satisfaction, and the viewer may feel like he/she has taken charge of the film as a whole.

The fact that the feelings of the viewer for a particular character keep changing may cause some unusual responses. One example can be the formation of perverse allegiances, where the viewer aligns him-/herself with a character that is morally suspect. This can happen because the (unreliable) narrator which shares his/her perspective with the viewer will often have the tendency to present him/herself, consciously or unconsciously, in a positive way, which can increase the chances of identification, empathy and/or sympathy on the side of the viewer.41_{The viewer may experience the}

discovery of the true nature of the character as a negative experience, but, on the other hand, the viewer may also take pleasure in aligning him-/herself with a morally suspect character, because this is something that is not something one can easily do in everyday life. Film, however, is a more risk-free environment. Related to this ethical side of unreliability is the fact that unreliable narrators tend to lay bare the moral predispositions of the viewers. This can lead to the viewer feeling offended by

unreliable narrators who adhere to a different ethical standard than their own.42

Feelings of identification, empathy and sympathy are also largely influenced by the manner (by the narrator him/herself or by a higher narrative level) and moment (the fact and extend may be revealed in different moments) in which the unreliability is revealed. Ferenz states that once the unreliability is discovered the positive feelings that were elicited in the viewer, before the moment the unreliability is revealed, are likely to diminish somewhat. Nevertheless, I would argue that this need not always be the case. Ferenz already states that the manner and moment of the revelation of the unreliability is important. However, Ferenz’s definition of narrative unreliability entails that the narrator always intentionally deceives other characters and/or the viewer. This matter of intentional deceit obviously influences the way the viewer reacts to the revelation of the unreliability. Besides the fact that I think narrative unreliability includes more than just willfully lying protagonists, what also

40_{Ferenz, Mementos of contemporary American cinema, 269.} 41_{Ibid., 266.}

(20)

20

seems of key importance to me, especially in the light of processes such as identification, empathy and sympathy, is why a narrator turns out to be unreliable. It might be possible that in some cases the reason behind the unreliability may even increase the identification/empathy/sympathy the viewer has with/for a character. For instance, identification can ensue when the viewer imagines he/she would have done the same as the character if he/she were placed in the same situation, with the same amount of knowledge. Sympathy may occur when the viewer pities the character for having been placed in a situation which required the character to be unreliable, whether they wanted to or not. The fact that an unreliable narrator recognizes his/her own unreliability early on in the narrative and tries to better him/herself may also elicit positive feelings in the viewer. An example of the viewer identifying in certain aspects/ sympathizing with a character could be the nameless protagonist from Fight Club. The viewer may identify with the character on a psychological-affective level or even empathize with the character, because, towards the end of the film, they share a sense of confusion about the manner in which certain events took place. Since the nameless protagonist is as clueless as the viewer as to what is going on at first, and discovers the truth at the same time as the viewer, (presuming the viewer experiences the narrative twist at the intended moment in the film) this could create a bond between the protagonist and the viewer based on this shared experience. This could lead the viewer to sympathize with the protagonist in the sense of pitying him, feeling hopeful

towards him that, after all this, his problems are resolved, and that he will now be able to lead a more normal life.

In this chapter I have looked at the concept of narrative unreliability in film. I began the chapter with a look at the cognitive processes that are needed to understand cinematic narratives. These processes are not only unconsciously used by the viewer to understand the narrative, but in an unconventional form such as the unreliable narrative they are also actively brought to the attention of the viewer. Then I elucidated on how the rise of films that use unreliable narration of the past two and a halve decades can be placed within the broader trend of the rise of complex narration in film. Next, I looked at the different forms an unreliable narrative can take. A distinction was made between unreliable narrators and unreliable narratives. Thomas Elsaesser explained that there can be several situations in a film that can lead to unreliability. Often (not always), films which feature an unreliable

(21)

21

(22)

22

Chapter 2 Identification, empathy and sympathy with fictional characters

After looking at unreliable narration and narrative twists in the previous chapter, it is now time to focus on the other theoretical concepts of this thesis, namely identification, empathy and sympathy. The use of the term identification in relation to film viewing is not restricted to the academic world, on the contrary, it is used frequently by film viewers as part of the description of their viewing

experience. It is often seen as an important factor in determining one’s opinion of a film and it plays a big part in the emotional experience of film viewing. Ed Tan states that identification is a response that can happen to anyone who watches a film.43

Before I go any further, I will first briefly summarize why I think these notions are deemed to be important to study both in general and within film studies. In explaining why the study of

identification (with fictional characters) is important, Jonathan Cohen, in his article Defining

Identification: A Theoretical Look at the Identification of Audiences With Media Characters, refers to a well-known study by Erik H. Erikson (Identity: Youth and Crisis, 1968). This study claims that

identification contributes to the development of self-identity, since self-identity is related to our perception of others and to how others view us. Thus, Cohen states, media images are related to self-identity because identifying with a media character (such as a fictional character in a film) allows us to experience social reality from a different perspective, thereby shaping the development of self-identity and social attitudes.44_{According to George Herbert Mead, identification is crucial to the}

socialization of children and adolescents, and the development of personal and social abilities throughout one’s life. For him people can, by identifying with both other people and fictional characters, “try on alternative ideas, images, attitudes, and identities,” a view that is also shared by Erikson.45_{Because it involves internalization, identification can have long-term effects when it is done}

repeatedly. Cohen states about this idea that “[t]hrough identification with characters in books, films, and television we extend our emotional horizons and social perspectives.”46_{I would argue that this}

makes media characters very influential, especially in the modern day society in which we are

surrounded by different media almost constantly. Besides the possible influence these processes have on the (emotional) development of us as human beings, on a smaller scale it is also important to research them within film studies for the large influence they have on the (emotional) viewing

43_{Ed Tan, Film als emotiemachine, 223.} 44_{Cohen, 2001, 245-246.}

(23)

23

experience of a film. That watching a film can have a deep emotional impact on the viewer, according to Ed Tan, is a fact that probably everyone who has ever watched a film has experienced to some extent. One of the main reasons for this is that the artifact we are looking at, in this case a film, is made especially for its audience.47

Despite (or maybe because of) the fact that the term is widely used, the notion of identification with fictional characters is extensively debated within film studies, and one of the main questions is how identification with fictional characters should be defined. According to Cohen the concept of identification “is understood in a variety of ways by different theorists and […] this confusion has inhabited the development of a comprehensive theory of identification and its consequences.”48_This

confusion about terminology also extends to everyday life, as Noël Carroll states:

Indeed, whether or not the notion of character-identification is merely a metaphor or is meant to be a literal description of a mental state is generally not determinable in the context of daily conversation. […] But the term is often used—even by professional critics—in a way that fails to specify exactly how we are to characterize the mental state to which the speakers are referring.49

In a broader perspective the concept of identification with media characters can also be related to the debate surrounding the paradox of fiction, which deals with the question of whether or not it is even possible for readers/viewers to feel genuine emotions towards fictional characters and situations. Because this debate is not the focus of this thesis however, I will assume here that it is indeed possible for film viewers to feel genuine emotions that concern the fictional characters in a film. Instead I will focus on finding clear definitions of the terms identification, empathy and sympathy, which will help us understand the impact these phenomena have in the viewing experience of

complex narrative films that make use of unreliable narration.

In spite of the fact that it is a widespread notion, a number of well-known film scholars have rejected the term identification and have instead proposed their own terms and models to describe the relationship between viewer and character.50_{I will look at some of these different perspectives to}

see which of them are the most suitable to describe the emotions that identification, empathy and sympathy evoke in the viewer. First I will briefly look at a narrow definition of identification, presented

47_{Tan, Film als emotiemachine, 3-4.} 48_{Cohen, 2001, 246.}

49_{Carroll, The Philosophy of Horror, 89.}

(24)

24

by Jonathan Cohen. It is exactly such narrow and limited definitions like these that have led several scholars to abandon the term identification altogether. Noël Carroll and Murray Smith have reacted to definitions such as these by presenting alternative terms and/or models which, according to them, describe the relationship between film viewer and character more accurately. Unlike Carroll and Smith, Berys Gaut does not reject the term but instead chooses to redefine it so it becomes more practical. Although all of these theories and models primarily focus on the term identification, the notions of empathy and sympathy are also interwoven in them. These notions are most commonly defined as feeling with a character (empathy) and feeling for/about a character (sympathy). In this chapter I will therefore also look at the roles these two concepts play in the different models

describing the relationship between the viewer and fictional characters. The goal of this chapter is to get clear definitions of all three concepts (identification, empathy and sympathy), in order to be able to apply them to the characters and viewers of films that use unreliable narration. By means of the application of these clearly defined terms I will, through the analysis of the three case studies, be able to see how these phenomena can be used by filmmakers to achieve specific effects in unreliable narrative films.

The narrow definition of identification

Cohen describes identification as an imaginative process that the audience experiences as a response to the characters that are presented to him/her in a media text (such as a film). He states that it demands extreme absorption into the text on the part of the viewer. A key element of Cohen’s definition of identification is that it results in “a cognitive and emotional state in which the audience member is aware not of him-/herself as an audience member, but rather imagines being one of the characters in the text.”51_{The intensity of this process is not constant, but varies throughout the film.}52

An important basis for this process is for the viewer to adopt the character’s goals. The events in the plot are then comprehended in accordance with these goals and the viewer experiences different emotions, depending on whether or not the character is hindered in accomplishing his/her goals.Also, despite the fact that the audience member often has a different amount of information than the character, identification is still possible. The knowledge of the viewer is processed from the

perspective of the character, which leads to empathetic emotions from the viewer for this character. Cohen’s definition of identification focuses on feeling with the character (empathy) instead of about

(25)

25

the character (sympathy). His 2010 study with Nurit Tal-Or Cohen elaborates on empathy and identification, stating that the process of identification consists of two components, namely an affective empathy component (feeling empathy and affinity towards a character) and a cognitive empathy component (adopting the character’s goals and point-of-view).53

Cohen gives us two main attributes of his concept of identification. First it is “defined not as an attitude, an emotion, or perception but rather as a process that consists of increasing loss of self-awareness and its temporary replacement with heightened emotional and cognitive connections with a character.”54_{Second, identification is the result of a response to textual features that are meant to}

provoke identification in the viewer. This means that characters are specifically created for this

purpose; diegetic characters are so called identification targets. As I already mentioned in the previous chapter, Ed Tan also points out that the artwork we are looking at (which includes the characters in a film) is made especially for its human audience. In their 2010 study Cohen and Tal-Or mention another important attribute of identification, which is the fact that it is focused on a specific character, instead of on the narrative as a whole. This means that the viewer experiences the text through a character, and the identification process thus determines the way the viewer experiences the entire text.

According to Cohen there are several factors which could lead to identification with a character. It can be initiated because of a certain production feature which leads to the adoption of the perspective of the character by the viewer, a viewer’s fondness for a particular character or the fact that the viewer becomes aware of certain similarities between him-/herself and a character. The process of identification can also be interrupted or ended by a number of factors, both external and textual, or by the ending of the narrative. Identification may result in both positive and negative feelings on the side of the viewer, depending on the achievement of certain goals and the nature of the character with which the viewer has identified.55

There certainly seem to be some valid points in Cohen’s definition of identification. It does make sense that the overall experience of the narrative is in part determined by the fact that the viewer identifies to a certain extent with a specific character in that narrative, and that the characters in this narrative have been specifically created for this purpose by the creators of the narrative. However, it also contains some aspects which seem quite radical. The main one is the fact that the viewer supposedly gives up his/her own identity in its entirety for that of the character during the viewing process. I

53_{Cohen and Tal-Or 2010, 404. Cohen seems to contradict himself here. In his 2001 study he clearly states that}

affinity is a different emotional connection to a character than identification, which involves more distant between the viewer and the character. However, in his 2010 study he makes affinity a part of one of the two key components of identification.

(26)

26

agree with Cohen that it is possible for the viewer to have an emotional and cognitive connection with a character in a film. Nevertheless it seems unrealistic to me that in this moment it is possible for the viewer to give up his/her own identity entirely in the way Cohen describes. To me, it seems, it would take a lot more than identifying with a fictional character to make one forget one’s own identity, thoughts, state of being, and surroundings completely.

It is no wonder then, that in the face of such a narrow and radical definition a number of film scholars, amongst them Noël Carroll and Murray Smith, have rejected the term identification. Instead they plead for alternative terms which, according to them, are more fit to describe the relationship between viewers and fictional characters. Noël Carroll especially argues fervently against a definition of identification as is given by Jonathan Cohen. Instead he presents his term of assimilation. Murray Smith also presents his own model of engagement to describe the relation between the viewer and fictional characters. I will now examine both of these theories to see how they can help in establishing a clear model for analyzing the function of identification, empathy and sympathy in films with

unreliable narration.

Noël Carroll’s theory of assimilation

In his book The Philosophy of Horror Noël Carroll discusses the notion of identification in relation to horror films, but also film in general. Carroll’s is that the term is not clearly defined and, in everyday use it could mean a number of things, such as liking the protagonist or sharing the protagonist’s values. However, when commentators speak of identification, Carroll says, they often have in mind a more radical notion, namely that of being identical to the character. In this case the viewer merges with the character thus becoming the protagonist of a fiction. The similarities with Cohen’s definition are obvious here. The main reason I include Carroll here is because I agree with his strong

argumentation against such narrow definitions of identification as proposed by for example Jonathan Cohen. I will first look at Carroll’s objections to this definition and then at the alternative term he proposes.

According to Carroll, the definition of identification proposed by Cohen, which is based on a notion of illusion, makes no sense when one looks at the behavior of viewers of horror fiction and fiction in general. Consumers of fiction give every indication that they know they are not the protagonist. Thus, according to Carroll, “if the notion of character-identification is to make sense, it cannot be based on postulating an audience illusion of being identical with the protagonist.”56_The

(27)

27

same goes, Carroll states, for the idea that the viewer only duplicates the viewer’s emotional state. Most of the responses viewers have to fictional characters are no such duplication. There are several reasons for this. First, the viewer often has more information than the character, which causes him/her to feel different then the character does about (upcoming) events on the basis of this information, for example when the viewer knows the character is in danger but the character does not. This reason alone seems proof enough that a definition of identification which requires a symmetry between the feelings of the character and the viewer does not describe the relation between the character and the viewer very well. Cohen argues that additional information the viewer has is processed from the perspective of the character. This seems not a very convincing argument however, because even though additional knowledge may be processed from this perspective by the viewer, this does not automatically mean that this results in symmetrical feelings between the character and the viewer.

Another reason Carroll gives as to why the responses from the viewer and the character to story events differ is that the internal motivation behind these responses is different. While the viewer’s emotional response is based on a context of entertainment, the emotional response of the character is based on belief. The response of the audience member also concerns another person, namely the character, and thus can be said to be altruistic, while the response of the character concerns him/herself, and is therefore egoistic. On the basis of this argument Carroll describes the response of the viewer as sympathy. Because of these differences in the responses of the viewer and the character Carroll concludes that “it very well may be the case that character-identification never supplies an account of the audience’s relation to the protagonist.”57_{He also states that if one were to}

say that identification with fictional characters only entails a partial duplication of emotion in the viewer, it would still not do as a definition because “if the correspondences are only partial, why call the phenomenon identification at all?”58_{We can conclude from this argument that according to}

Carroll the process of identification cannot be partial. As I will show later on in this chapter, Gaut begs to differ.

Carroll stresses that while it is possible for a viewer to have parallel emotions to the character, this does not entail identification. The viewer, unlike to the protagonist, responds to what Carroll calls ‘the larger situation,‘ which is defined by the fact that the viewer is outside of this situation while the fictional character is in it. This means that the response of the viewer includes both the character and the character’s emotional response. Here we see, once again, the difference between the altruistic response of the viewer and the egoistic response of the character. The final argument Carroll poses against the narrow definition of identification, which requires a fusion between the viewer and the

(28)

28

character which results in in a symmetrical emotional response, is that the notion seems to be based on what he calls “a very radical sort of egoism.”59_{By this he means the idea that people can only}

respond emotionally to something that pertains their self-interest. However, according to Carroll there is enough evidence that this is not the case and that people are perfectly capable of responding emotionally to situations that concern others but not themselves.

Carroll argues that is would be best not to use the term identification anymore, at least not within an academic context, to prevent confusion on the subject. Instead, Carroll proposes, in the case of fiction “what we do is not identify with characters but, rather, we assimilate their situation.”60_{The term}

assimilation is Carroll’s answer to the debate on character identification. An important part of this activity entails having an understanding of how the characters assess the situations they find themselves in. But besides understanding the character’s internal perspective, the viewer also takes an external view and assimilates aspects of the situation they see onscreen that are not focused on the protagonist. Therefore the viewer also sees the situation which involves the protagonist.

According to Carroll, identification, in the sense of a merging between viewer and character resulting in symmetrical emotions, is not necessary when it comes to understanding fictional situations. The only thing that is, is understanding why the reaction of a specific character makes sense in a specific situation, which results in different emotions in the viewer then in the character. Carroll concludes therefore that “[t]his logical asymmetry indicates that identification, a symmetrical relation, is not the correct model for describing the emotional responses of spectators.”61

Murray Smith’s model of character engagement

Murray Smith also proposes an alternative model for the relation between the viewer and fictional characters in his theory of character engagement. This theory is based on the concepts of sympathy and empathy. Since most theories tend to focus more on the concept of identification and less on the concepts of empathy and sympathy, I will use Smith’s theory to delve deeper into those terms, especially sympathy.While Smith sees these two faculties as the things that together make up what is commonly called ’identification’, I find them helpful as an addition to the theoretical framework I am building. As I will show in the next part of this chapter, empathy and sympathy can operate beside identification. In the third chapter of his book Engaging Characters: fiction, emotion and the cinema

Open Your Eyes, and Watch Closely