163 Chapter 4 Emotion Expressions The first important gift for which we have film to thank was in a certain sense the rediscovery of the human face. Film has revealed to us the human face with unexampled clarity in its tragic as well as grotesque, threatening as well as blessed expression. The second gift is that of visual empathy: in the purest sense the expressionistic representation of thought processes. No longer will we take part purely externally in the workings of the soul of the characters in film. We will no longer limit ourselves to seeing the effects of feelings, but will experience them in our own souls, from the instant of their inception on, from the first flash of thought through to the logical last conclusion of the idea. Fritz Lang1 Character Emotions
164
Many cognitive film theorists have argued that film and emotion are closely connected, yet they tend to be interested more in spectator emotions than in character emotions, or to be more precise, their analysis of character emotions is focused on their role in generating spectator emotions.2 As a device of character construction, emotion expressions (principally of the voice and face) are another versatile tool that demand to be considered in their own right. And like typing and mindreading, emotion expressions can be shaped toward aesthetic ends. In this chapter I argue that expressions of basic emotions such as anger and fear are universally produced and recognized, and implicit in this discussion is the notion that one appeal of cinema is its ability to represent these expressions that transcend time and culture. But I also argue, more importantly, that emotion expressions can be understood only in a social or narrative context, and that filmmakers can manipulate both the context and the expressions to generate interest in character. It follows, naturally, that one feature of many films with an interest in character, such as American independent films, is their interest in exploiting emotion expressions in relation to narrative situations. As I argue in relation to Welcome to the Dollhouse and Hard Eight, carrying over the discussion from Chapter 3, one especially significant technique is the withholding of conventional expressions of emotion or of reliable contextual cues
165
necessary for the recognition of clear character emotions. As with those films’ efforts to frustrate mindreading and attribution processes, the careful frustration of our attempts at emotion recognition also produces effects of complexity and depth and demands a more active engagement with the construction of character. It also functions within the basic viewing strategies outlined in Chapter 1, demanding the exploration of identity, making narrative form playfully engaging, and offering a contrast with the comparative simplicity of mainstream cinema’s characters. So as with the other aspects of social cognition I have introduced, emotion expressions are natural processes that are exploited by a cultural form.
The Face The human face has always been a key feature of cinema’s appeal to filmmakers and spectators alike, one source of its special powers. This is hardly surprising, given the face’s significance as the emblem of our very identities. If you ask to see a picture of someone, you typically aren’t interested in an image of their teeth, knees, or liver. When you look at a picture of your face, you say, “That’s me!” The face is the only part of a person that can be abstracted in this way from the rest of the body; it is the best synecdoche we have for ourselves.
166
Scholars have long recognized the communicative power of faces. Study in this field dates to the enormous interest in physiognomy in ancient Greece and “facereading” even before that in China and the Near East.3 The study of physiognomy, which the Oxford English Dictionary defines as “the art of judging character and disposition from features of the face,” has long been recognized as pseudoscience, but interest in the face as an index of mental states has never abated. The Darwinian study of facial expressions as a product of evolution long ago filled the place once held by physiognomy and face reading, though more recently research has shown that social perceivers agree about links between facial features and character traits, which may sometimes even have some validity.4 From a Darwinian perspective, interest in faces is not as the outward expressions of “character” but as a primary means, along with our bodies and voices, of expressing emotions. Indeed, from such a perspective, emotions and facial expressions have a close evolutionary connection.5 Many of the emotionexpressions that are universally recognized, such as baring teeth to show anger and opening the mouth to show surprise, have a functional purpose that precedes the communication of emotional states. Baring teeth threatens an enemy, while opening the mouth is part of taking in breath. Each of these actions is practical given the situation in which one feels either angry (because one
167
has been wronged and will have to defend oneself) or surprised (breathing in will facilitate alertness to the cause of the surprise and readiness in response to it). Darwin theorized that emotions preceded the universally recognizable facial expressions that express them. Emotions themselves have adaptive value: it pays to be afraid of things that could harm you, to be angry at those who wrong you, to be disgusted by things that could make you sick. The facial expressions that accompanied these emotions, which originally had only practical functions, became habituated over time to the point that they took on the significant communicative role they have always since had. The ability to communicate our emotional states in turn is greatly beneficial to any species that requires the extensive social organizations characteristic of primate societies. Being able to communicate your feelings to your conspecifics gives you a definite advantage. One of Darwin’s arguments in his classic work on this topic, The Expression of the Emotions in Man and Animals, concerns the universality of emotions. There has long been debate over whether specific emotions themselves, and also whether the facial expressions of those emotions, vary from culture to culture, or whether they are universal or pancultural. As an evolutionary theory, Darwin’s argues for the universality of emotions. More recent research has basically confirmed this position, identifying several
168
basic emotions and facial expressions of them that are recognized panculturally, even by members of preliterate tribes who have had very little or no exposure to Western culture and media. The seven emotions, as Paul Ekman and his fellow researchers proposed to name them, are happiness, surprise, fear, sadness, anger, disgust/contempt, and interest.6 Ekman and his colleagues went farther than just ascertaining whether and how many facial expression are universal. They set about to code the muscles of the face to better understand how each facial expression is produced anatomically. They found that the muscles can be divided into action units (AU’s), of which we have 43, and identified about 3,000 meaningful combinations of these 43 action units. The catalogue of all of these facial actions and their combinations is called the Facial Action Coding System, or FACS. The few hundred people who can read the face using FACS report that their interpersonal experience is completely transformed.7 And apparently, animators at Pixar and Dreamworks use FACS to create facial expressions for characters in feature films.8 What about cultural specificity? What accounts for differences among cultures in the expression of emotion? The standard account in the facial expression literature holds that what varies is not the experience of emotion itself, nor the facial expressions corresponding to basic emotions in different cultures. It’s not as though a smile of joy for Westerners means terror in some other culture. What does vary is customs for who may
169
display emotions and in what situations. These are called display rules.9 People often try to display an expression for an emotion they are not experiencing, or to mask one emotion with another. For example, if you are trying to tell a lie or to smile for a picture you might feel one emotion but display—or attempt to display—another.10 To test this hypothesis, Ekman ran an experiment using Japanese and American subjects.11 It is widely believed that in Japanese society, there are cultural expectations regarding the facial display of negative emotions (which are also gender and age specific). In particular, Japanese are known to mask negative emotions with smiles, and as a result the smile means something different in Japanese culture than it does in the West, because Japanese expect many smiles to be less than genuine.12 In the Ekman experiment, a “stress” film is shown to two groups of students, one a group of American students in California and one a group of Japanese students in Tokyo. This is a film made up of things that are difficult to watch and that leave ordinary viewers pretty shocked. They showed each film to the students twice: once with the student alone in a room (but observed through a hidden camera), and once with the student observed by a technician (of their own culture) in a white coat and carrying a clipboard. Ekman found that when alone in the room, the American and Japanese students reacted with the same facial expressions. But when another person was present, the Japanese subjects attempted to
170
mask their emotions with smiles. The display rule in Japanese society is to hide or mask emotions when in the presence of authority figures, but this is not the rule in America. However, the Japanese subjects were betrayed by their faces, as many people find it impossible to effectively hide strong emotions of disgust or fear and Ekman, using the FACS, could even detect indications of emotions under the masking. Indeed, people are typically fairly proficient at distinguishing real emotional expressions from sham emotional expressions. If we were more proficient, we would be better detectors of lying and cheating, but in general we are able to distinguish a true (or “Duchenne”) smile, produced by genuine amusement, from an affected one, put on when asked to pose for a picture. A Duchenne smile combines a movement of muscles around the mouth (called a “simple smile”) with a movement of muscles around the eyes. A simple smile is easily recognized as nongenuine.13 More recent research has identified a more finetuned sense of cultural variability of emotional expression. The overwhelming majority of studies of emotional universality find that subjects identify emotions displayed by someone of a different culture at a rate greater than chance, making the case for some degree of universalism incontrovertible. Yet they also find that ingroup subjects identify each other’s emotion expressions at a higher rate than those of outgroup members. Outgroup expressions are more accurate
171
when the different cultures are closely connected (e.g., by sharing a national border). Within a given culture, minority subjects judge majority members’ expressions more accurately than the reverse. Indeed, minorities judge majority expressions more accurately than majorities judge each other! This makes sense intuitively: the success of minority members is increased as they are better able to interact with majority members.14 So this suggests that rather than absolute, the recognition of facial expressions of emotion is modestly universal. Different societies do not possess totally different systems, but neither are they identical.
The Voice The representation of the voice is another cinematic narrative technique with functions and possibilities similar to those of the face, yet historically the voice has been given less attention by film critics, theorists, and historians. This lesser emphasis may be explained by two factors. First, cinema existed for three decades before sound recording and sound image synchronization became standard, so the representation of the voice did not seem “cinematic” enough to warrant substantial attention as a means of storytelling. Second, the voice is not uniquely cinematic; music, literature, and drama used voices for
172
thousands of years before cinema was invented. So paying attention to the storytelling possibilities of vocal expression in cinema did not help differentiate cinema from its rival arts at a time when this kind of differentiation was a paramount interest of film critics.15 Like the face, the voice is an emblem and index of an individual’s identity. Like the face, the voice is a means communication that facilitates social organization, and like the face, some vocal expressions of emotion are recognized crossculturally.16 The most commonly identified emotions recognized vocally across cultures include anger, fear, sadness, and joy.17 There is evidence that the physiological and environmental components of emotional experiences have a causal relation to vocal characteristics: increases in rates of breathing and blood flow when a subject is taken by surprise affect the voice. Yet cultural specificity also has its part in this story. The display rules that apply in explaining the cultural variability of facial expressions affect vocal expressions in more or less the same way in the same circumstances. A person smiling to mask her fear would also try to speak in even, measured tones and not in a breathless fit of anguish.18 The evolution of vocal expressions of emotion runs parallel to the evolution of facial expressions of emotion. Darwin considered facial and vocal expressions to be equally significant. Vocal expressions of emotion, like facial ones, are products of natural selection. Many species communicate vocally in ways we might consider simplistic by
173
comparison with our own species’ vocal abilities, yet, as Darwin argued, our nonverbal vocal expressions evolved from these ancestors of our species. Some of the sounds produced by the human vocal apparatus, such as shrieks of pain, are produced the same way in many other species. But only human beings have the control over the acoustic properties of their vocalizations necessary to be able to sing and to communicate using language. We often say that when someone is experiencing a given emotion, regardless of what they may say or how they may look we can “hear it in their voice.” Research confirms that this is indeed the case: the human voice functions to express emotions regardless of which words are spoken. Listeners can detect basic vocal emotions when produced by actors, with some exceptions.19 We express emotions vocally (nonverbally) by varying our voice’s acoustic properties; as with facial expressions, we are able to recognize crosscultural vocal expressions of basic emotions such as anger and sadness. Researchers into nonverbal communication of emotion have coded the acoustic properties of some vocal emotion expressions much in the same way as FACS correlated facial musculature to emotion expressions. The most important acoustic aspects of the voice for emotional expression are loudness (intensity, measured in decibels), pitch (fundamental frequency, measured in
174
hertz), and time, which includes both rate of speech and pauses between words.20 Each of these can be further broken down into components, and each may vary to greater or lesser extents. For example, anger is generally characterized by a quick rate and a high frequency, and by high variability in both of these qualities. Sadness is generally characterized by a slow rate, low frequency, low intensity, and little variability of these qualities.21 For joy, sadness, and anger, recognition of vocal expressions is four or five times greater than chance.22 Yet some emotions are more difficult to isolate in discrete vocal expressions: disgust is recognized at a rate scarcely better than chance. However, disgust may be a special case, as it is less likely than other emotions to be expressed in ordinary conversational speech. Disgust typically comes out in very brief “affect bursts,” which combine vocal and facial expressions. A typical case would be making a face and exclaiming a monosyllabic expression of displeasure.23 Still more modes of expression exist. Gesture or body language is another means of emotional display that psychologists have studied, though to a lesser degree than voices and faces. Of course, the body is a significant tool in the actor’s kit. Yet more and more, the mainstream film conventions of framing and cutting emphasize faces at the expense of bodies.24 If John Wayne were acting today, his cowboy gait would be less prominent, yet we would see more of his friendly smile and his angry glare. Bodily expressions of
175
emotion are a significant form of cinematic narration and characterization but the literature on this topic is rather thin, and I leave further discussion of it for future research.
Emotion Expressions : A Closer Look There are three basic claims arising out of this discussion of research on expressions of emotion. The first is that the basic emotions themselves are universal: no matter the cultural or historical specificity of a person’s background, she still experiences the same happiness, fear, disgust, etc., as any other person. There are no remote tribes in which happiness itself is unknown, or in which they have a different basic emotion, distinct from happiness, unknown to Westerners. The second claim is that basic expressions of happiness (smiling), fear (furrowing the brow), disgust (pursing the lips), etc., are also crosscultural. There are also no remote tribes in which no one ever smiles, or in which everyone uses some expression we have never seen to communicate happiness. Indeed, the notion of display rules suggests that not only are expressions of emotion universal, but that people know how to produce many of them independent of experiencing the emotions that they communicate. If
176
asked, you could make a happy, fearful, or disgusted face without having to think much about how to do it. Both of these universality claims have significant implications for understanding how characterization works in cinema, as we shall see. 25 A third, implicit, claim of this research is more basic yet. The face may or may not be an accurate index of a person’s character or disposition, as physiognomists thought, but it is a reliable source of information about a person’s emotions. This is the kind of information people seek when they scan the faces of others. By looking at someone’s face, it’s hard to tell with much certainty whether or not they are virtuous or illhumored. You can generally tell their age, ethnicity, and sex (though not always). But what you are generally looking for in a face, within the parameters of a given context, is information not only about personality traits and the intentional states, but also about affect states such as emotions and moods. These are what faces are best at communicating. This is also a universality claim: in all cultures, people treat faces as emotionindicators and scan faces for feelings. And most significantly, it follows that the ability to read emotions in faces must universal, just like the ability to express them. Both the sending and receiving of facial emotion information is a product of natural selection.26 The same claim applies in a slightly modified form to vocal expressions. The voice is used to communicate more than just emotion; the words chosen may convey
177
many kinds of meaning. So too the face may communicate several kinds of information (a wink might mean “yes,” a wince might say “that hurt” or “this is embarrassing”). But, among other things, the nonverbal aspect of speech functions as an index of emotions. In this way it is similar to music; both voices and music communicate emotion even without the meaning conveyed by words. The Beatles’s “Help!” is a cheerful song in spite of its desperate lyrics. There is also evidence that the acoustic properties corresponding to particular emotional meanings in music are the same as those applying to vocal expressions (i.e., the high pitch and intensity characteristic of fearful speech is understood in music to be fearful—think of Bernard Herrmann’s Psycho violins).27 These claims have significant implications for the study of cinematic narrative in its specificity as a distinct storytelling medium.28 Unlike literature, movingimage media such as cinema use images of faces as a primary narrative technique. While there are, of course, many films that contain no closeup images of faces, rare is the narrative feature film that eschews the face. To make such a film would be to engage in a kind of avant garde practice, an experimental effort to break the rules and see what happens. There is also a distinction here to be made between movingimage media such as film, television, and video games on one hand, and theater on the other. Since the introduction of moving pictures, the liveness and immediacy of theater have been its selling points. But what
178
cinema lacks in these qualities, it makes up for in its ability to represent faces on a large scale. Filmmakers rely heavily on images of the face to communicate emotion information, and to generate emotional response in spectators.29 The universality claims above explain, to a large extent, the power of movies to appeal across cultures and historical periods. Spectators approach the image of the face primed to recognize the same emotions, in the same expressions, regardless of their background. Of course, literary narrative creates emotions for characters too, but by other means. The immediacy and directness of the face make for vivid and direct characterizations and allow the audience to connect with the characters’ interior states in a powerful way. However, the literary comparison illustrates well that there are other means of representing a character’s emotional experience, such as verbal description (much more common in literature than film or television) and folkpsychology inference. Movingimage media make use of these techniques too, but the availability of images of the face make them secondary. Like facial expressions, vocal expressions are a more direct and vivid means of characterization than is available in literature. Yet there is no argument from medium specificity to be made regarding the voice. The silent era produced many masterpieces
179
that relied on an aesthetic of voiceless performance. Since then, however, the aesthetic has changed to the point that a narrative feature film without voices would be highly unusual.30 The contemporary aesthetic, which is the basis of cinematic characterization, places the face and voice at the top of all stylistic hierarchies. Faces (and by extension, bodies) are the most important component of mise en scene and are the paramount concern of cinematography, especially in terms of framing. Voices are the most important sound component, taking precedence over music and “noise.” In the vast majority of scenes, filmmakers cut to follow both facial and vocal cues above all other forms of narrative information.
Problems in Expression Recognition Underlying the above discussion are several assumptions about expressions of emotion that must be qualified to understand how such expressions function both in real life and in fictional representations. One assumption is that each expression, whether facial or vocal, conveys or displays one and only one emotion. Most research in this field follows just this logic, asking subjects to identify emotions expressed in a static image of a face or in a brief recording (a few seconds long) of a person speaking. The examples are typically
180
chosen to be typical expressions of a single emotion which the researcher predicts the subjects will recognize clearly. The subjects are often given a fixed set of choices from a list of basic emotions. Thus the results of the research show that subjects identify single emotions for each display.31 Yet in actual settings, it is common for subjects to experience more than one emotion, and it would follow that they also may express more than one. When subjects were asked to describe another person’s emotional expressions, they generally used more than one term for each emotional episode, e.g., they described someone being angry and jealous, or sad and grieving.32 There is typically far more nuance and complexity to any instance of emotional experience and expression than the division of facial and vocal expressions into seven basic categories would suggest. Second, implicit in this discussion has been the suggestion that all instances of expressions of emotion are equally recognizable. That is, either you are expressing anger in your voice or face, or you are not. Similarly, it has been implicit that all of the various different emotion expressions are equally recognizable. I made an exception for vocal expressions of disgust, but in general the various affect states were not shown to vary in strength. Yet some emotions are more easily and universally recognized than others.33 In practice, there is a threshold of recognition for emotional expressions, and varying
181
degrees of intensity. It is often difficult to discern another person’s affect states, whether because of their efforts to mask them or because of other factors. We often speculate about what someone else is feeling—we even speculate about our own emotions. It is not always easy to put your finger on what a given pattern of behavior really means. Some of the most interesting filmic representations of emotion, such as European art films of the 1960s, show a character’s face and voice to be, to some extent, inscrutable. Sometimes it’s obvious that a person is afraid; but often it’s hard to say what a face means. Bergman’s Persona and Resnais’s Last Year at Marienbad are good examples of films in which faces function as mysteries. In one study of particular relevance to the question of how emotion expressions function in film, psychologists James Carroll and James Russell analyzed facial expressions of emotion in four Hollywood films, Dead Poets Society, Terms of Endearment, Kramer vs. Kramer, and Ordinary People, all of which had been praised by Leonard Maltin for their realism, acting, or general quality. They identified more than 100 episodes in which viewers (psychology students) agreed that a character was experiencing one of the basic emotions, and agreed on which emotion. They found that, with the exception of happiness, the basic emotions were rarely coupled with the facial
182
expressions Ekman’s influential theory would predict. Most of the time, no prototype emotion was displayed, yet some action units, such as a brow raise, were consistently seen for some emotions (surprise, in this case). This point underlies Carroll and Russell’s theory that facial expressions generally function at the level of the AUs, and that the seven basic expressions are rarely—if ever— seen in everyday life. They suggest that the seven expressions of basic emotions are prototypes combining many of the different AUs that one would display for a given emotion. They are mental prototypes used in miming emotions or in acting guides, but are really idealizations of expressions and are ecologically rare or nonexistent.34 This experiment shows that in cinema as well as in real life, emotions may be expressed in subtle or partial ways that do not involve the classic facial patterns we think of as an “angry face” or a “surprised face.” Yet with happiness, the Duchenne smile was commonly observed. This is clearly the most recognizable, unambiguous, and universal of all facial expressions. It is produced in instances of genuine happiness or amusement but not the other contexts in which one might smile, as in polite smiles, miserable smiles, smiles of fear or contempt, and supercilious “Chaplin” smiles.35 Third, this discussion has assumed that emotions may be detected from brief, discreet expressions of an isolated face or voice. But in ecological settings a person’s
183
emotional experience is expressed in multiple modes. One does not ordinarily use an angry voice absent an angry face, angry body language, angry words, and a context producing anger. While facial expressions are often seen to offer the best evidence of emotional experience, in practice people are receptive to all of the modes of interpersonal communication. When asked to describe another person’s emotional experience, those same subjects discussed context in narrative terms (they told a story about the person’s situation), and referred to various vocal and bodily expressions.36 This demonstrates that emotion expressions, and the understanding we have of them, are a product of many cues; some of them may even be contradictory. Contradictory cues may coexist just in the vocal channel, as one’s words and one’s voice might express different feelings. While there are emotional qualities that people express purely nonverbally, in practice vocal expressions occur in combination with language. It may often be the case that subjects must discern someone’s emotions when their words do not make them explicit. Indeed, when asked to describe emotional episodes they observed, subjects would rarely refer to explicit descriptions such as ,“I’m so angry!”37 Yet it would be foolish to doubt that the verbal component of vocal expression is negligible. Words are one part of the multi channel array that is emotional expression.
184
This leads to a fourth point, which is that people’s understandings of each other’s emotional expressions may be influenced to a large extent by context.38 The methods of studying expression recognition typically use still pictures of faces abstracted from any situation. Yet contexts can make a big difference in determining what a face means. For example, athletes often express the joy of victory with faces that Ekman would code as angry (accompanied by shouting and aggressive gestures such as fistpumping) or sad (crying). No one who is watching the game—or even highlights of it that make clear that these are winners’ faces—would mistake the players’ affect states. Expressions of emotion always occur in context and the judgments people make of other people’s emotions are based on a combination of information about the person and the situation. Indeed one of the defining features of emotion is appraisal: one generally becomes afraid only in the presence of something to be feared. Ecologically, it would be highly unusual to encounter emotionally expressive faces absent any context. Even if we grant that clear facial expressions of basic emotions exist, which the above example might cast some doubt on, they have meaning only within the realm of the social. This means that vocal and facial expressions are not the only way that spectators have of understanding a character’s emotions. Inferences based on narrative context (i.e., folk psychology) and causal attributions are also possible means of establishing emotion,
185
as are the direct techniques of narration such as voiceover. The claim of this section is, rather, that faces and voices are primarily emotionexpressers, and that narrative feature films make extensive use of them for this purpose. Yet because of the interaction of modes of characterization, the emotional content of a character cannot be split off from the other information we have about him. A person’s environment (i.e., a narrative situation) is fundamental to this and every aspect of characterization. Fifth, this discussion has assumed that the most significant, if not the only, function of faces is as emotion expressers. But although the majority of research into facial expressions is concerned with emotion expressions, there are obviously other functions of facial expressions. Faces and voices may be analyzed on a scale of arousal, from minimally aroused to neutral to highly aroused.39 Arousal can be associated with emotion expressions (surprised faces show high arousal, sad faces low arousal) or they may not. Someone may just look or sound sleepy or hyper. Given specific contexts, faces and voices can be indices of many nonemotion states, such as alertness, hunger, physical or mental health, or boredom. A facial expression may have nonemotional referential meaning about intentional rather than affect states. Rolling the eyes can communicate doubt or derision; furrowing the brow can communicate skepticism; licking the lips can communicate pleasure; opening the eyes wide can communicate eagerness to learn more;
186
and smiling and nodding can communicate agreement. Facial expressions can also have clearly coded meanings within a given culture or subculture. Exaggerated winking can be a silly way of saying, “I mean this ironically.” Ordinary winking can convey a variety of meanings. Finally, although there is strong evidence supporting modest universality claims, there are still significant differences of degree of recognition among subjects of different backgrounds. Some of the difficulty in identifying an emotional expression as such may be a function of cultural difference.40 Recognition follows a pattern of gradation, and not all subjects’ responses fall on the same point of the scale. So while cinema’s universal, crosscultural appeal might be explained in part by the scientific research I have described, the preference of Americans for American cinema might also be explained by it, since each culture is better at recognizing its own expressions of emotions than it is at recognizing those of other cultures. To summarize these clarifications, the existence of universal emotions and universal expressions of them does not mean that there is no room for ambiguity and nuance in our understanding of how expressions function ecologically, and by extension, in representations of human experience. Emotions may be expressed in combination with each other and may be easy or difficult to recognize. Emotion expressions are a multi
187
channel phenomenon, not simply vocal or facial displays, and they are understood and interpreted in a social context. They are combined with other facial and vocal communication which may not contain emotion content, and which may be culturally variable. And emotion expressions themselves are to some extent culturally variable and thus not absolutely universal. Not all subjects are able with 100% accuracy to understand all other subjects’ emotion expressions. There are degrees of fallibility. All of these qualifications offer direction for the application of theories of emotional expression to the analysis of cinematic narrative.
Emotion Expressions in Film The main obstacle to applying these ideas about expressions of emotion to film is that a film is a representation of human actions, not the actions themselves. An actor portrays a character who is sad or afraid because of events in her world. One might think that unless she is a hardcore method actor, however, she is simulating the emotion. So how do actors convey emotions so impressively that they hold audiences captivated? This is not the topic of this project, which is concerned primarily with the spectator’s comprehension.41 However, in the case of emotions it is impossible to divorce
188
comprehension from other levels of involvement, since one way that people ordinarily respond to the emotional expressions of others is by feeling what they feel. This phenomenon is known as emotional contagion. When you are around a cheerful person, you become cheerful, but a depressed person makes you feel down it the dumps.42 This is true of films as well: an angry character, such as Peter Finch’s madashell newscaster in Network, causes the audience to feel his anger. Obviously, not all film characters’ emotions are duplicated in spectators; unsympathetic characters are less likely to generate this effect. Granting that, how do actors express emotions so convincingly that they are able to infect the audience with them? Emotional contagion depends on facial, vocal, and postural mimicry. Darwin recognized that many species mimic facial expressions seemingly automatically, and there is substantial evidence showing that infants mimic their parents’ facial and vocal expressions and vice versa. Facial expressions are caused, among other things, by autonomic nervous system (ANS) activity, and it would seem that the causation runs both ways. When you smile in mimicry of someone else’s genuine happy smile, your ANS activity is that of a smiling, happy person. In experiments, researchers had subjects form facial expressions without telling them to display a particular emotion, by giving them instructions about which muscles to move, and measured their ANS activity. They found
189
that even without any context in which to feel particular emotions, the subjects showed the ANS patterns of someone experiencing them. Film spectators mimic facial expressions of film characters and “catch” their feelings in the process. A new conception of acting, and in particular of method acting, is suggested by the idea of emotional contagion. One standard simplified description of method acting is that an actor uses a memory of an emotional experience to produce a performance of a character going through the same emotion.43 This is supposed to produce a more authentic rendering than mimicking a character’s outward appearance or pretending to be that person. Emotion memory is an actor’s way of getting into the part and has become a standard notion in American film acting, to the point that even if the technique is not being used the idea of the actor taking on the character’s interiority is commonplace. Yet mimicry is an effective means of producing an emotional experience, as emotional contagion demonstrates. Facial and vocal expressions are both causes and effects of emotional experience. They are all part of the same process, not the endpoint results from a stimulus. It may sound odd, but by pretending to be happy, it is possible to become happy. Still, aren’t there differences between acted expressions of emotion and genuine ones? Undoubtedly there are, but it would take a specialist in the coding of facial and
190
vocal expressions to detail them insofar as they constitute variations in how facial action units or vocal acoustic parameters are utilized. On a more general level, though, we might assume that representations of expressions function roughly in the same way as other aspects of representation. This is essentially a matter of realism. I contend that expressions of emotion derive from and refer to phenomenal reality, especially from social intelligence, but in what way do representations of emotion expressions seem real and in what ways do they seem notreal? When they seem notreal, is this an aesthetic effect or a failure of realism? In general, representations are realist or stylized in varying degrees and fashions. That is to say that they necessarily bear some relation to reality, but that this relation has many permutations and combinations. It would be a gross simplification to say that any representation is more or less realistic than any other. (Is Rules of the Game more realistic than On the Waterfront? According to what terms? In what ways? By whose conception of what is real or stylized? Compared to which other films?) Specific representations function by appealing to realism in certain ways and by applying the stylization or abstraction of reality in certain ways. In cinema, emotion expressions may often be exaggerated or understated, and they may be intentionally incongruous or
191
frustrating. There also may be variations in quality of acting technique that make some performances more or less realistic or stylized, obvious or subtle, clear or obscure. These variations are all historically and culturally specific, as realism at one point in film history will later seem laughable, and realism in one national cinema may look contrived against examples from another.44 But the basic materials of emotion expression are the same facial and vocal features that express emotions in reality. This is because we know of no basic emotion terms or descriptions of facial or vocal activity that are specific to artistic representation. There is no genre of film in which one observes emotion x and corresponding facial expression x, which is specific to the genre or more generally to cinema and is never observed in the real world. Like the other aspects of characterization I have discussed, it is assumed that the same processes used in understanding people in everyday life are those used in understanding cinematic narrative. This is the simplest explanation I can think of; the burden of proof is on those who support a notion of cinemaspecific processes to show first that they exist and then that they explain our engagement with cinema better than the processes I have described.
192 Facial Expressions in the Construction of Character: Welcome to the Dollhouse and Hard Eight Spectators attribute emotions to characters in various ways. Characterization creates an emotion profile on the basis of multichannel cues. Each character is the product of this ongoing process combining information about narrative situations with visual and aural input. Dawn in Welcome to the Dollhouse is a good example because her emotional experience is defined by simple situations and clear, basic emotions. Welcome to the Dollhouse is a film in which understanding the character’s emotions are of central concern to the spectator, as we are invited to empathize with her plight as an awkward social misfit. Dawn’s mental states are generally unambiguous and sympathetic. At various points in the narrative, she experiences humiliation (shame), anger, fear, and even happiness. Emotions are generally brief, and each one Dawn experiences is a response to a welldefined situation. In the first scene, Dawn is humiliated by being unable to find a place to sit in the cafeteria. She is humiliated quite frequently in subsequent scenes, as when she is forced to watch her siblings eat her cake, when Steve implies that she is “retarded,” and when her family laughs at the video of her being pushed into the pool.
193
At the assembly, Dawn is angered when she is showered with spitballs by the bullies. She is also angered quite frequently, as when she calls her friend Ralpie a “faggot,” when she calls her sister “Lesbo,” and when her brother, Mark, orders her out of his room. Her anger also arises whenever she feels she has been treated unfairly, e.g., when she is given a detention and when she is forced to dismantle her clubhouse. The best example of Dawn’s fear comes when Brandon threatens to rape her. She is also made to fear Lolita, Brandon’s friend, who forces Dawn to go to the bathroom with the stall door open and speaks to her in menacing tones. Dawn’s happiness is generally connected to Steve, her brother’s bandmate whom she has a crush on. She is happy when she watches him sing, when she dreams about him, and when she entertains him with Hawaiian Punch and leftover fish sticks and plays piano (badly) for him. We can also identify blends of emotion, i.e., episodes in which more than one emotion can be identified. When she is brought before the principal for returning spitball fire and injuring a teacher, Dawn sits next to her parents and is questioned about her social life. Over the principal’s shoulder, we see the bullies through the window mocking and taunting Dawn. She can see them, but the others cannot.
194
Like so many other scenes in the film, the overriding appraisal that the spectator makes is that Dawn is a victim of unfairness. But her emotions are harder to pin down: we might believe that she is shamed by being told she has no friends, angered by being the one who got in trouble when the instigators got off, afraid of being punished, and remorseful about having caused her teacher harm. Note first that all of these emotion episodes are contextdependent. These emotions are clear from these descriptions of them, whatever their representations. They are represented in a multichannel fashion, as the theories discussed above would predict. Her voice rises and quickens when she pleads her case to her teacher and her parents. Her facial expressions soften with smiles and wideopened eyes when she is around Steve. In general, though, Dawn’s emotions are hardly evident at all her in facial expressions. Heather Matarozzo rarely uses any classic basic emotion faces, as described by Ekman. Her expressions are much more often just blank. This is not Solondz’s general style; indeed most of the other characters are facially very expressive. Dawn’s mother displays many clear, even exaggerated expressions of happiness, disappointment, anger, and fear. Brandon displays menacing, aggressive expressions to Dawn and plaintive, supplicating ones to Cookie, the popular girl who snubs him. Missy is
195
frequently seen with smiles of happiness and Dawn’s teacher typically sneers with contempt. Yet Dawn’s face is typically inexpressive. By deemphasizing Dawn’s face as an expressive technique, Solondz underscores Dawn’s difference from the other characters and her alienation from her family and schoolmates. This technique calls on us to activate assumptions about display rules: her stoicism in the opening scenes as well as later ones is likely a product of the suppression of affect states such as sadness, anger, fear, and frustration. This in turn leads back into assumptions and inferences about intentionality and dispositions. We might figure that an awkward misfit might suppress the display of negative emotions as a means of easing her social experiences, or as an attempt to overcome the negative emotions. This might suggest a resilient or defiant personality, which squares with Dawn’s stubborn actions, or a denial of serious problems, which squares with her lack of selfawareness. The facial display is itself a multidimensional cue, and by virtue of its inexpressiveness Dawn’s face becomes a site of ambiguity, upping the interest in Dawn’s character and, by contrast, making the others seem flatter. To add to that, the film’s style underemphasizes her facial expressions by underlining her emotional experience using other means. Matarozzo’s voice modulates frequently, from quiet and endearing to loud and whiny. She mimics the bullies’ firm
196
tones of voice when speaking to Missy or Marc, but she speaks in a hush when threatened by Lolita and Brandon. Her humiliation when reciting her essay on “dignity” is conveyed by her unsure posture and her weak, hesitant voice. This is duplicated in the assembly scene when she addresses the school and is, again, humiliated. Other stylistic means deemphasize Dawn’s face. Solondz cuts to long or extreme long shots at moments of heightened feeling to detract from Dawn’s face as an expressive device. For example, when Dawn runs away from Brandon, in the scene in which he has said he is going to rape her, she races to a chain link fence, where we are shown her in an extreme long shot. Similarly, her humiliation at being pushed into the pool is a long shot seen on a (TV) screen within the screen. Solondz also uses a musical cue, with distorted, rhythmic electric guitar, uptempo bass, and tom drums in scene transitions. This cue becomes a shorthand for Dawn’s frustration, anger and humiliation, a refrain to sum up each episode of her troubled adolescence.
Murray Smith has discussed how Takeshi Kitano, Robert Bresson, and Wong Kar
wai, all encourage a style of acting that is intentionally facially inexpressive or enigmatic, each fitting this approach to the face into a larger aesthetic system, achieving specific effects. He contrasts this with the clear basic expressions in films by Hitchcock.45 I would suggest that facially inexpressive acting is one baseline technique of modern
197
independent and art cinema, which may function to involve the spectator in constructing the character’s interiority, to increase narrative ambiguity, and to make the characters seem complicated and interesting. Dawn’s emotions are fairly clear from her situation; but there is something intriguing about the stoicism with which she faces her circumstances. This sets her apart from characters in more conventional dramas about adolescence. It makes her experience seem highly specific and individuated, but it also makes adolescence seem especially hellish. What makes Welcome to the Dollhouse distinct as an independent film about adolescence is its unflinching portrayal of unfairness as a basic condition. By restraining Dawn’s facial expressions, Solondz suggests an acceptance of this unfairness, of her routine humiliation, and an almost nihilistic expectation that it will continue unchecked for as long as Dawn is a child, and in many other children after her. He suggests that this is a natural state for adolescents and that their best chance at surviving it is by accepting it as a fact of life. In the film’s final scene, Dawn sings the Hummingbird theme song on the bus to Disney World, with the saddest happy face one can imagine, which underscores both the pathos of her character and the extent to which situations determine our attribution of character emotions. Extracted from the film, this scene would have little emotional impact, but in context it is powerfully pathetic. Ultimately, the effect of the
198
characterization of Dawn in Welcome to the Dollhouse is one of futility, of condemnation to suffering which bespeaks the incompatibility of adolescence and individuality. It is typical of both the film’s director and of independent cinema more generally to characterize adolescence differently from the Hollywood mainstream. For much of Hard Eight, the character emotions are much less evident, largely because of the enigmatic nature of the exposition. There are virtually no unambiguous basic emotion episodes at all in the first third of the film, which include the scenes of Sydney and John’s meeting and the development of their friendship, the episodes introducing Jimmy and Clementine (Gwyneth Paltrow), and the development of the relationship of Clementine’s relationship with Sydney and John. In terms of affect generated by the film in these scenes we do best to speak of mood rather than emotion. Moods are more generalized and longer lasting than emotions.46 Those moments that seem like they might build into genuine emotion moments tend not to. For example, John seems to get angry at Syd for suggesting that they return to Las Vegas together, because he is wary that Syd might have a sexual interest in him. But Syd assures him that this is not the case and John’s feelings are defused. Similarly, in the scene in which John introduces Syd to Jimmy, it seems that Syd is angry with Jimmy for making sexually
199
explicit comments about the cocktail waitresses, but their discussion of it never builds into a genuinely angry dispute. In the scene in which Syd observes the cocktail waitress Clementine emerging from a customer’s room, clearly indicating that she was having sex with him for money, it is suggested that Clementine might feel shame. Yet her words might contradict this inference, as she insists that she doesn’t do anything she doesn’t want to. It could be that she feels mixed emotions: perhaps she is proud to earn money but ashamed do it by debasing herself. Moreover, in all of these moments, no facial or vocal expressions clearly display the classic basic emotion expressions. What the first scenes of the film do have in significant quantity are smiling faces. Especially in the encounters in the cocktail lounge, the characters smile at one another frequently but never, it seems, from genuine happiness or amusement. These social smiles actually signify apprehension as much as positive emotion. Clementine and Sydney smile at each other even though Sydney tells her of his disapproval of her behavior. Jimmy and Sydney smile at each other even though they clearly have some mutual dislike. In these cases, facial expressions serve to obscure rather than illuminate the characters’ interior states, and spectators’ inferences about other aspects of characterization are necessary to establish the character’ emotions or lack of them. In Hard Eight, the tone shifts dramatically in the scene beginning with Sydney’s
200
appearance outside a motel room. Inside, he discovers that John and Clementine have kidnapped a man who hired Clementine as a prostitute and then refused to pay. The man is lying unconscious, with his face in a bloody pillow and his hands cuffed to the headboard. In this scene, the characters express their emotions boldly using every available channel. They bare their teeth in anger, cry in sadness, raise their voices in fear. They shout at each other and pace back and forth. Clementine covers her face in shame and cowers on the floor. As the situation is made clear, the narrative context supports these responses. Clementine and John have committed several felonies and are at risk of being discovered by the police. By asking Sydney to come, they have involved him as an accessory. They also say that they have contacted Jimmy, so that he too is involved and could potentially affect their effort to escape without getting into trouble with the law. We are suspicious of Jimmy not only because of a snap judgment based on his character type (slick casino security consultant, which is made to sound somewhat euphemistic) but because Sydney tells John that he doesn’t like Jimmy. We are inclined to trust Sydney, who takes on a protective, fatherly concern for John. The situation prompts our interpretation that the characters are angry, afraid, and anguished. As is so often the case in cinematic narrative, a variety of redundant cues support the same meanings.47 In the final third of the film, Clementine and John have fled to Niagara Falls, and
201
the story turns to a drawnout confrontation between Jimmy and Sydney. Jimmy tries to shake Sydney down, blackmailing him by threatening to reveal Syd’s secret to John. This is when the narrative’s revelation is made: Sydney acts as he does toward John because Sydney killed John’s father. Sydney seeks, in effect, to take the place of John’s father without letting John know who he really is. What is remarkable about this last part of the film is the extent to which Sydney maintains his stoical, generally expressionless demeanor. With the exception of a single short scene, in which Jimmy threatens him with a gun, Sydney’s face is a blank screen. This is in contrast to images of Clementine laughing in the wedding video that Sydney views, of Jimmy shouting and showing anger and excitement, and John’s soft sobs and sad expression when Sydney tells him, over the phone, that he loves John as though he were his own son. Sydney typically stares straight ahead. At times, Anderson deemphasizes the face, as Solondz does in Welcome to the Dollhouse, by cutting to long shots, by framing unconventionally as in a scene in a car with the camera set up behind Jimmy and Sydney’s heads, and by cutting away to seemingly irrelevant objects such as Sydney’s waist and a coffee cup on a table. But these techniques are brief and function more as punctuation or pause than as a dominant aesthetic.
202
The effectiveness of the narrative depends on the fascination Anderson finds in Philip Baker Hall’s face. As Sydney, Hall makes the most of his face’s craggy lines, deepset eyes with multiple bags under them, and distinguished, angular features. The audience studies Hall’s face to try to get insight into Sydney’s emotion. This is not unlike Solondz’s use of Matarozzo’s face in Welcome to the Dollhouse, and in each case the face seems to suggest an underlying default stance or attitude. Dawn’s face is an index of her suffering, even when she is not shown to suffer, because her character is defined most of all by this quality. Sydney’s face, on the other hand, is constantly purposeful. Sydney is always thinking, it seems, but never lays out his thought processes in words. He feels no less than the others, but his love for John and his hatred for Jimmy are almost always kept to himself, as are his anger and his fear. This is a classic toughguy image from gangster pictures and films noirs. The stoical thinking man never shows his cards until it is most advantageous to him. Like Welcome to the Dollhouse, the protagonist’s face functions in contrast to all of the other faces, which are more conventionally expressive. But in Hard Eight, as in Welcome to the Dollhouse, there are key moments in the narrative in which the character’s shell is cracked and Sydney’s passion is displayed. When Jimmy threatens him with a gun, Sydney looks and sounds afraid, just as in the hotel room he looks and
203
sounds angry. These bursts humanize him and establish that he does feel as strongly as the others. They underline the significance of these scenes in the narrative, as they are rare expressions from Sydney. Hard Eight is a film of shifting emotional tones, and of shifting narrative approaches. It begins by raising many more questions than it answers, in a rather mysterious fashion. It is not at all clear why the characters are behaving as they do. This is the case of largescale narrative developments, such as Sydney’s taking John on as his protégée. It is also the case of smallerscale details, such as Sydney’s habit of passing many hours in a casino hotel cocktail lounge playing keno. It is typical of independent film narration to create enigmatic characters based on genre or social types; the work of spectatorship becomes the effort to decipher the clues leading to an interpretation of the character’s identity. Thus they are able to combine their most basic appeals: on the level of the fascinating individual who stands for a particular social experience (in Syd’s case, an old man, a father, a Vegas hustler, a mentor); on the level of formal play (uncovering the mysteries of the character as a game we play with the narrative); and on opposing Hollywood conventions of more straightforward emotional characterization. Midway through the film, the emotion tone shifts from being cool and contemplative to being more straightforwardly melodramatic, arousing pathos and fear on
204
behalf of the central characters who are caught in a bad situation. We still do not know enough about their backstories at this point to feel the full emotional weight of the relationships they have with each other. In the end, what is most fascinating about Hard Eight is Sydney, who is capable of acts of selfless love for John at the same time that he is capable of vicious cruelty toward Jimmy, who is foolish enough to think he can outtough a tough guy. The contradiction in Sydney’s emotional makeup is developed in a perfectly modulated performance, breaking through the internal norm of restrained stoicism with flashes of full emotion, and relying on modulations in narrative situation to further sketch out the character’s psychological states. The clarity of the other characters’ emotional expressions makes Sydney’s inexpressiveness all the more interesting.
Conclusion: Combining Appeals Spectators understand character psychology in various ways. They categorize characters, which focuses attention on some traits and not others, and guides attention and interest. They are given explicit information through dialogue and subjective techniques. They infer intentionality from narrative scenarios and observed actions using folk psychology. They attribute the causality of narrative events to character dispositions and thereby
205
create a sense of the character’s personality. And they read emotions in characters’ expressions or attribute emotions in spite of those expressions on the basis of other cues and inferences. All of these processes are mutually reinforcing. Information about any one aspect of characterization impacts upon the others. For example, a sense of the character’s personality may bear upon any understanding of their emotional states. A deceitful character may be more likely to mask his or her true emotion. Knowledge about the character’s intentions may also affect emotion assessments. A character who is trying to pull off a swindle will be expected to display different emotions from a character who is the victim of the crime. All of the techniques of characterization can tilt our understanding of character, which affects spectators’ use of all of the techniques. The importance of narrative context cannot be overstated, yet each technique is part of the process that creates a narrative context. Connecting all of these dimensions of characterpsychologyconstruction is an overarching goal of narrative comprehension: the data of the narrative must be made to cohere as best they can. We prefer that all the parts of the social puzzle fit together, so that my inferences about your behavior, your intentional states, your emotions, and your traits and types are not mutually contradictory, and that explanations of some of these are helpful in explaining other parts. As we have seen in considering stereotypes, we tend to
206
see what we are looking for, so narrative contexts direct our expectations and make coherence more attainable. This is why exposition has come up so often in these discussions: the design features of narrative demand that beginnings clarify a context—at least minimally—so that social cognition processes can function efficiently and coherence can be achieved. Maximum coherence is the spectator’s goal, but not always the filmmaker’s. Most narratives challenge our abilities to make character psychology inferences and judgments cohere in a modest fashion by making traits and actions inconsistent with each other, by frustrating typing, by creating ambiguous or contradictory situations and emotion expressions, and by withholding information for purposes of suspense or surprise. Characterfocused narrative traditions such as American independent cinema challenge our coherence goal more than modestly, making character comprehension a more involved and extensive process, but also potentially a more satisfying one. Of course, narratives can go too far into incoherence or can fail to cohere because they are badly made, and there is little satisfaction in that. But the best independent films pose interesting and welldeveloped challenges of coherence that seek to maximize our interest in engaging with the process of figuring out character psychology. .
1
Fritz Lang,, “The Future of the Feature Film in Germany” in Anton Kaes, Martin Jay and Edward Dimendberg (eds.), The Weimar Republic Sourcebook (Berkeley: U of California P, 1994), 623. 2
Carroll, Philosophy of Horror and A Philosophy of Mass Art; Tan, Greg M. Smith.
3
Alan J. Fridlund, Human Facial Expression: An Evolutionary View (San Diego: Academic P, 1994), 2.
4
Charles Darwin, The Expression of the Emotions in Man and Animals (New York: Oxford, 1998). On physiognomy, see Leslie A. Zebrowitz, Reading Faces: Window to the Soul? (Boulder: Westview P, 1997). 5
Ibid; see also Fridlund; Ross Buck, The Communication of Emotion (New York: Guilford P, 1984); Richard S. Lazarus, Emotion and Adaptation (New York: Oxford UP, 1991); Robert Plutchick, Emotion: A Psychoevolutionary Synthesis (New York: Harper & Row, 1980); Jonathan H. Turner, On the Origins of Human Emotions: A Sociological Inquiry into the Evolution of Human Affect (Stanford: Stanford UP, 2000). Paul Ekman (ed.), Darwin and Facial Expression: A Century of Research in Review (New York: Academic P, 1973); and Paul Ekman, Joseph J. Campos, et al. (eds.), Emotions Inside Out: 130 Years After Darwin’s The Expression of the Emotions in Man and Animals (New York: The New York Academy of Sciences, 2003). 6
Paul Ekman, Emotion in the Human Face vol. 2 (Cambridge: Cambridge UP, 1982). 43.
Ibid; see also Paul Ekman and Erika Rosenberg (eds.), What the Face Reveals: Basic and Applied Studies of Spontaneous Expression Using the Facial Action Coding System (FACS) (New York: Oxford UP, 1997). 7
8
Malcolm Gladwell, “The Naked Face” The New Yorker (5 August 2002), available online at http://www.gladwell.com/2002/2002_08_05_a_face.htm. 9
Paul Ekman, “Expression and the Nature of Emotion” in Klaus R. Scherer and Paul Ekman (eds.), Approaches to Emotion (Hillsdale, NJ: Erlbaum, 1984), 319343. On facial expressions and lying, see Paul Ekman, Telling Lies: Clues to Deceit in the Marketplace, Politics, and Marriage (New York: Norton, 2001). 10
11
This experiment is described in Ekman, “Expression and the Nature of Emotion.”
12
13
Daniel MacNeill, The Face (Boston: Little, Brown, 1998), 242243.
Antonio Damasio describes the neurophysiology of this distinction between two kinds of smiles and
applies it to distinguishing between method acting and its alternatives: Damasio, Descartes’ Error: Emotion, Reason, and the Human Brain (New York: Quill, 1994); for a detailed discussion of smiles, see Mark G. Frank, Paul Ekman, and Wallace V. Friesen, “Behavioral Markers and Recognizability of the Smile of Enjoyment,” in Ekman and Rosenberg, 217242; and Millicent H. Abel (ed.), An Empirical Reflection on the Smile (Lewiston: Edwin Mellen P, 2002). 14
Hillary Anger Elfenbein and Nalini Ambady, “On the Universality and Cultural Specificity of Emotion Recognition: A MetaAnalysis” Psychological Bulletin 128 (2002), 203235. 15
One influential—and typical—example of this position is Rudolph Arnheim, Film as Art (Berkeley: U of California P, 1957). 16
Klaus R. Scherer, “Vocal Affect Expression: A Review and a Model for Future Research” Psychological Bulletin 99 (1986), 143165. 17
Tom Johnstone and Klaus R. Scherer, “Vocal Communication of Emotion” in Michael Lewis and Jeannette M. HavilandJones, Handbook of Emotions 2nd ed. (New York: Guilford P, 2000), 220235. 18
Ekman, Telling Lies, 9298.
19
Johnstone and Scherer.
20
Ibid.
21
Patrik N. Juslin and Petri Laukka, “Communication of Emotions in Vocal Expression and Music Performance: Different Channels, Same Code?” Psychological Bulletin 129 (2003), 770814. 22
Johnstone and Scherer.
23
Klaus R. Scherer, “Affect Bursts” in Stephanie H.M. Van Goozen, Nanne E. Van de Poll, and Joseph A. Sergeant (eds.), Emotions: Essays in Emotion Theory (Hillsdale, NJ: Erlbaum, 1994), 161193. David Bordwell, “Intensified Continuity: Visual Style in Contemporary American Film” Film Quarterly 55.3 (2002), 1628. 24
25
The existence of basic emotions and universally recognized expressions of them is a matter of debate among experts in the field. See James A. Russell, “Is There Universal Recognition of Emotion from Facial Expressions? A Review of CrossCultural Studies” Psychological Bulletin 115 (1994),10241; Paul Ekman, “Strong Evidence for Universals in Facial Expressions: A Reply to Russell's Mistaken Critique” Psychological Bulletin 115 (1994), 26887; Carroll E. Izard, “Innate and Universal Facial
Expressions: Evidence From Developmental and CrossCultural Research” Psychological Bulletin 115 (1994), 28899; and James A. Russell, “Facial Expressions of Emotion: What Lies Beyond Minimal Universality?” Psychological Bulletin 118 (1995), 37999. However, no one denies that some emotions are recognized crossculturally at a rate of better than chance; the questions motivating these debates refer to the extent of recognition. 26
Buck.
27
Juslin and Laukka.
28
One application of facial expression research to film narrative theory is Murray Smith, “Darwin and the Directors: Film, Emotion and the Face in the Age of Evolution,” Times Literary Supplement 7 February 2003, 1314. 29
Smith, Engaging Characters; Carl Plantinga, “The Scene of Empathy and the Human Face on Film,” in Carl Plantinga and Greg M. Smith (eds.), Passionate Views: Film, Cognition, and Emotion (Baltimore: Johns Hopkins UP, 1999), 239255. 30
Unusual but not unheard of: in the “Hush” episode of Buffy the Vamire Slayer originally aired 14 January 1999, a demon’s spell leaves everyone in Sunnydale literally speechless. 31
James A. Russell and José Miguel FernándezDols, “What Does a Facial Expression Mean?” in James A. Russell and José Miguel FernándezDols (eds.), The Psychology of Facial Expression (Cambridge: Cambridge UP, 1997), 330; James A. Russell, JoAnne Bachorowski, and José Miguel FernándezDols, “Facial and Vocal Expressions of Emotion” Annual Review of Psychology 54 (February 2003), 329349. 32
Sally Planalp, “Communicating Emotion in Everyday Life: Cues, Channels, and Processes” in Peter A. Andersen and Laura K. Guerrero (eds.), Handbook of Communication and Emotion: Research, Theory, Applications, and Contexts (San Diego: Academic P, 1998), 2948. 33
Elfenbein and Ambady.
34
James A. Carroll and James A. Russell, “Facial Expressions in Hollywood’s Portrayal of Emotion,” Journal of Personality and Social Psychology 72.1 (1997), 164176. 35
Ekman, Telling Lies, 150160.
36
Planalp; Ursula Hess, Arvid Kappas, and Klaus R. Scherer, “Multichannel Communication of Emotion: Synthetic Signal Production” in Klaus R. Scherer (ed.), Facets of Emotion: Recent Research
(Hillsdale, NJ: Erlbaum, 1988), 161182. 37
Planalp.
38
José Miguel FernándezDols and James M. Carroll, “Is the Meaning Perceived in Facial Expressions Independent of its Context?” in Russell and FernándezDols (eds.), 275294. 39
James A. Russell, “Reading Emotions From and Into Faces: Ressurecting a DimensionalContextual Perspective” in Russell and FernándezDols (eds.), 295320. 40
Elfenbein and Ambady.
41
The emotional response of spectators is a topic addressed in many cognitivist film theories, e.g., Carroll; Smith, Engaging Characters; Plantinga and Smith; Greg M. Smith; Tan; and Torben Grodal, Moving Pictures: A New Theory of Film, Genres, Feeling, and Cognition (Oxford: Clarendon, 1997). 42
Elaine Hatfield, John T. Cacioppo, and Richard L. Rapson, Emotional Contagion (Cambridge: Cambridge UP, 1994). 43
The idea of “emotion memory” comes from Constanin Stanislavski, An Actor Prepares, trans. Elizabeth Reynodls Hapgood (New York: Theater Arts, 1936). 44
Kristin Thompson, Breaking the Glass Armor: Neoformalist Film Analysis (Princeton: Princeton UP, 1988). 45
Smith, “Darwin and the Directors.”
46
Greg M Smith.
47
Bordwell, Staiger and Thompson 31.