CHAPTER 7
A Song in Your Heart
Music has charms to soothe a savage breast. WILLIAM CONGREVE, The Mourning Bride
Music is joy. HSÜN TZU (XUNZI), Basic Writings
MUSIC AS EMOTIONAL
A traditional basis for asserting the centrality of music to human nature is its relation to the emotions. Deryck Cooke straightforwardly claims, “music is, properly speaking, a language of the emotions, akin to speech.”1 Beethoven, he notes, wrote on the manuscript of Missa Solemnis, “From the heart—may it go back—to the heart!” 2 Psychologist Carroll C. Pratt also claims that music is a language of the emotions that is “unequalled in this regard by any other art.”3
The idea that music has a special relationship to emotion is not uniquely Western. According to the “Great Preface” in the Chinese Book of Songs (The Shih Jing),4
The feelings move inwardly, and are embodied in words. When words are insufficient for them, recourse is had to sighs and exclamations. When sighs and exclamations are insufficient for them, recourse is had to the prolonged utterances of song. When those prolonged utterances of song are insufficient for them, unconsciously the hands begin to move and the feet to dance.5
When the ancient Confucian philosopher Xunzi claims that “music is joy,” he is not only expressing his own opinion. He is pointing out that the Chinese character for music is identical to that for joy.
Joy is only one of the emotions that cultures commonly link with music. The twelve dastgāh, or modes, of the Persian musical system are associated with various emotional tones.6 The ragas7 used in classical Indian music (both Hindustani and Karnatic8) are similarly associated with particular moods, as well as with times of the day. According to Rowell, “the tradition of rāga has become one of the primary means by which Indian culture has become sensitized and perhaps even instructed in emotive life.”9
The Temiar of Malaysia describe the heartbeatlike rhythm of their bamboo percussion as “moving one’s heart to longing.”10 Steven Feld describes the Kaluli of Papua New Guinea as singing with the purpose of moving listeners to “feel sorrow for the performer,” whose songs typically involve recalling and lamenting someone who has died.11 Music is a primary means of expressing grief throughout the world.12
A common theme across cultures is that music renders participants emotionally open (regardless of whether they are performing or listening). Medieval philosopher and mystic Al-Ghazali (1058–1111) claims of the Qu’ran, poetry, and serious music, “There is no way of extracting their hidden things save by the flint and steel of listening to music and singing, and there is no entry into the heart save by the antechamber of the ears.”13 Gilbert Rouget observes that the association of music and emotion is so strong in the Arabic world that tarab, the Arabic word for musical emotion, “has in fact come to signify music.”14
Association with emotion seems at least a near-universal in connection with music. But how are emotions and music related? First, I will consider some of Western philosophy’s answers to this question and suggest that the competing answers need not be seen as mutually exclusive. Instead, we should recognize that listeners can relate to music in a fluid manner that allows for varying degrees of identification or detachment. The various competing theories tend to focus on different ways the listener can engage with music.
Second, I will consider the question of the object of musically aroused emotion. Emotion is often said to require an object, for example, something at which one is angry when one experiences anger. But how is an object provided when emotion is aroused by music? Is one angry, for example, at the music? I will argue that in musically aroused emotion, the object is typically vague. Our attention is certainly not focused on as clearly structured an object as we encounter in everyday emotional experience.
I will go on to suggest that the listener’s fluid relationship to the emotional character of music as well as the typical lack of a definite object of emotion enable music to prime listeners for a relatively open-ended empathetic stance that can encourage feelings of affinity with others with whom one does not usually feel connected. I will then discuss the extent to which musical expressiveness and arousal of emotion transcend cultural boundaries and to what extent they are more restricted.
THE CONNECTION: ALTERNATIVE VIEWS
The relationship of music and emotion has traditionally received at least three different explanations in the West, although many thinkers combine more than one of them. Some theorists claim that music imitates or cognitively represents emotion, because of either natural resemblances or conventional associations between features of music and characteristics of emotions and/or their behavioral manifestations. Others argue that music expresses emotion.15 A third group holds that music arouses emotion in musical participants. I will discuss each of these positions in turn, noting some parallel positions formulated outside the West.
1. Music Imitates (or Cognitively Represents) Emotion
The idea that music imitates emotion may strike us as a highly abstract thesis, but this was taken for granted at an earlier point in Western history, when music was not sharply differentiated from other activities, such as dance, poetic presentation, and theatrical performance (as it remains in much of the world). Thus Plato, whose character Socrates argues that music imitates (as well as arouses) emotion, seems to have believed that the character and rhythms of music resemble the characters and rhythms of human beings in the grip of certain feelings, a view that seems quite plausible when the music goes along with the movements of an actor or dancer.16 Medieval Islamic thinker Ibn Sina (980–1037) proposed an alternative imitation theory, according to which the basis of our delight in music is the resemblance between the appearance and disappearance of motives and themes in music and the comings and goings of beloved individuals.17 In the Renaissance and early Baroque a debate ensued over whether Plato associated imitative qualities with the mode (i.e., the scale) or the specific melodic fragment.18 But even these later theorists did not expect the imitation to be established by the music alone; they assumed that music would be linked with texts and, in the case of opera, action.
The eighteenth century saw both renewed defense of imitation theory and the inception of doubts as to its plausibility.19 The doctrine that the fine arts in general were inherently imitative, a view ascribed to Aristotle, motivated a defense of music’s imitative powers, and emotion was typically taken to be the object of imitation. However, as instrumental music developed into a phenomenon in its own right, this categorization began to strike some theorists as dubious. Adam Smith (1723–1790), for example, reveals skepticism.
Instrumental Music . . . though it may, no doubt, be considered in some respects as an imitative art, is certainly less so than any other which merits that appellation; it can imitate but a few objects, and even these so imperfectly, that without the accompaniment of some other art, its imitation is scarce ever intelligible: imitation is by no means essential to it, and the principal effect which it is capable of producing arises from powers altogether different from those of imitation.20
James Beattie (1735–1803) goes further and contends that music should not be considered among the imitative arts, for its power is its ability to affect listeners, not to represent anything.21
A more recent critic of imitation theory has had perhaps the strongest influence on subsequent Western philosophical views on the music-emotion connection. Eduard Hanslick (1825–1904) sought to debunk the fashionable theories of his day concerning music’s relation to emotion, including the view that music represents emotion. In On the Musically Beautiful (1854), he asserts: “The representation of a specific feeling or emotional state is not at all among the characteristic powers of music.”22 The only ideas that music can represent with its own resources, according to Hanslick, are “those ideas which relate to audible changes in strength, motion, and proportion,” including ideas “of increasing and diminishing, acceleration and deceleration, clever interweavings, simple progressions and the like.”23 Music’s content is not emotion, but “tonally moving forms.”24 We can use emotional terms to describe the character of passages of music—they reflect our felt experience and “indeed” Hanslick adds, “we cannot do without them.” Strictly, however, we should keep in mind that “we are using them only figuratively.”25
Thus shared dynamic character leads people to associate particular emotions with music, according to Hanslick. Several emotions might share the same dynamic, however, and this enables people to associate the same passage of music with a variety of emotional states.26 Hanslick also observes that composers frequently use the same passages of music to express different emotions. (We might think of the music for “Greensleeves,” which has amorous words attributed to that wife-killer Henry VIII, but is also utilized as a Christmas hymn.) Because particular musical passages are not uniquely correlated with definite emotional content, Hanslick concludes that the imitation theory (and the broader view that music represents emotion to the mind) should be rejected.27
Some who accept Hanslick’s position that music and emotions share dynamical characteristics have drawn the opposite conclusion, a route that is available to them because they do not take representation to require, as does Hanslick, a one-to-one correspondence between a musical passage and a particular emotion.28 Susanne K. Langer is among these thinkers.29 She characterizes music as a “presentational” (as opposed to “discursive”) symbol of emotional life, with which it is isomorphic. Music presents the “logical form” of human feeling.30 “Logical form,” in this case, means such patterns as “motion and rest, of tension and release, of agreement and disagreement, preparation, fulfillment, excitation, sudden change, etc.,” which characterize both music and “inner life.”31
2. Music Expresses Emotion
Although often conjoined with the view that music imitates or cognitively represents emotions, the view that music expresses emotions is a distinct account of the music-emotion connection.32 The expression theory has a long track record in the West, and it had particularly enthusiastic adherents among those who elaborated specific correspondences between musical elements and passions in the sixteenth through eighteenth centuries and among the Romantics.33
The expressiveness of music is broadly accepted around the world.34 In the fall of 2004, Haitian victims of Hurricane Jean were shown on a CNN clip singing about their tribulations, including their lack of food.35 In Japanese Noh theater harsh music, punctuated by cries, is used before a ghostly character appears on stage to express both the attachment that is pulling the ghost through the gap separating the living and the dead and the pain that is caused by this crossing. In ancient China, as the previously cited passage from the “Great Preface” of the Chinese Book of Songs (The Shih Jing) attests, music was considered an outgrowth of a natural human need for emotional expression and an enhancement to the expressive power of words.36 Xunzi similarly contends with respect to music, “Man must have his joy, and joy must have its expression.”37
Expression theory also has its opponents. Igor Stravinsky claimed, “music is, by its very nature, essentially powerless to express anything at all.”38 Hanslick argues, “Music consists of tonal sequences, tonal forms; these have no other content than themselves.”39 China, too, had its Hanslick or, perhaps more accurately, its Langer.40 Although he considered musical patterns to be revelatory of “patterns of mind and nature,” the neo-Daoist Ji Kang rejected the correlation schemes devised during the Han dynasty to link particular sounds with particular emotions.41
The idea that music expresses emotion has led some philosophers to ask whose emotion is expressed. The obvious candidates are the composer or the performer, but the activities of composing and performing are not necessarily expressive activities. The technical details of writing or performing music are quite absorbing in themselves, and attention to these may be incompatible with giving complete attention to emotional expression. Moreover, both composers and performers are able to concentrate on producing music, even quite moving music, when their personal emotional states are quite different from those expressed by the music. But if the emotion expressed in music is not that of the composer or the performer, whose is it?
Eighteenth- and nineteenth-century defenders of expression theory, including J. G. Herder, C. P. E. Bach, and Heinrich Heine, held that the composer expressed his own emotions through music, but that these were not his emotions as a particular individual. Instead, they were universal emotions, expressive of the “intelligible self,” the noumenal self, independent of particular circumstances and motives that differentiate individuals from each other.42 Music, expressive of such a self, conveyed universal emotions that anyone could understand, because everyone was such a self.43 One might compare this view to that of eleventh-century Kashmiri Shaivite Abhinavagupta, who held that the true savor (rasa) of the emotion (bhāva) produced by the performing arts arises only when one experiences it in its essence, as a universal mode of experience.44
In the twentieth century, however, artistic concepts of expression in the West have focused on the artist’s personal emotions, although the supposition is usually that human drives and their vicissitudes are sufficiently similar to render such expression communicative to others. By contrast, twentieth-century Western philosophy became suspicious of the notion of emotional communication on any basis besides intersubjective awareness of a particular person’s behavior (including speech). For some time, under the influence of Ludwig Wittgenstein, anglophone philosophy took the postulation of mental states to be unwarranted. In this respect, philosophy seemed more concerned to avoid speculating beyond what is provable than to acknowledge the possibility of communicating anything beyond what is observed in behavior. While many contemporary philosophers reject the proscription on reference to mental states, debate continues to emphasize external behavior and what is intersubjectively verifiable.
A growing number of musicologists and philosophers have recently urged a kind of restoration of a sense of a person behind musical expression. They do not, however, propose that the emotions expressed are the occurrent emotions of the composer or performer. Somewhat resembling adherents of eighteenth-century expressionism, these theorists propose that the communicator of musically expressed emotions is not the particular individual who composes or performs, but a more abstract and more universal character, a persona, particular to music.45
As Jerrold Levinson characterizes the notion, the persona is “the individual indefinitely imaged as the subject of the state being expressed.”46 This seems a reasonable answer if one considers that persona means mask, or the face one presents to the world. The persona is not the full-fledged character, but the personality as it appears. If we take the putative musical persona to be something like this, it would be the hypothetical person who experiences the apparent emotions expressed in music, with no further characterization than that.47 Superficially, “someone’s” emotions are expressed; the persona is, essentially, that “someone.” Levinson has associated the persona with the sense of human intentionality and agency that we recognize in music. I am inclined to agree with him: we do interpret music, for the most part, as human communication, even in instrumental music, in which the source of the communication is minimally identified. The persona is the mask, or placeholder, for this source.
The persona is something of a persona non grata to many philosophers of music, Davies and Kivy among them. They hold that musical expression is in the music; we need not posit some person to whom the expressiveness belongs. Music is expressive by virtue of its resemblance to the kind of behavior or utterance characteristic of a person experiencing an emotion.48 Hence the persona is unnecessary.
Davies and Kivy diverge in their further grounds for criticizing the persona idea. Davies acknowledges that a listener might contemplate a work of music in terms of a persona; he objects to the idea that imagining a persona is necessary for a full understanding of the work and doubts that music constrains the imagined narrative enough for it to be useful for illuminating what specific music is expressing.49 Kivy, by contrast, dislikes the very idea of the persona, primarily on the ground that it is too vague to be very helpful. “The musical persona is such a vague, abstract, shadowy being that even ‘its’ sex cannot be determined,”50 he complains.
No doubt Kivy is right that the musical persona is ungendered, but I think this vagueness is a virtue, enabling listeners and performers with various personal traits to identify with it. The fact that the musical persona is in many respects devoid of detail allows the listener the possibility of an unobstructed vicarious joining in with the movement of the music, including the expressive movement. Once again, I suggest that a type of vagueness in music is a virtue. Because the persona does not have specific traits besides those exemplified in its behavior, many of the situating characteristics that identify individuals in everyday life are lacking, along with the interference these sometimes present for empathetic identification with others. And the modes of movement that music conveys are not restricted to persons of any particular body type, a fact that yields considerable delight in childhood games of imitating animals. A petite preschooler can display a lumbering gait, for example. The postulation of a largely indefinite persona reflects the fact that musically expressed emotion can be, but need not be, recognized by anyone and everyone participating in the musical experience—including composer, performer, and listener.51 Whether or not imaging a persona’s experience will lead to convergence in judgments about what a given piece of music is expressing is another question.
Aaron Ridley rightly suggests that some works of music are more likely than others to suggest an impression of a persona doing the expressing, and that the responsiveness of the individual listener plays an important role in the experience of a work of music.52 Importantly, Ridley acknowledges transformations of the listener’s affective response over the course a particular experience of music. He suggests that initially the feeling state expressed by a musical persona may strike us as “out there” and as “a state of mind that is not ours . . . but the persona’s.”53 We may well sympathize with this state of mind, but this somewhat distanced entertaining of the persona’s feeling often gives way to a more identified empathetic response. “It seems reasonable,” claims Ridley, “to describe the net effect upon the responsive audience as empathetic rather than sympathetic in nature . . . even though the persona to whom the person . . . may be said empathetically to respond has been partly constructed by him, on sympathetic grounds.”54
What strikes me as particularly important about these observations is their acknowledgment that a listener’s affective perspective on the music can vary over the course of a particular work. We may listen from a primarily intellectual state of mind, in which case we may not respond to the expressive content with either sympathy or empathy.55 Instead, we occupy a relatively detached perspective. At such moments the music is no less expressive, but we don’t particularly identify the expression with our own individual state. Even if we are responsive on an intellectual level, as Ridley notes, we may shift from an initially somewhat distanced attitude toward a more fully empathetic one. I would add that our attention and stance can shift over the course of listening to music, particularly while listening to long and complex works.56 Sometimes we are particularly identifying with the emotions expressed. At other times we may occupy a relatively detached perspective on the music and its emotional content. At such moments the music is no less expressive, but we don’t particularly identify the expression with our own individual state.
I think the notion of a persona is useful for suggesting the range of relationships we can take toward musically expressed emotion. At times we identify with the persona (in which case we do not take the detached analytical view in which the term or impression persona would likely come to mind). At other times, we are more emotionally detached from the music, and the term persona is serviceable for indicating a sense that the expressive content pertains to beings like us, categorically considered.
3. Music Arouses Emotion
Cultures appear to be fairly unanimous in the belief that music can arouse emotion. We have already mentioned the Kaluli of Papua New Guinea and the Temiar of Malaysia, and the texts of ancient China, as well as the raga system of Indian music. Merriam reports that members of the Basongye tribe of Africa believe that even outside of its cultural context, music can arouse emotion.57 Tuamotus, a Polynesia people, believe that mana, or the supernatural power of the ancestors, is conveyed through their music. Mana is more than emotion, but emotional nonetheless.58 Anthony Seeger similarly reports of the Suyá people of the Amazon rainforest: “Suyá singing arouses sadness in some and creates euphoria in the rest.”59
Music’s power to influence emotion is one of the key reasons that Chinese philosophers typically believe that music has significant ethical influence. They often emphasize music’s potential as a means of emotional regulation.60 The neo-Confucian thinker Chou Tun-I (Zhou Dunyi) observes, “As the sound of music is calm, the heart of the listener becomes peaceful, and as the words of the music are good, those who sing them will admire them. The result will be that customs are transformed and mores are changed.”61 The ancient “Record of Music” and the Confucian tradition generally emphasize the importance of structuring music in accordance with right principles so that the influence it exerts is good.62
Like Chinese philosophy, Indian thought takes for granted that music arouses emotion. The classical text that elaborates on the connection is the Nātyaśāstra. Insofar as it provides techniques for achieving a basic emotional tone in performance, the Nātyśāstra offers a theory of emotional imitation through the arts, including music; but it is primarily concerned with arousing emotion in the audience. The aim in performance is to arouse one or more bhāvas, or basic emotional states.63 Ideally, however, performances arouse one of eight rasas, each of which is the universalized flavor or savor corresponding to one of the bhāvas.64
Ancient Greek philosophers also held that music arouses emotion. The Pythagoreans may have originated the Greek idea that music could have ethical effects, that is, an impact on listeners’ spiritual state that would influence their behavior. This idea was linked to the notion that each tribe or city-state had its own characteristic mode, and that the modes themselves had an impact on the character of the people. Damon of Athens (fl. 460 BCE), Plato, and Aristotle all endorsed this view, commonly called the ethos theory.65
Many of the great Islamic philosophers of the Middle Ages explicitly endorse some version of arousal theory. Al-Kindi (ca. 800–870) contends that although the emotional impact of particular instruments is culturally relative, some music can instill courage, strength, military fervor, or delight in any listener.66 Al-Farabi (d. ca. 950) stresses both the expressive function of music and its ability to communicate the emotions expressed to listeners.67 He classifies the affects linked to music into three categories: affects that make the soul stronger (e.g., anger and hostility), affects that soften the soul (e.g., fear and mercy), and affects between the other two categories (e.g., calm).68 Ibn Sina suggests that the listener’s soul conforms to the affective quality of the music heard. He also claims that music (and sound in general) can help release intense emotions and tranquilize disruptive ones.69 Al-Ghazali contends that music can arouse various emotions, both energizing and relaxing ones, and it can stir the spiritually prepared person to longing for God, the highest potential of music.70
Like many medieval Islamic thinkers, seventeenth- and eighteenth-century European theorists often emphasized music’s power to arouse feeling and held that the emotions expressed in music were the ones aroused in the listener.71 The idea that listeners mirror emotions expressed in music continues to hold sway, but this view is hard to verify empirically. Experimental subjects may describe their emotional responses to music inaccurately because they read into the music what they think they are supposed to feel on the basis of “demand characteristics,” cues that indicate what the researcher wants to find.72 Listeners may also identify the emotions they think a musical passage expresses as their own feelings, even if they have not really experienced these emotions.73
Music’s arousal of emotion is probably widely taken for granted, but some theorists have challenged the view. Langer denies that music arouses emotion, even though it symbolizes emotional life. Composer Paul Hindemith argues that emotions aroused by music are not real, but apparent. “The emotions which music evokes are not feelings, but they are the images, memories of feelings.”74 Psychologist Carroll C. Pratt, while acknowledging that some listeners do feel aroused by the emotion expressed by music, contends that most listeners (and all who are sophisticated) do not.75 Peter Kivy similarly rejects the view that listeners typically mirror expressed emotion (though he is decidedly in the minority).76 The lack of unanimity among listeners is one of the principle arguments against the arousal view: not everyone seems to experience the emotions that are allegedly aroused.77
One of the most innovative accounts of the music-emotion relationship, that of Leonard B. Meyer, deflects this criticism. Meyer’s theory is based on the idea that as we listen to music in a familiar style, we have certain expectations of the way the music will proceed. “Affect or emotion-felt is aroused,” argues Meyer, “when an expectation—a tendency to respond—activated by the musical stimulus situation, is temporarily inhibited or permanently blocked.”78 According to Meyer, listeners typically have affective responses at points in the music where their expectations are thwarted. Alternatively, a listener may approach the music less emotively and more analytically, attending to the way it has been structured and noting the points at which it takes a surprising turn. These are the points at which affect predictably occurs.
Meyer’s indication of alternative approaches that a listener can take to music, with different impacts on affective response, strikes me as important and surprisingly underappreciated. If the arousal of emotion depends in part on a listener’s attitude in approaching the music, there should be no mystery as to why one listener experiences musically induced emotion while another one does not. Meyer makes the important point that alternative types of engagement can be taken toward the same piece of music. Although Meyer emphasizes different personality types and educational backgrounds as inclining listeners toward different orientations toward the music, his point acknowledges the possibility that listeners can be flexible in their focus, a point that I will make much of in what follows.
One of the ingenious features of Meyer’s theory is its attempt to connect affective response with objective features of a musical work: “granted listeners who have developed reaction patterns appropriate to the work in question, the structure of the affective response to a piece of music can be studied by examining the music itself. . . deviations can be regarded as emotional or affective stimuli.”79 If Meyer’s strategy succeeds, it dispenses with the need to track the subjective reactions of particular listeners. As long as a listener is sufficiently habituated to the style of a piece of music, his or her emotional responses will correspond to its structural features (unless the listener defuses these by rationalizing them).
Meyer’s theory of the mechanism for musical arousal of emotion is only one of many. Even if we restrict our attention to the emotional impact of intramusical developments, we need not limit ourselves to the fulfillment or nonfulfillment of expectations. Early in our exposure to a style of music, for example, before our expectations have become very detailed, emotions of satisfaction can be aroused by virtue of our tracking a trajectory of music and recognizing its gradual development of clear form. The goal-directed character of many cultures’ music (as we noted in chapter 3) is one structural component that can give shape to a piece of music. Many musical forms across the world encourage a gradual sense of the music “taking shape.” For example, the alap in Hindustani music, which involves a rhythmically free exploration of the raga (scale-melody) on which a piece is constructed, typically gives way to a section with a steady pulse (the jor), and later reaches a stage characterized by a definite rhythmic cycle imposed by a drum (jhala).
The purposive growth and renewal that are experienced in the unfolding of a musical form are also emotionally satisfying. Japanese music tends to be organized in a sequence that moves from relative rhythmic freedom (jo) toward greater rhythmic regularity (ha) and ultimately toward a conclusion in which density is intensified (kyū).80 In many forms of Western classical music, particularly in sonata-allegro form, preliminary statements of themes give way to more exploratory development of some of their features, ultimately to give way to a recapitulation, typically in the tonic key. The move from the development (in which bits of the form are allowed to evolve on their own) to the recapitulation signals a new level of integration in which the earliest statements of themes are renewed but are more replete with meaning by virtue of the elaboration of its elements over the course of the development. The structure follows the course of what John Dewey analyzes as meaningful experience generally: the evolution of themes in tension with each other toward a new integration. The development of form enables participants to undergo an integral experience that runs its course to a culmination and aftermath, just as valued experiences in life do. By sharing the shape of many meaningful developments within life itself, the musical piece resonates with meaningful experience in life generally.81 In this way, it stimulates a sense of life through musical means.
Other mechanisms for arousal are not based on musical structure. Patrik Juslin and Daniel Västfjäll indicate multiple mechanisms that can act independently or in tandem:
1. Brain stem reflexes (the arousal of the nervous system by acoustic effects);
2. Evaluative conditioning (the pairing of “positive” and “negative” stimuli in music with other positive or negative stimuli);
3. Emotional contagion (emotional mirroring or mimicry of the emotions seemingly expressed in the music);
4. Visual imagery (which is conjured by the listener in connection with music, and which itself has affective associations);
5. Episodic Memory (personal associations that have affective character);
6. Musical expectancy (the type of structural expectancy that Meyer describes); and
7. Cognitive appraisal (the sense that the music may positively or negative affect one’s goals or plans).82
Because these arousal mechanisms may operate in combination, and because they operate at different levels of mental functioning and interact at different levels, Juslin and Västfjäll suggest that mixed emotions may arise in connection with music.83 I will discuss their research further in connection with the cultural variation in musical arousal of emotion.
4. Music Induces Moods or Feelings
A fourth candidate for an explanation view of music’s relationship to emotion should be mentioned. This is the position that music arouses moods or “feelings” instead of emotions. The motivation for this view is largely theoretical. Emotion is often analyzed as requiring an object.84 Indeed, one of Kivy’s grounds for denying that music evokes “garden-variety” emotions is the argument that emotion must have an object, and music does not present an appropriate object for such responses. One alternative response to Kivy is to claim that music elicits moods, which do not take objects. Moods simply set an affective tone for experience of whatever sort.
The difficulty with this approach is that there is no consensus on how to distinguish emotions and moods, although a number of criteria have been suggested. Sloboda and Juslin contend that moods are lingering conditions, while emotions are relatively brief; that moods, unlike emotions, cannot be identified by means of a particular stimulus event; and that moods do not have distinctive facial expressions, while emotions do.85 All of these points are controversial, however. While psychological experiments tend to define emotion operationally, as brief episodes, emotions can certainly be extended over time. Some emotions, such as resentment, even seem to be long-termed by definition.86 Although moods may develop in the absence of a defining precipitating event, they sometimes do commence as a response to some occurrence. I recall several friends becoming depressed by the murder of John Lennon, and many Americans became jumpy after the terrorist attack on the World Trade Center on September 11, 2001. As for the emotions having distinctive facial displays, a theory proposed by Darwin and developed in great detail by Paul Ekman, even Ekman sometimes suggests that a distinct correlation applies only to affect programs, which in turn specify only basic emotions. This would imply that nonbasic emotions are not distinguishable on the basis of identifiable facial expression; and the display of nonbasic emotions is generally acknowledged to admit of culturally relativity.87 Ekman’s point, of course, assumes that one can distinguish between basic and nonbasic emotions, a matter that is itself controversial.88
R. J. Davidson suggests another way of putting the distinction between emotions and moods: “emotions bias action, whereas moods bias cognition.”89 In other words, emotions function to assist an individual’s action readiness, while moods influence the way information, both incoming and remembered, is processed.90 But surely moods affect action readiness. A person in a chipper mood is more geared to energetic action than one who is despondent. Emotions influence the processing of information, too. Anger at a particular person, for example, can certainly affect whether some new instance of behavior on that person’s part appears as evidence of his or her bad character.
Aaron Ridley, taking emotions to require objects, contends that music arouses feelings, not emotions, but I think his category of “feelings” amounts to what other theorists often mean by “moods.” Ridley holds that feelings “do not take particular material objects but instead consist in a tendency to regard things in general as fitting the description given by the relevant formal objects.”91 Feelings “color our world,” and such world-coloring affect is what we experience in connection with music.92 Ridley does not contend that musically aroused feeling lacks an object, but he considers it a generalized object, not a particular material one. He submits, however, that “some feelings can be interpreted as attitudes of a certain kind, where an attitude is a tendency to regard things in a particular way, often an evaluative way.”93
Certain philosophers who deny that moods lack objects take a similar approach to the issue of the affective object. Robert C. Solomon contends that moods have objects, but that their object is the world in general. He sees emotions and moods as being different domains along a continuum.94 Martha Nussbaum similarly argues that some emotions have “a highly general object (fear of unnamed and unnamable dangers, hope for something good),” which music can embody.95 Gabriela Husain, William Thompson, and Glenn Schellenberg propose that moods and emotions lie on a continuum, defining moods as “relatively long-lasting emotions.”96 Neurologist Antonio Damasio also rejects a strong distinction between emotions and moods; he calls moods “background emotions.”97
Another alternative is to deny that emotions must take objects. On this basis, Colin Radford challenges those who defend the idea that sad music does not necessarily make us sad.
Only people confused by philosophical theory about the emotions would say this cannot be sadness or, if it is, that it is unintelligible (and so probably it is not sadness!). For sadness cannot only lack an object, it can lack any obvious cause, and can be experienced in situations in which there is no focused attention, or if there is, we cannot see why our focusing on that should do anything more than coincide with the onset of our sadness. People who are sad may be puzzled as to why they are feeling sad, and what they are feeling sad about, and why it overcame them as they heard the clock strike, stepped off the bus, etc. Everyone is familiar with these experiences.98
Radford concludes that this demonstrates that an emotion need not have a particular object. A defender of the view that emotions take objects might reply that sadness is sometimes a mood (e.g., in music, where it lacks a definite object), although sometimes it is an emotion, or that one’s sadness may have an object that one recognizes only unconsciously.99 In any case, the contention that music arouses moods is sufficiently vague, and the criteria of the emotion/mood distinction sufficiently debatable, that the mood theory has not been a leading contender in the debate about music’s relation to the affective domain.
A Proposal: All of the Above
Which of the accounts just considered is right? My answer is, “All of them.” At least sometimes music has each of these relationships to emotion. Fortunately, these candidates are not mutually exclusive. The connection between emotion and music depends on the participant’s perspective, and one’s sense of one’s position in relation to the music is fluid, particularly if one is a listener. Music flows, and so does our relationship to it. One can move back and forth among the possibilities described by the various emotion-music theories. Each alternative suggests a particular relationship between the perceiver or performer and the music, and one can shift from one relationship to another during the course of listening or performing a single piece of music.
Imitation/representation theory suggests a relatively detached point of view in which a listener recognizes similarities between features of music and features of emotional experience. Some listeners will find resemblance between the structural forms of music and some features of emotions along the lines of Hanslick’s account. But a more common basis for finding imitation of emotion in music arises as a consequence of the context of music. When music is used to accompany dance, drama, or film, for example, it seems to underscore the resemblance between the music’s movements and those of emotionally motivated actors. In these contexts the resemblance between music and emotional surges, and that between music and behavior, seem unquestionably parallel, although I doubt that most audience members contemplate the resemblance dispassionately for extended periods.
Does such resemblance provide evidence in favor of imitation or expression theory? I would say, “Both,” depending on whether one is more abstractly observing the parallel forms (noting the music’s imitation of or resemblance to emotional experience) or whether one is more generically empathetic (appreciating the similarity between the music and the behavior of a person with whom one might empathize). In general, I think the various music-emotion theories operate along a continuum, with the imitation/representation theory corresponding to a relatively disinterested mode of listening, expression theory corresponding to a somewhat more engaged mode (in that one is appreciating the kinship between the music and the behavior of an agent with whom one might empathize), and arousal theory corresponding to a highly empathetic mode of engaged listening.
The mood theory also seems an apt description of some cases of music arousing affect. This is evident when music is enlisted in connection with dance, drama, and film. Usually the reasons for employing music in these contexts are to arouse the emotions and to stimulate the physiology of audience members. But this often works so effectively because music first stimulates general moods or background emotional tones. Once the mood is induced, the audience is primed for the arousal of emotion. The case is very similar to the Pavlovian case of the low two-note cello motif in the movie Jaws.100 After an association is established between this motif and sharks swimming in the deep, the motif primes audience members to jump at any hint of a surprise, for they expect a shark to appear.
Musical soundtracks are geared to emotional arousal. For example, sighing wails of the bagpipe early on during the film Titanic101 prime the audience for the poignant emotions the plot will provoke. The association of such poignancy with sweeping violin music is so common that a clichéd way of indicating that a person is expressing self-pity is to play “air-violin,” a wordless sarcasm suggesting “How sad!” In the case of such soundtracks, the music is prompting a mood, not an emotion, but at the same time it is priming one for the experience of emotions on cue. The use of “mood music” in everyday life is another case in which the mood set by music primes the occurrence of kindred emotion.
The very same similarities that justify a comparison between musical and emotional patterns—which underlies imitation theory—strongly suggest a basis for music’s ability to arouse emotion. Meyer is right in claiming that interpretive distance distinguishes emotional from analytic attention to music. He observes that one can listen from a detached analytical point of view or from a more engaged stance in which one is receptive to emotional arousal. The imitation/representation point of view presupposes the more detached point of view; the expression view presupposes a less detached outlook; and the arousal view presupposes an engaged one, which enables more direct identification with the music’s movement than the other two stances.
Davies points out that it is quite common for a person who recognizes similarities between musical patterns and the patterns of behavior typical of some kinds of emotion to feel the emotion in question, even though other emotional responses are possible as well.102 A mirroring relationship between expressed and aroused emotion is presupposed by the common tendency to reinforce one’s occurrent emotion (or mood) with music that expresses it. In this case, however, the emotion is not newly aroused by the music. The teenager who reinforces an alienated feeling by listening to music that expresses alienation (almost always texted music) uses the feeling as the principle of selection in choosing music for listening. But the chosen music effectively adds fuel to the feeling. The music in this case is something of a two-way mirror for the emotion, both reflecting and renewing it.
I suspect that mirroring emotion in connection with musical experience is more common than mirroring other people’s emotion in everyday life. In daily life one often occupies some specific role or set of roles (even if they are temporary) with respect to someone who displays emotional behavior. Such roles help to determine what emotional response is appropriate, for emotional appropriateness is often a matter of interpersonal positioning. Sometimes emotional mirroring is called for, but sometimes it is not. For example, if someone behaves angrily toward you, you may have a response that is not merely mirroring, even if you are moved to anger as well. You may feel fear if the angry person is perceived as a real threat, or contempt if the angry person is not.
In the case of music, however, one is usually not in a specific role or position vis-à-vis another person that determines the dramatically appropriate response. Accordingly, the possibility of mirroring the emotion suggested by the music—a purely empathetic response—is not blocked by one’s own practical stance toward another person who is behaving. (Conceivably, one could stand in some kind of relationship with another performer or a composer that might restrict one’s responses, such as competitive rivalry or close personal attachment. But these are not, I think, the usual circumstances of either performers or listeners.)
Common as the mirroring response is, it does not preclude other stances a listener might take toward emotion expressed through music.103 One can imaginatively move around the music’s emotional terrain. One can shift back and forth between a less engaged and a more empathetic mode of relating, to the point of identifying with the music. Efforts to adjudicate among the theories of music’s relation to emotion usually ignore the fact that the listener is able to relate to the music in a fluid way, moving back and forth between the extremes of emotionally identifying with the music and taking an analytic approach to music as an object.
Performers while practicing often experience such shifts when they go back and forth between playing a passage as they might perform it, in an emotionally engaged manner, but analytically alert to flaws in their playing, enough to be able to stop and attend to those areas that need specific work. Performers in ensembles, particularly in improvisatory contexts, maintain a dynamic equilibrium between involvement and keen attention to other performers. Paul Berliner describes the situation of improvisatory performers:
Listening is typically a dynamic activity and performers continually adopt different perspectives on the musical patterns that surround them. Their constantly fluctuating powers of concentration, the extraordinary volume of detail requiring them to absorb material selectively and developments in their own parts that periodically demand full attention, together create a kaleidophonic essence of each artist’s perception of the collective performance. Moreover . . .improvisers sometimes deliberately shift focus within the music’s dazzling texture to derive stimulation from different players.104
Similarly, it is possible for a listener to attend emotionally to an unfolding piece of music, sometimes being more detached and taking the music as an object of emotional display, sometimes focusing on the music as animate (perhaps semiconsciously thinking of the music as tracking the experience of an agent), and sometimes feeling emotional identification with music. The various types of theories describe alternative possibilities that a listener or performer might take up at different moments.
I do not mean to suggest that listeners and even performers are lurching from one state to another, overcome by emotional arousal at one moment and completely switching gears at another. Such shifts may happen, but I doubt that this is typical. Instead, what I am suggesting is more akin to Paul Thom’s notion of “playful attention” during the course of a performance. Describing the situation in the context of musical performance, Thom plausibly suggests that audience members can and optimally do engage in active play of attention over the course of listening. He itemizes six types of playful attention that can factor into a listener’s activity of interpreting the work.
1. “An audience’s attention can play between the performer’s present action and recollected past actions or anticipated future ones.
2. “An audience’s attention can play between one performer and another.
3. “An audience’s attention can play between content and vehicle.
4. “An audience’s attention may play between a particular performance as a whole and another performance of the same work.
5. “The audience’s attention can play between aspects of the performance and aspects of their own lives.
6. “An audience’s attention can play between what occurs inside the performance space and what has occurred or may occur outside it.”105
Some of these types of attentive play require a somewhat detached outlook— attention playing between content and vehicle, for example—although it would be possible to do so while largely mirroring the expressive content and flashing occasionally on how glad one is to be present for a performance that conveys it with such skill. Others of the types of attentive play—for example, the comparison of the present performance with another performance of the same work—probably necessitate sufficient distance that one is not at the moment of comparison fully identifying with the music. It would be possible, for example, to relate the present performance to another one in terms of technique or in terms of the affective feel of the performance, or to compare the two performances, sequentially, in both respects as different features draw one’s attention or come to mind. Although one might compare aspects of the performance to things going on in one’s life in a dispassionate way, I suspect that often such comparisons involve emotional assent. Each of the types that Thom mentions admit of variety with respect to the extent of detachment involved. Importantly for my purposes, Thom points to the fact that shifts in attention occur for even “consummate” audience members. In fact, he considers such active mental engagement with the music to be the hallmark of consummate listening.
Fluidity in the focus of immediate attention while attending to music is suggested, too, by reports of listeners. Alf Gabrielsson, in a study of strong emotion in connection with music, quotes a young woman who had attended a performance of Mahler’s unfinished Tenth Symphony, reconstructed.
The orchestra is breaking out in a warmth that is fascinatingly painful. I remember tears filling my eyes. I felt as if I understood a message, from one time to another, from one human to another. I sat as if turned to stone, with my fingers gripping the armrest. After a while I became quite dizzy. Then I realized that I had forgotten to breathe. . . . The last movement opens with eleven beats of an enormous wooden club against the largest bass drum. The effect is horrible. A flute, followed by the strings, starts hesitatingly to develop a melody, but it is suddenly crushed by the beat of the wooden club. I was totally taken by surprise. Each time the club beat, I was seized by terror. Each time an instrument began to sound, doubtfully, I anticipated the hit of the wooden club, I mourned. Here, too, I felt like I was a receiver of a message . . . I felt like I had not missed a single note . . . the feeling of being spoken to, strongly and directly, lived inside me.106
The young woman, admittedly giving an after-the-fact report, suggests that she comes out of a state of extreme absorption enough to realize at one point that she has forgotten to breathe. The impression of receiving a message, which she mentions more than once, seems to imply a different degree of identification with the music’s behavior than is indicated by her claim that she was “seized by terror.” I suspect that in much of our emotional lives we have occasion to move from a state of unself-conscious emoting to a slightly more analytical condition of reflecting on what we are feeling. Alternatively, sometimes we are bracing ourselves with steadying “thoughts” when we feel a surge of inchoate feeling rising to awareness, only to find that surge overwhelming our composure. I see no reason to think that our focus is completely steady in music, any more than it is in these cases.
I emphasize the fluidity of the listener’s emotional engagement as well as the relative freedom from situational features that would inhibit mirroring of expressed emotions because both are relevant to the musical participant’s relative freedom to empathize through music, regardless of practical positions vis-à-vis particular individuals in nonmusical everyday contexts. The ability to relate without attention to ordinary roles is one of the reasons that music is particularly conducive to forging emotional connections even among strangers.
CROSS-CULTURAL SIMILARITIES
Theorists who emphasize listeners’ differences with regard to emotional recognition and arousal through music presumably refer mainly to individuals with similar cultural backgrounds. We might surmise that differences would be more obvious cross-culturally, and this expectation, as we shall see, is empirically justified. However, there is evidence of much cross-cultural convergence at least in recognizing emotional expression in music. The interrelation of emotion and prosody appears to be shared cross-culturally.107 Juslin and Laukka, in their previously cited study, found specific patterns of acoustic cues for particular emotions, regardless of culture. They point out that listeners recognized the emotion performers aimed to express, with particularly strong cross-cultural correlations for anger and sadness.108
Juslin elsewhere notes that the same acoustic codes are used cross-culturally for “songs of different ‘emotional’ types (e.g., mourning, love, war, lullaby),” and that this enables people to recognize these song-types in other cultures’ music. “Mourning songs,” he observes, “usually have a slow tempo, low sound level, and soft timbre, whereas festive songs have a fast tempo, high sound level, and bright timbre.”109 He concludes that “some aspects of music (e.g., tonality, melody, and harmony) are relatively more culture-specific, whereas other aspects . . . are more culture-independent (because they are based on nonverbal communication of emotion).”110
In a series of experiments Laura-Lee Balkwill and William Thompson investigated listeners’ ability to recognize intended emotion in unfamiliar music.111 In one study musical excerpts in Hindustani ragas were presented to Canadian subjects, who rated them in terms of tempo, rhythmic and melodic complexity, pitch range, and the emotions of anger, joy, sadness, and peacefulness. These subjects’ judgments of joy, anger, and sadness (but not peacefulness) correlated significantly who had those of listeners with considerable experience with Hindustani music.112 Another study, which involved experiments with subjects from Japan and Canada, corroborates the general finding that subjects from outside a musical culture could be fairly accurate in recognizing intended emotional expression.113
Balkwill and Thompson took pitch range, tempo, and rhythmic and melodic complexity to be straightforwardly perceptual cues and considered the emotional associations with the ragas to be enculturated cues, though they acknowledge that the two types of cues can be redundant. The distinction between the two types may not be as sharp as the experimenters hypothesize, given the influence that learned schemata may have on perception. What sounds deviant from the point of view of an uninitiated listener may not sound deviant from the point of view of a musical insider, and yet the impression of deviance may seem perceptually evident. Such an impression could well have an impact on the listener’s ability to recognize emotional content, distracting attention from such content or prompting affective response.
Hanslick, with his insistence on absolute agreement, would probably be unimpressed by these convergences, given that experimenters offer their subjects a limited number of available emotion categories, so that responses within the same categorical range is taken to be agreement. Contemporary psychologists do not, however, take the establishment of constant regularities in spontaneous judgments of emotions expressed to be a reasonable goal. Juslin, for example, postulates that the apparently species-invariant acoustic cues for expressing emotions were selected by evolution for clear communication of broad emotion categories, not for nuances.114
If we do want more specificity than broad emotion terms provide, we may well emphasize cultural—and individual—differences, or at least grant their significance. One’s views about cross-cultural correlations regarding musically expressed or aroused emotions depend in part on one’s willingness to grant context a role in specifying particular emotions. The specific affective associations with particular ragas are evident only to the extent that one has absorbed these conventions, for example, even if the general emotion tone of a particular raga is more widely recognizable.
A particular difficulty that attends efforts to determine the extent of cross-cultural convergence in emotional expression is the fact that cultures do not categorize emotions in the same way. Marc Benamou, studying the use of affective terms to describe the expressive character of music among Western and Javanese subjects, ascertained that some Javanese emotion terms did not straightforwardly correspond to Western categories.115 This raises some doubts about how much we can trust studies that purport to compare cultures. Presumably, we can assume that when Javanese subjects report expressiveness in music using words for which English-speaking subjects have no term, the two groups of subjects are not recognizing the same expressive content. But more generally, we should be alert to the possibility that imperfect translations lead us to imagine greater agreement about musical expression than we would find if we had a more nuanced sense of the way the terms are used in the respective languages.
With regard to musical arousal, we cannot assume that the sharing of emotions by listeners extends to nuances of experience. In the first place, much about the emotional experience of listening to music is listener specific. Such factors as disposition to mirror expressed emotions (on the basis of the listening stance that the listener adopts, which may be influenced by temperamental factors), mental imagery, personal associations, and connection with the listener’s own goals can differ from one individual to another. The specific level of empathy adopted in a given listening experience can also affect how arousal mechanisms might interact.
Moreover, as we have already noted, one’s responses can shift as one interacts with the music, just as the emotions expressed are not constant, but sequential. This may help to explain why many find it difficult to articulate the emotions aroused in precise terms and why listeners may spontaneously choose different emotion terms in labeling the emotion(s) aroused. The fluidity with which listeners can move from one degree of emotional distance or empathy to another renders the emotional character of a particular encounter with a piece of music (as performed by oneself or by others) nonstandardizable.
Another reason that the arousal of emotion by a specific piece of music cannot be precisely delimited is that performance variables can have a powerful impact on emotional arousal. To a significant extent arousal is performance-specific, not work-specific. Charles Keil analyzes the affective impact of aspects of performance in terms of vital drive and participatory discrepancies.116 Vital drive is a term Keil borrows from André Hodeir, who contends that it is a part of “swing” and characterizes it in terms of “rhythmic fluidity.”117 In jazz, Keil observes, “To the extent that the rhythms conflict with or ‘exhibit’ the pulse without destroying it altogether, we have engendered feeling, and for a solo to grow the feeling must accumulate.”118 Vital drive is missing from Meyer’s analysis of emotion in music, Keil comments, and he offers the following principle to supplement it: “Paralleling Meyer then, the greater the processual tension and gestural uncertainty a jazz piece has the higher its value.”119
Participatory discrepancies are those nuanced deviations from precision in articulating the rhythms and pitches in music as notated or otherwise abstractly indicated. These, according to Keil, are the points at which we gain an affective impression of another human being. They also explain why human performance moves us while a computer reading a score with mechanical precision usually does not. Keil asserts provocatively, “Music, to be personally involving and socially valuable, must be ‘out of time’ and ‘out of tune.’”120 He is not particularly interested in offering labels for the emotions aroused by participatory discrepancies or vital drive. He seems to see the affective upshot of participatory discrepancies as the “feelingful” recognition of other human beings, who share and express emotion with individual precision.121
We will have occasion in the following chapter to consider further the feelingful recognition of others in connection with music. For now, I will conclude this one by moving back from our grounds for doubting that the nuances of emotional arousal are identical for any two individuals toward the more optimistic view we are likely to take if we accept broader emotional categorizations. A rather surprising aspect of emotional arousal in connection with music is the common association of music with happiness or joy, despite the fact that music can arouse other, more specific emotions, including some that are far from cheerful. Judith Becker points out:
“Happiness,” meaning some kind of strong positive emotion, is the feeling most frequently cited in connection with music listening and may constitute one of the universals of cross-cultural studies of music and emotion. From the “polka happiness” of the Polish-American parties of Chicago . . . to the !Kung of the Kalahari desert: “Being at a dance makes our hearts happy” . . . to the Basongye of the Congo who “make music in order to be happy” . . . to the extroverted joy of a Pentecostal musical service music has the ability to make people feel good.122
In a study of the musical arousal of emotion by Juslin, Laukka, and their colleagues, 1,500 adult Swedish citizens similarly ranked happiness the emotion they experience most commonly in connection with music.123 What should we make of this?
We could follow the tendency among psychologists to analyze in terms of valence, that is, in terms of positive or negative emotion. We might, accordingly, consider the characterization of music as arousing happiness, even alongside more disturbing emotions, as indicating an overall positive valence for musical experience. The notion of valence would not, however, be very informative in this context. One problem with valence is that the positive/ negative poles are not consistently defined. Are they equivalent to pleasure and pain? To “desirable” and “undesirable”? To conducing to and interfering with well-being? To stimulating approach and avoidance?124
Another problem is that various levels of musical detail might be related to valence. For example, dissonance might be perceived as “negative” in the sense of requiring movement (and hence suggesting a kind of aversion), while consonance might be perceived as “positive” because it does not. Anne Blood and Robert Zatorre point out, however, that the positive affect that coincides with chills while listening to music involves different brain structures than those related to consonance and dissonance. Moreover, the pleasure associated with consonance is mild by comparison to that involved with chills. If valence is interpreted in terms of listener enjoyment, it appears that listeners enjoy music that is sad as well as music that is happy.125 Husain and her colleagues suggest the further complication that “positive” and “negative” moods may be influenced by different neurotransmitters that may be activated.126
Perhaps a more straightforward psychological explanation of the association of music with joy and happiness concerns the activation of brain structures. Blood and Zatorre, conducting PET studies of subjects who experienced “chills of varying intensity while listening,” found brain activation patterns associated with euphoria and/or pleasant emotion, including that involved in cocaine use. “Activity in these regions in relation to reward processes is known to involve dopamine and opioid systems, as well as other neurotransmitters.”127 They suggest that the physiological reward system is engaged (and not differentially on the basis of consonance and dissonance). They conclude:
Music recruits neural systems of reward and emotion similar to those known to respond specifically to biologically relevant stimuli, such as food and sex, and those that are artificially activated by drugs of abuse. . . . Activation of these brain systems in response to a stimulus as abstract as music may represent an emergent property of the complexity of human cognition. Perhaps as formation of anatomical and functional links between phylogenically older, survival-related brain systems and newer, more cognitive systems increased our general capacity to assign meaning to abstract stimuli, our capacity to derive pleasure from these stimuli also increased.128
Reference to “the reward systems,” while perhaps accurate, contrasts with the outpouring of affect in the descriptions of those who are deeply moved by music. Many philosophers and scientists may urge caution in ascribing everyday emotion terms to music, but the musical participant rarely seems concerned to restrict his or her emotional vocabulary.129 Recall the previously cited passage from a study by Alf Gabrielsson, quoting a young woman who described her responses to a performance of Mahler’s unfinished Tenth Symphony. This woman’s report of the emotions she experienced in listening to Mahler emphasizes an aspect of emotion we have not yet attended to, the impression that music involves emotional communication. (“I felt like I was a receiver of a message . . . I felt like I had not missed a single note . . . the feeling of being spoken to, strongly and directly, lived inside me.”)
Despite the variability of emotions experienced, the association with “happiness” seems to be basic to what draws people to music. Besides the physiological impact of music on the rewards system, I submit that part of what grounds this association of music and happiness is our sense of interpersonal participation in something larger than ourselves. Even if we are listeners, we gain “that quickened sense of life” that Walter Pater describes as the ambition of the arts, and a sense that we are dynamically involved in the living world that surrounds us.130
In the following chapter I will defend the idea that music impresses upon us a powerful awareness of sharing life with others in the world, and that this awareness is a source of basic security. I will argue that one of music’s cross-cultural emotional functions is to evoke and reinforce feelings of security, and that this depends both on specieswide aspects of musical perception and on culturally specific associations. In promoting feelings of security, musical experience reinforces our sense of comfort in the world, replenishes our sense of life, and invigorates our recognition that we share a world with others.