Expressiveness as a Property of the Music Itself

Transcription

1 SAAM TRIVEDI Expressiveness as a Property of the Music Itself... the sadness is to the music rather like the redness to the apple, than it is like the burp to the cider. O. K. Bouwsma 1 I. THE PROBLEM We often say that musical passages and pieces are sad or joyous, or that they are expressive of sadness or joy. But music itself is inanimate and does not possess emotions. So, then, what does it mean to say that music is sad or joyous when it does not have any emotions? Are we using a metaphor; or do we mean in a nonfigurative (and nonliteral) 2 sense that some musical passages and pieces are sad and some others joyous? If we mean this nonfiguratively, then whose emotions, whose sadness, is sad music expressive of? Is it the sadness of the composer; or that of the performer; or that of the listener; or somehow that of the music itself? 3 I cannot do justice here to views that say talk of musical expressiveness is metaphorical; or to views (expression theories) that say such ascriptions are nonmetaphorical and that music is expressive of the emotions of the composer or performer; or to views (arousal theories) that say music is expressive of the emotions of the listener. Instead, I focus here on the intriguing idea that music is somehow expressive of its own emotions, that expressiveness is somehow a property of the music itself. This intriguing idea has been cashed out in two different ways. The first way is that of resemblance-based theories of musical expressiveness that analyze musical expressiveness in terms of music resembling something to do with the emotions. The second way is that of persona-based theories of musical expressiveness that say music is expressive of the emotions of an indefinite make-believe musical persona. 4 I proceed to examine critically both of these views, as found respectively in Malcolm Budd and Jerrold Levinson. 5 I then sketch and defend a proposal that tries in part to reconcile what I think are insights in these two views, while avoiding their respective drawbacks. II. RESEMBLANCE-BASED THEORIES In explicating musical expressiveness, resemblance-based theories of musical expressiveness differ about the locus of the resemblance between music and the emotions. Some resemblance-based theories claim the sound of music resembles our vocal expressive behavior, 6 while some others claim that the sound of music resembles the gestures and movements associated typically with certain emotions in our bodily expressive behavior. 7 Malcolm Budd has recently proposed that the appropriate resemblance lies not between our vocal or bodily expressive behavior and music, but rather between the sound of music and the feeling or affective component of the emotions. In brief, following Carroll Pratt, 8 Budd claims that music sounds the way emotions feel and that this is what accounts for musical expressiveness. This cross-categorial resemblance is further explained by Budd in terms of there being mirrorings, correspondences, and analogies between music and the affective component of emotions. For instance, the way musical tension is resolved mirrors how our desires are satisfied and our tensions released; similarly, there is a correspondence between the The Journal of Aesthetics and Art Criticism 59:4 Fall 2001

2 412 The Journal of Aesthetics and Art Criticism way music attains its intermediate and final goals and how we attain our desired ends; between musical motion and certain felt movements essential to certain emotions (e.g., tremblings intrinsic to acute agitation); and so on. Budd claims that in hearing music as expressive, we perceive, often subconsciously, these cross-categorial resemblances between how music sounds and how emotions feel. In addition, Budd claims that while the basic and minimal concept of musical expressiveness is that music sounds the way emotions feel, there are nevertheless many concepts of musical expressiveness and no single one. To his basic and minimal concept of musical expressiveness Budd claims we must disjunctively add three accretions or emendations as concepts of musical expressiveness so as to capture the variety of experiences of musical expressiveness. One accretion involves the idea that we experience music as expressive when we resonate emotionally with it, and experience the music with the feeling it sounds like; a stronger version of this involves the claim that only if music arouses feelings in us do we hear it as expressive. A second accretion involves the claim that music that sounds like a feeling may induce us to imagine our undergoing that feeling we hear the music as sounding like; in such cases we imaginatively identify with the music and imagine we are the music. 9 A third accretion involves the idea that we imagine the music we hear (as sounding like how an emotion feels) as an instance of the emotion, without our figuring in it; we need not conceive of the imagined emotion as our own, nor as that of the composer s, but can imagine the emotion as belonging to the music s imagined persona. 10 Although I think Budd is right to say there are a variety of experiences of musical expressiveness that cannot be captured by a single, monolithic concept of musical expressiveness, I would register the following reservations for his view. To begin with, considered in itself, without the accretions, the basic and minimal concept that Budd gives us is a concept of resemblance that loses the link with our ordinary concept of expressiveness. Our ordinary concept of expressiveness involves the outward manifestation of inner mental states. As a species of expressiveness in general, musical expressiveness must relate somehow to this ordinary concept. Moreover, a theory of what it means for music to be expressive must relate somehow to what it ordinarily means to be expressive, as given by our ordinary concept of expressiveness. But Budd s basic and minimal concept in terms of music s sounding the way emotions feel gives us a concept of resemblance that does not relate intelligibly to this ordinary concept of expressiveness. Thus there must be doubt whether it is really a basic and minimal concept of musical expressiveness. Budd might reply that the three accretions connect the basic and minimal concept of resemblance to the ordinary concept of expressiveness, for they involve the idea of someone s (who is either ourselves really, or ourselves make-believedly, or a musical persona) states being expressed, i.e., manifested outwardly. In other words, he might claim that it is via this link between his basic and minimal concept and the three accretions that we are told what it means for music to be expressive. But, even so, the problem remains that Budd s basic and minimal concept does not do so by itself, and there must remain doubts about whether it is not too minimal to be a basic concept of musical expressiveness. Furthermore, and relatedly, mere resemblance by itself would not seem sufficient to account for expressiveness, unless the resemblance in question is tied to someone s or something s mental states being really or imaginarily manifested outwardly, as urged by our ordinary concept of expressiveness. For example, the mere fact that snails and turtles move slowly, resembling the manner in which sad people often move slowly, is not sufficient to show that the slow motion of snails and turtles is thereby expressive of their or someone else s sadness, unless we link this motion to the outward manifestation of the real or imagined sadness of snails and turtles or something or someone else. Likewise, the mere fact that something crosscategorially resembles how emotions feel, and that these likenesses are perceived, does not show that the thing is thereby expressive, nor that it should be, or is, seen to be so. I suggest in this vein that some foods may taste the way some emotions feel and that we perceive these cross-categorial likenesses. For example, an extremely hot and spicy food may taste the way extreme anger, or anguish, feels in that there may be a fiery, burning affect common to both;

3 Trivedi Expressiveness as a Property of the Music Itself 413 similarly, a soothing beverage may taste the way calmness, or contentedness, feels in that there may be a pacifying, cool feeling in common. The ingestion of such foods may cause or evoke such emotions and feelings, but that is a separate causal story with which I am not much concerned. Rather, my concern is with the very experience of tasting these foods, getting to know what the gustatory feel of such foods is, and what their phenomenological what-it-feels/ tastes-like aspect is. Now, if indeed there are such likenesses between some foods and some emotions, and if indeed we perceive them, that alone does not suffice to show these foods are expressive of these emotions, nor that they are, or should be, judged to be so. I thus suspect that we need something more than Budd s too-broad resemblance concept to account for musical expressiveness; at the very least, we need some connection with our ordinary concept of expressiveness that is part of the basic and minimal concept itself that Budd gives us. A second worry for Budd is this. What we want to know from a theory of musical expressiveness is what it means to say that music is expressive of sadness, what it means to say that music is sad, given that music is inanimate. However, in giving us a concept of resemblance, Budd has not told us what it means for music to be expressive; that job is better done by theories that relate what it means for music to be expressive to what it means ordinarily to be expressive. Instead, Budd has given us merely the causal basis or ground of musical expressiveness and experiences of it. What causes or allows music to be, and to be experienced as, expressive is the fact that the sound of expressive music resembles the affective feel of emotions in the ways Budd specifies, and also resembles our vocal and dynamic expressive behavior. Indeed, composers and improvisers often consciously or otherwise exploit resemblances between music and the emotions to make their music expressive. Moreover, as listeners we hear music as expressive precisely because it resembles these aspects of the emotions and we consciously or otherwise perceive these resemblances. But this causal story does not tell us what it is or what it means for music to be sad or joyous, given that it is inanimate; it only tells us the causal factors in virtue of which music is expressive. 11 The two worries raised so far for Budd apply to all resemblance-based theories of musical expressiveness, irrespective of where they locate the relevant resemblances between music and emotions. In addition, a third minor reservation I have is that Budd s first accretion captures not an experience of expressiveness but rather one of arousal and thus must be excluded from an account of musical expressiveness that tries to capture the variety of genuine experiences of musical expressiveness. In contrast, Budd s second accretion captures not an experience of real arousal of emotions, but rather one of, at best, an imagined arousal as the listener imaginatively identifies with the music and imagines the music is expressing her or his emotion as she or he imagines undergoing that emotion. This second accretion may involve a genuine experience of hearing musical expressiveness, for the imaginative identification involved in it may be a very powerful feature of some experiences of hearing music as expressive: the imaginative identification may involve the experience that it is as if we are the music. III. PERSONA-BASED THEORIES I turn now to the persona-based account advanced by Jerrold Levinson. 12 Levinson s view applies the ordinary notion of expressiveness to musical expressiveness and its concept, claiming that musical expressiveness and its experiences involve readily and immediately hearing the music as the manifestation of the emotions (or other mental states) of an imagined, indefinite musical persona. What this amounts to is that when we hear sad music as sad, we make believe that an indeterminate persona s sadness is being expressed musically: it is as if an indeterminate someone in the music is crying, or wailing, or otherwise expressing their sadness, though in a musical manner. This imagined persona, it should be noted, is a type of individual, not a particular individual. Levinson s view relates quite easily to our ordinary concept of expressiveness, and it also does not face the problem of giving us a mere causal story behind musical expressiveness. However, my worry is that its concept of musical expressiveness may be too narrow to be adequate, for it does not cover the variety of experiences of musical expressiveness. To be adequate, a theory of what it means for music to be

4 414 The Journal of Aesthetics and Art Criticism expressive must cover, or apply to, all genuine experiences of musical expressiveness. To find whether it is adequate, a concept of musical expressiveness must be tested empirically against experiences of expressiveness: its answer to what it means to say that music is expressive must be true to the various experiences of musical expressiveness. Reflecting on my own (possibly idiosyncratic) experiences of musical expressiveness, I find that whereas I sometimes do make believe that there is an indeterminate, imagined musical persona that is expressing its emotions musically, 13 I nevertheless do not always do so in experiencing music as expressive; indeed I do not even do so most of the time. There are other experiences of musical expressiveness that Levinson s view does not accommodate, notably the second accretion described by Budd, and also the experience I will presently describe wherein we make-believedly animate the music itself as the very being that is alive and expressive of its own emotions. If this is right, then Levinson s view may only be partially adequate. 14 IV. A POSITIVE PROPOSAL: VARIATIONS ON A THEME From the preceding discussion it emerges that an adequate theory of musical expressiveness must meet at least three requirements: (i) as a species or at least an intelligibly related cousin of expressiveness in general, it must relate to our ordinary notion of expressiveness involving outward manifestation of mental states; (ii) it must explicate what it means for music itself to be expressive, rather than give us a mere causal story about what allows music itself to be, and to be heard as, expressive; and (iii) it must be a nonmonolithic theory that embraces the variety of experiences of musical expressiveness. I now sketch a proposal that meets these three requirements better than any other theory does and that tries partly to reconcile what I think are insights in the theories of Budd and Levinson. The proposal consists of three variations, all of which involve make-believe, 15 on a theme that relates to our ordinary concept of expressiveness. The theme is that when we hear music as expressive, we make believe that someone s or something s mental states are being expressed, i.e., outwardly manifested, musically. Who or what this someone or something is may vary across at least three different sorts of experiences of musical expressiveness, and thus we get at least three different answers to the question, Whose emotion is the music expressive of? There are at least three sorts of experiences, if not more, of musical expressiveness that all play on this theme that make-believedly someone s or something s emotions (or mental states in general) are being expressed musically. One sort of experience, as Budd claims in his second accretion, involves our make-believedly identifying with the music so that we imagine that the music is make-believedly expressive of our own emotion. 16 A second sort of experience, as both Levinson and Budd, in his third accretion (as presented above) claim, involves our making believe that the music is expressive of the emotion of an indefinite musical persona. A third sort of experience involves this: though we know fully well that music is inanimate, nevertheless, on some occasions, in hearing music as expressive we make-believedly animate the music itself so that the music itself is imaginatively experienced as the very being that is animate and whose emotions are being expressed musically. 17 So far, I have sketched a disjunctive account positing a disjunction of at least three sorts of experiences of musical expressiveness that vary the same theme on which there may also be other variations. 18 I suspect the third sort of experience described above may be the most common sort of experience of musical expressiveness, and I proceed to describe it more. The third experience, then, involves our make-believedly animating the music itself, even though we know full well that music is not animate. 19 In hearing the music, we makebelievedly bring it to life (at least in our imaginations) and imagine that it is the very being whose emotion is being expressed, so that it seems as if it is the music itself that is sad or joyous. The sadness expressed musically is felt as belonging to the music itself, as opposed to belonging to something other than the music, such as a persona, or us as listeners, or the composer, or the performer. Imaginatively, readily and immediately, on such occasions we endow the music with life and lifelike qualities, and this make-believe animation of the music involves imaginative projection of emotional and other qualities onto the music. 20 Specifically, our animating the music involves four things: (i) we imagine the music is the kind

5 Trivedi Expressiveness as a Property of the Music Itself 415 of thing that is alive; (ii) we imagine the music is the kind of thing that has psychological states; (iii) we imagine the music itself is sad or joyous (or whatever state it is appropriate to imagine the music as possessing); and (iv) we imagine that the music is an audible thing or an audible being that expresses its sadness or joy by presenting an aural appearance apprehended aurally. 21 It should be noted that we make believe the music is sad without making believe, or without having to make believe, that the music has complex, functional organizational states of the sort that seem necessary, according to currently orthodox theories of the mind, for it to have psychological states. This is similar to the way in which children can make believe that a block of wood is a truck without making believe, or at least without having to make believe, that the block of wood has an engine, a fuel tank, and so on; indeed one can, somewhat prosaically, point out to a child that the block of wood cannot be a truck because it lacks an engine, a fuel tank, etc., and yet the child can continue to make believe that it is a truck. As an example, consider the opening passages of the second movement of Beethoven s Ninth Symphony in D Minor, op We hear these passages, I suggest, as expressive of a vigorous and exuberant joy that is almost manic, or something in that ballpark. But whose vigorous and exuberant joy is being expressed musically? Often in hearing the music as expressive in these terms, it is felt as if it is the music itself that is exuberantly joyous, not something other than it such as a persona. We imagine often that the music itself is alive and possessive of psychological states, that it is exuberantly joyous, and that its joy is manifested through the sounds we apprehend aurally. Our animating the music itself is analogous to the way in which comic strips often imaginatively animate many inanimate objects such as cars themselves as possessing eyes (the headlights), a mouth (the radiator), and a smile or a glum look, without constructing an indefinite make-believe automobile persona that has these features, and without our imaginatively identifying with cars. It is not as if, makebelievedly, there is an indeterminate someone in and distinct from the car let us call the car Donna that is smiling. Rather, makebelievedly it is as if it is the car, Donna, itself (or herself, if you like) that is smiling. Within the world of the comic strip, when Donna the car is damaged and splutters, we imagine that it is Donna the car itself, and not an automobile persona, that has been damaged and that is expressing its sadness visually. When the car is then repaired and back in racing condition, within the world of the comic strip, we imagine that it is Donna the car itself that, restored to health, is now expressing its happiness visually, by means of a smiling mouth (the radiator) and eyes (the headlights). 22 Our animating music is, I suggest, an abstract instance of the same kind of activity that is involved in the animation of inanimate things such as cars in comic strips. To illustrate this point, it will be helpful to move away from music, precisely because of its abstract and often intractable nature, and consider two drawings, the first of which depicts a glorious day and represents the sun as smiling, and the second of which depicts a witheringly hot day and represents the sun as angry. In such instances, we see the representations of the sun showing it as smiling or as angry. But who is it that is smiling or angry: Whose emotions are expressed pictorially? There is not an indeterminate make-believe solar persona that is distinct from the sun and in the sun or in its pictorial representation, and whom we imagine to be smiling or angry; nor do we imaginatively identify with the smiling or angry sun so that we might say that it is make-believedly our emotions that are expressed pictorially. Rather, it is the sun itself that is smiling or angry; that is to say, we imagine the sun itself is alive and smiling or angry. We make-believedly animate the sun itself, even though we know the sun is inanimate. The causal story behind this that allows us to do so involves the resemblances between the faces of smiling or angry people and the features of the sun (the eyes, mouth, and forehead) as represented by the artist. The case with hearing music as expressive is often similar, though it involves an abstract and hard-to-detect instance of the same kind of activity: we often imagine it is the music itself that is alive and sad or joyous. There is, though, the difference that these pictures involve representations of the sun, not the sun itself, whereas the musical case involves not representations of it but the music itself. But this difference can be noted and left aside, for it is not clear that it amounts to much. What sort of being is the music that we animate and make-believedly bring to life? It is an

6 416 The Journal of Aesthetics and Art Criticism indeterminate, audible, abstract being, not ordinarily familiar to us, and it make-believedly possesses and expresses mental states such as emotions. Moreover, our make-believe animation of the music may happen without our realizing this is what we are doing. Sometimes our imagined animation of the music may be more specific, in that we may make-believedly anthropomorphize the music, so that the music is experienced not merely as the very being but as the very person that is crying, wailing, dancing, and the like. There are two respects in which this third experience is different from the second one, involving an indefinite make-believe musical persona. First, the third experience does not involve a mediating indefinite musical persona that is in the music and thus distinct from (not in the sense of being detachable from but rather in the sense of being other than or different from) the music itself. In this sense, the persona theory is arguably not an account of musical expressiveness that involves the music itself; or at the very least, it does not involve the music itself as much as the third experience described above does. Secondly, the third experience is thus a more ready, immediate, and direct experience of musical expressiveness than the second one, which involves positing a mediating make-believe persona. What permits or causes us to animate the music may be resemblances of different sorts between music and the emotions, resemblances between music and the affective feel of the emotions, and between music and our bodily and vocal expressive behavior. But I do not know why we make-believedly animate the music itself, even though we know fully well it is not animate; the answer may have something to do with us as human beings, our imaginative faculties, and our mental states and lives. 23 Notice now that the way I posed the problem of musical expressiveness relates easily to this third experience. The problem is to account for what is meant by saying that music is sad or joyous, given that it is inanimate. The answer suggested relates to our ordinary notion of expressiveness in that we make believe the music is expressing someone s or something s mental states in either of three different ways. The third of these ways involves our make-believedly animating the music itself so that it is felt as if the music itself is something that is alive (even if abstract) and that make-believedly expresses its own states, such as sadness and joy, through its musical properties, gestures, development, and the like. This third experience dispels in the simplest way possible the mystery about how music that is inanimate and known to be so can itself be said to be sad or joyous. It also tells us in the simplest possible way how the emotion expressed musically can make-believedly be a property not of the composer or performer or listener but of the music itself, without appealing to resemblances or make-believe personae. Moreover, it may be the very simplicity and directness of this third experience that may account for why it is the most common experience of musical expressiveness by my lights: it is the simplest, most direct, most ready, and most immediate experience of musical expressiveness, in that it does not involve positing a make-believe persona, nor our make-believedly identifying with the music. Note also, to go back to the quotation from Bouwsma at the beginning of this paper, that the account proposed above makes sense of the idea that sadness is a property of the music itself in the way that redness is a property of the apple. Redness is a secondary property of the apple, a dispositional property not independent of us as perceivers: the apple is red if normal observers are disposed to see it as red under standard conditions. Likewise, on the account proposed, sadness is like a secondary property of the music, a dispositional (and not strictly intrinsic) property of the music that is perceiver-dependent, and in particular dependent on our capacity for make-believe. A musical passage or piece is sad (or is expressive of sadness) if competent listeners (those who are musically sensitive, informed to some minimal degree, and so on) are disposed, under standard conditions (they are not bored or tired or distracted or satiated by the music, and so on), to hear it as make-believedly sad (or as make-believedly expressive of sadness). Whose make-believe sadness the music is expressive of varies in at least three different ways, as proposed in the three variations above. 24 V. OBJECTIONS AND REPLIES I now want to respond to some possible objections to the proposal sketched above. As the experience of make-believedly animating music itself is what is most distinctive about the pro-

7 Trivedi Expressiveness as a Property of the Music Itself 417 posal, I focus here on possible objections to that part of my suggestion. One possible objection call it the incredulous objection may be that the suggestion that we animate the music itself is a simplistic solution to our problem. After all, if the solution is indeed as simple as my suggestion makes it to be, how could it possibly have eluded us so far? In response, I suggest that the fact that we often animate the music without realizing this is what we are doing and the abstract nature of music, these two things together, make it hard to detect this kind of experience. Furthermore, I suggest that flirtation with metaphor-based views, under the influence of the linguistic turn in philosophy, and also flirtation with arousalist views, and thus with our aroused reaction to expressive music, may have led us away from focusing on the very experience of hearing music as expressive, thus missing acknowledgment of the third experience described here. A second line of possible objections is this. On the third variation above, music itself make-believedly has emotions, and it also make-believedly expresses emotions. But the two features should be kept distinct as involving distinct things, as claimed by persona-based views that claim it is the persona that makebelievedly has the emotions, while it is the music that make-believedly expresses them. In response, I see no problem with one and the same thing both having and expressing emotions. I see no reason why there cannot be a partial analogy between music and persons (and indeed other animate things) in this respect: sad music can be like sad persons who both have and express sadness, with the difference that, unlike persons, music only make-believedly has and expresses emotions. The objector may, at this stage, continue and say the analogy is not a happy one, for sad persons have certain features in virtue of which they express sadness. My response is that music too has certain features, sonic features, in virtue of which it expresses sadness and is heard as doing so. These causal features are resemblances between how music sounds and different aspects of the emotions, as well as musical gestures, musical properties, and their development, and so on. But, the objector may continue, the analogy does not hold up because while music is the sum of these sonic features that make it expressive, persons are not the sum of the features in virtue of which they express their sadness. There are two possible responses to this objection, a less promising one and a more promising one. The less promising move is to defend some sort of principle of organic unity as applied to music, claiming that music is not the sum of the sonic features in virtue of which it is expressive, but rather an organic whole. I cannot develop this move here, which in any case I do not find very plausible. The more promising move is to claim that even if music just is the sum of these features, that need not be a problem, for the analogy between music and persons is only partial. Here is a third objection. We often say of our behavior that it itself is exuberantly joyous. So, how is the music being joyous different from behavior being so? How is it that we do not animate behavior that we often say is itself joyous, but we do animate music itself, on the proposal above? My response is this. We do not animate behavior for we do not need to do so. For, with the exception of fictitious oddities such as the Cheshire cat s smile, there is no such thing as disembodied behavior: there is always a behaving agent associated with behavior, a behaving agent of whom we can say, in response to the question Whose behavior is exuberantly joyous? that her or his behavior is exuberantly joyous. In contrast, to answer the question whose exuberant joy music is expressive of, given that music is nonsentient, we do need some sort of agent of whom we can say that her joy is being expressed musically. On the proposal above, often the agent is make-believedly ourselves, and often the agent is an indefinite make-believe persona. But very often the agent is the music itself, animated in the manner suggested above as the very being or thing that is exuberantly joyous. 25 A fourth objection is this. In all the talk of animation above, the examples used have been those of things such as the (comic-strip) sun, (comic-strip) cars, toys, trees, and the like. But these are all more or less stable substances, unlike the temporal art of music, works of which are dynamic processes that unfold over time. Given this difference, the analogy between music, on the one hand, and the sun, trees, etc., on the other, is structurally imperfect. Besides, it is not clear that there are any everyday processes

8 418 The Journal of Aesthetics and Art Criticism that can be animated as beings in the manner in which it is suggested that we animate the musical process. And if we cannot find any process in life that could count as a being, then perhaps one cannot represent the musical process either as a being. This is a good objection, I think. In response, the first thing I wish to say is that if one adopts a process-ontology (of the sort associated with Heraclitus, Whitehead, and the Buddhists) and this is certainly an open option then one can view the sun, trees, and indeed many other everyday things as processes that are subject to constant, even if slow and imperceptible, change. Specifically, one can adopt a weak or a modest process-ontology according to which all these things are processes in that at any given time, there is at least one property they possess that is subject to change, even if all other properties remain constant at that time. So, in the case of the sun, there is change in its position in that the sun rotates on its axis, in its temperature in that it is going through a long drawn-out cooling process, in its intensity in that it is constantly emitting solar radiation, in its age in that it is constantly growing older and nearing its end, becoming a white dwarf some five billion years from now, and so on. But this sort of process-ontology may not convince some, so let me say, by way of a second line of defense, that even if the analogy between the musical process and (alleged) substances such as the sun, trees, and the like is not perfect or total and only partial, it is not clear that that amounts to much. There is no reason why one cannot animate familiar processes if one wants to, though usually we do not do so, nor is there usually need to do so. As an example, I mention the seasons, which can be viewed as annually recurring processes. Take winter, for example, which we often do animate as Old Man Winter: winter arrives on the scene as fall departs, takes some time to settle in, has us in its grip after a while, and then wanes a bit, only to come back and often hit us with one last pounding before fading away and giving way to spring. A different example of a process that we animate is provided by meteorological processes such as hurricanes, which we animate variously as Hurricane Floyd or Hurricane Georges and so on. Another example of a process that is animated, at least in some cultures and for sure in pagan traditions, is rivers, which are constantly flowing and thus the subject of the famous Heracleitian quote. We talk of raging rivers and calm rivers; some cultures even animate and worship rivers as goddesses, as do the Hindus with the river Ganges, most notably. Moreover, even if one cannot directly animate everyday processes, such as the earth s rotation on its own axis or its revolution around the sun, one can associate them with an animated being, Mother Earth let us say, who rotates and revolves. In such a case, the rotation of the earth itself is not a being or a person, true, but as a response to the question Who or what is rotating? we can animate the earth as the very thing that is rotating. Similarly, as a response to the question Who or what is expressing emotion E through music M? we can answer that it is the music itself, and, amongst other ways acknowledged in the variations above, that makes sense if we animate the music. I have attempted to sketch and defend above an account of how expressiveness can be a property of the music itself in a way that builds on and refines earlier attempts at similar accounts; the attempt is akin to the way in which Functionalism built on and refined earlier Materialist theories of the mind such as Behaviorism and the Identity Theory. I leave it to the reader to decide whether such a full-blown account of how music itself can be sad or joyous is plausible. The next task would be to show why the above account s main current rivals metaphor-based theories and arousalist theories of musical expressiveness are inadequate. That task cannot be performed here, and shall have to await another occasion. 26 SAAM TRIVEDI Philosophy Department Simmons College 300 The Fenway Boston, Massachusetts INTERNET: saam.trivedi@simmons.edu 1. See O. K. Bouwsma, The Expression Theory of Art, in Aesthetics and Language, ed. W. Elton (New York: Blackwell, 1954). 2. It seems to me that there are at least three ways in which one can construe The music is sad. Taken literally, it is false, for given that music is inanimate and devoid of emotions and other mental states, music cannot literally be sad. The second way is to claim that music is metaphorically sad,

9 Trivedi Expressiveness as a Property of the Music Itself 419 but this runs into the problem that any possible paraphrase of the alleged metaphor will be in nonemotive terms (either in technical terms of musical analysis or in layman s phenomenological terms) and will fail to capture the emotive significance of the alleged metaphor. This raises doubts whether the locution The music is sad is metaphorical in the first place, assuming metaphors are paraphraseable. (For similar criticisms of metaphor-based theories of musical expressiveness, see Malcolm Budd, Music and the Emotions (London: Routledge, 1985), chap. 2. See also Stephen Davies, Musical Meaning and Expression (Cornell University Press, 1994), chap. 3. Finally, the third possible construal is to claim that music is make-believedly sad, that it is imagined by the listener to be sad. This is, I think, the most promising construal, and my positive proposal in this paper tries to build on it. 3. The problem of musical expressiveness is seldom stated in the by now vast literature on musical expressiveness, and the few statements of it one finds leave something to be desired by way of clarity; a notable exception to this trend is Stephen Davies in the preface to his Musical Meaning and Expression. I believe stating the problem clearly can be insightful in proposing solutions to it. 4. Perhaps a tighter formulation is to say that the music is make-believedly expressive of the emotions of an indefinite musical persona, as the formulation above seemingly suggests quantification over make-believe musical personae. 5. See Malcolm Budd, Values of Art (London: Penguin, 1995), chap. 4. See also Jerrold Levinson, Musical Expressiveness, in The Pleasures of Aesthetics (Cornell University Press, 1996). 6. See Peter Kivy, Sound Sentiment (Temple University Press, 1989). 7. See Stephen Davies, Musical Meaning and Expression, chap See Carroll Pratt, The Meaning of Music (New York: McGraw-Hill, 1931). 9. Budd also claims as part of this second accretion that we imagine the feeling from the inside ; we imagine of our auditory experience of music that it is an experience of our undergoing the emotion we hear the music as sounding like and expressing. The latter claim was originally made by Kendall Walton in What Is Abstract About the Art of Music? The Journal of Aesthetics and Art Criticism 46 (1988), and later developed further in his Listening With Imagination: Is Music Representational? The Journal of Aesthetics and Art Criticism 52 (1994). 10. More accurately, perhaps, the third accretion involves the idea that the music is imaginatively heard as an impersonal instance of the feeling or affective component of emotions, a feeling that is not anyone s in particular, not even perhaps that of a persona. One might ask how a feeling can be disembodied, as it were, belonging to no one in particular, neither really nor make-believedly. The reply to this worry would be that in the relevant case we are asked to imagine merely that the music is an instance of feeling, without having to imagine anything further than that, and in this sense the question Whose feeling? somehow drops out as illegitimate; it is not that the feeling is nobody s, but rather that we are not asked to imagine that the feeling is any particular person s. Be that as it may, one worry for this representation of Budd s third accretion is that given that it involves a mere feeling that is not anyone s, it is not clear it gives us an account of (or a description of an experience of) expressiveness understood as the outward manifestation of someone s inner mental states. Perhaps what it does instead is to give us an account of what it is sometimes for music to be, and for us to experience as, emotional. 11. This worry was raised by Budd himself in The Times Literary Supplement (1981): 762 in a review of Kivy s vocal resemblance-based theory of musical expressiveness, and also elsewhere ( Music and the Expression of Emotion in The Journal of Aesthetic Education 23 [1989]: 27) before he revised his view: It is mistaken to attempt to use...aperceived resemblance of any kind in an elucidation of the concept of the musical expression of emotion. Such an analysis is wide of the mark, whether the resemblance is... between music and the vocal expression of emotion, the expression of emotion in the body, or emotion itself (emphasis added). 12. Levinson s view modifies, and adaptively applies to music, Bruce Vermazen s view about artistic expressiveness in general. See Bruce Vermazen, Expression as Expression, Pacific Philosophical Quarterly 67 (1986). But while Vermazen s judgment-based view, as applied to music, seems committed to the idea that we infer the music is expressing the emotions of an indeterminate utterer, Levinson rightly rejects the ideas of inference and utterer and claims we directly and immediately hear music as expressive. A persona-type view has also been proposed by psychologist Roger Watt at Stirling University, who suggests that music creates in our minds a virtual person whose character, feelings, and behavior may be inferred from the music; see John Sloboda, Brain Waves to the Heart, The BBC Music Magazine, November Most notably, I have on occasion experienced the opening clarinet glissando in Gershwin s Rhapsody in Blue in terms of a make-believe persona that is wailing or crying through the music. Likewise, I have sometimes heard John Coltrane s music in similar terms, as if an indeterminate, imagined persona in the music (more precisely in the sound of the sax) is expressing emotions musically. 14. Similar reservations concerning how often in hearing music as expressive we make believe an indeterminate musical persona is expressing its emotion, if we do so at all, have been raised for Levinson s view by Robert Stecker (Artworks [Pennsylvania State University Press, 1997]); by Stephen Davies ( Contra the Hypothetical Persona in Music, in Emotion and the Arts, ed. Mette Hjort and Sue Laver [New York: Oxford University Press, 1997]); by Malcolm Budd ( Music and the Communication of Emotion, The Journal of Aesthetics and Art Criticism 47 [1989]); and by Aaron Ridley (Music, Value and the Passions [Cornell University Press, 1995]). 15. What is it to make believe? As an answer to this query, I follow Kendall Walton s analysis (see Pictures and Make-Believe, Philosophical Review 82 [1973]; and also Mimesis as Make-Believe [Harvard University Press, 1991]) of make-believe: make-believing (or making-believe, if you prefer) is a kind of imagining, and imaginings can be spontaneous, nondeliberate, passive rather than intended. Imaginings can also be unconscious or subliminal or implicit, and one may imagine something without noticing that one is doing so. 16. Compare the similar idea put forth in Chapter 6 of Aaron Ridley s Music, Value and the Passions. 17. The positive proposal made here can also be seen as

10 420 The Journal of Aesthetics and Art Criticism reconciling aspects of what Levinson calls make-believe-based views of musical expressiveness and what he calls appearance-of-expression-based views (see Levinson, Musical Expressiveness ), and in claiming this I tend to agree with Robert Stecker that the two sorts of views are similar, though he thinks the make-believe-based view may be a variant of the other sort of view (Robert Stecker, Expressiveness and Expression in Music and Poetry, n. 5, The Journal of Aesthetics and Art Criticism 59 [2001].) On my proposal, due to its resemblance with the feeling component of human emotions and their vocal and bodily behavioral expressions, we make believe that music presents us with an aural appearance of someone s emotions being expressed musically, and who or what this someone is varies in at least three different ways, all of which involve make-believe. 18. A possible variation of the third variation (or a fourth variation, if you like) is that sometimes we may animate not the music (i.e., the sound we hear) itself but the orchestra (or the chamber group or the string quartet or the duo) or whatever instruments we hear as the very being that is sad or joyous; it is interesting to note in this connection that there is at least one popular song that sings of one s guitar gently weeping. 19. In his book Sound Sentiment, Peter Kivy also talks of our animating music so as to experience it as expressive: We tend to animate sounds as well as sights...just as we see the face in the circle, and the human form in the wooden spoon, we hear the gesture and the utterance in the music (p. 58). In listening to... music, we need not be, and often are not, conscious of the life in the music, when, by means of it, we hear the expressiveness in the music (p. 173). I agree with Kivy that we often unconsciously animate, and especially anthropomorphize, many things, ranging from animation in mythology of such inanimate objects as the sun (animated by the ancient Greeks as Helios and by the ancient Indians as the sun-god Surya) to animation of the elements (earth, air, fire, water, rain, wind, etc.) in pagan religions to more contemporaneous animation of things such as decorative coffee-mugs (I have in mind a particular coffee mug I once saw, which involved a life-like depiction of the handle as the nose, and carvings of eyes above the nose and a huge smile below the nose) to animation of toys by children. But I am inclined to go beyond Kivy and say that we do not just hear expressiveness and gesture in the music; often we animate it as the very thing doing the expressing (or gesturing or uttering, if you like). (And I am also tempted to claim that in calling willow trees sad due to their droopiness, which resembles the bodily expressiveness of sad people, we are not using a metaphor, nor making arousalist claims, nor talking in terms of our imaginative identification with the willow tree, nor always imagining there is a persona in the tree. Rather, at some level, we often imagine, and are asked to imagine, that the willow tree is not only alive, it is the sort of thing that has psychological states, it itself is sad, and it is expressing sadness through presenting the visual appearance of drooping.) 20. My talk of projection here echoes Wollheim, who could, I think, claim similarly that we make-believedly animate landscapes themselves in imaginatively projecting lifelike qualities onto them and experiencing them as expressive. See Richard Wollheim, Correspondence, Projective Properties, and Expression in the Arts, in The Mind and Its Depths (Harvard University Press, 1993). 21. A similar suggestion was made, I believe, by the Italian Renaissance philosopher Marsilio Ficino ( ), who claimed in his music-spirit theory that music is like a living spirit or an animal that itself moves. See D. P. Walker, Spiritual and Demonic Magic (Leiden: The Warburg Institute, 1958). 22. I am quite sure that children, who have greater and fresher imaginations than many adults, will agree more readily and immediately than will many grown-ups to this point that we imagine that it is comic-strip cars themselves, and not automobile personae distinct from and in them, that are sad or happy. If in doubt, ask a child who it is that we see as sad or happy when we see comic-strip depictions of cars. 23. Maybe we like to animate, and especially anthropomorphize, various inanimate things in the world, such as objects in nature, so that we can then imaginatively identify with them and get emotional and other rewards from doing so: amongst other things, we realize make-believedly through doing so that we are not alone in feeling certain emotions. In addition to its being psychologically reassuring, perhaps we also do so because it is playful fun to do so: we can thus identify with or relate to more things in an imaginative and creative way. 24. An account of musical arousal consistent with the above proposal for musical expressiveness would run roughly as follows. We are moved in real life by the sight of someone in distress because we identify with (or sympathize, or empathize with, if you prefer) their plight. Similarly, we are often moved or aroused by plays, stories, and novels because we identify with the characters therein, their feelings and predicaments. Likewise, we are often aroused by music because we identify with the make-believe musical persona or with the music itself animated as the very thing or being that is sad or joyous. Sometimes, though, we can be musically aroused to a sadness (or a joy) that is about nothing in particular, neither an imagined persona nor the music itself animated as the very being that is sad, in which case it is a nonintentional mood or a feeling, not an emotion, that is musically aroused. 25. It is important to some degree, I think, that the reader focus on her or his very experiences of hearing music as expressive on various occasions. Perhaps my proposal will be more cogent to those who find that at least sometimes they do feel as if the emotion expressed musically belongs to the music itself, not to something other than it. And perhaps such experiences are more likely to be had when one plays an instrument or improvises on it. 26. I am grateful to John Brown, Malcolm Budd, David Cummiskey, Berys Gaut, Jerrold Levinson especially, Allen Stairs, Susan Stark, and Robert Stecker for comments and suggestions on this paper s various ancestors. A version of this paper was read at Bates College, and I am grateful to those present on that occasion for their feedback. I would also like to thank an anonymous reviewer for this journal.