In thinking of imagination in relation to music, it seems clear that the creative imagination is involved in such musical activities as musical composition and improvisation. Perhaps composers, in writing musical works, imagine musical forms, timbres, textures, and the like by creating images of these in their auditory imaginations, and then later often but not always test their hypotheses about these images through actual music-making (Levinson 1992: 84–5). The same might hold true of improvisers in jazz and other oral traditions, such as Indian classical music, which call essentially for improvisation, though the auditory imagination must work much quicker here, since improvisers play or perform what they imagine, or some variant thereof, soon after imagining it, leaving aside what has been imagined prior to the commencement of the improvisatory performance.
Be that as it may, it seems appropriate to begin an inquiry into music and imagination with at least a brief discussion of the nature of imagination. We imagine things in a variety of ways, not all of which are highly conscious or foregrounded (Ryle 1949: ch. 8; Walton 1990; Kieran and Lopes 2003). What follows is a short, non-exhaustive list of different kinds of imaginings.
Imaginings often involve visualizing some thing or event or scene that is not present, as when one tries to picture an ice-cream cone. But imaginings can also involve forming mental images associated with senses besides sight, such as forming an auditory image of the distinctive timbre of a trumpet.
Forming mental images, however, is not the only way of imagining things. Additionally, imaginings can involve fancying or supposing something such as when we are asked to imagine or suppose the denial of a certain proposition at the outset of a reductio ad absurdum proof. And imaginings can include pretending to oneself or make-believe, something children often engage in when they play games such as imagining that a tree stump is a bear, or imagining that a block of wood is a truck (Walton 1990: 21–4). Imaginings can also involve entertaining possibilities without actually believing or affirming them, such as when we are asked to imagine Louis XIV is the King of France today, or Lincoln is the US President. Sometimes imaginings can be delusions, such as when a deranged person imagines she is Queen Victoria. Dreaming and daydreaming are also instances of imagining, ones which clearly show we need not always notice that we are engaged in certain kinds of imaginings; we do not always notice that we are daydreaming, but sometimes merely lapse into it, and we rarely if ever realize we are dreaming.
This last fact about dreaming and daydreaming points to something important about the nature of imagination and various kinds of imaginings. Imaginings can be voluntary, that is, under our control, but they can also be spontaneous, non-deliberate, passive rather than intended. They can be constant or they can be intermittent, of a long or a short duration. And one may imagine something without being aware that one is doing so. We can also be engaged in imaginings while caught up with other activities, such as the daydreaming many students do while in class. In what follows, it is important that the reader bears in mind that imagination is not always highly foregrounded and we can engage in certain kinds of imaginings without being aware of doing so.
As we will see below, the particular notion of imaginative perception (or imaginative hearing in the specific case of music) is applied to the experience of music by some thinkers (such as Roger Scruton). So it seems appropriate to clarify here before proceeding further what imaginative perception is, generally, and how it is different from related phenomena such as perceptual imaginings. To do so, we must briefly look at recent work on imagination in general before turning to music.
A fair bit of recent work on imagination by philosophers and psychologists has focused on engaging with fiction and fictional characters empathetically (Currie 2004: 173–88) or, relatedly, on recreating others’ mental states and perspectives (Currie and Ravenscroft 2002), or else on the imaginings of children (Harris 2000). Let us briefly look, as a recent sophisticated example, at the view of imagination provided by Brian O’Shaughnessy (2002: 339–78). O’Shaughnessy identifies several varieties of imagining, but let us restrict ourselves to what he has to say about three sorts of non-propositional, direct-object imaginative experiences: (i) imaginative perceptions, as when we look suitably at and “see three dimensions in” a two-dimensional photograph; (ii) will-susceptible perceptual imaginings, as in the case of common mental imagery; and (iii) will-impervious perceptual imaginings, as in the case of visual hallucinations. Note that while O’Shaughnessy’s examples are visual, we will see auditory analogs of these later in this chapter (in the section on musical expressiveness).
In discussing imaginative perceptions, O’Shaughnessy tells us that these are imaginative non-imaginings where the imagination helps generate the internal object of the perceptual experience, that is, what we see. For instance, when seeing a photograph of a landscape, the imagination imposes a second-order interpretation upon our first-order experience of seeing colored expanses on cardboard. We see these in such a way that while remaining expanses of color, they simultaneously bring a landscape into view in a special imaginative sense. There is, O’Shaughnessy claims, one complex phenomenon here with two internal objects – the colored expanses and the landscape – the latter being dependent upon the former; put differently, there is one complex experience here involving two mental representations – one of the colored expanses and one of the landscape. Thus the phenomenon is fundamentally a seeing (an imaginative seeing, that is) rather than an imagining. Moreover, O’Shaughnessy claims it is vital that the colored expanses on the photographic surface share some similarity with the landscape (as seen from a point of view), such as their common contours and color-distributions. This combination of some similarity and yet some dissimilarity prompts the imagination to this imaginative seeing, though the imagination need not follow the prompt. O’Shaughnessy claims that a landscape is visible in these marks to those who know the look of landscapes and can also impose second-order imaginative interpretations upon suitable marks on surfaces. The landscape that “appears” to one is not really there; all that is literally there is marks on cardboard, which one sees, even as one also goes beyond them simultaneously in imaginatively seeing a landscape.
Turning to O’Shaughnessy’s discussion of perceptual imaginings, which is focused on visual imaginings, we are told that there are three kinds of visual imaginings: mental imagery, visual hallucination, and “dream seeing” (about which O’Shaughnessy does not have much to say). Mental imagery comes in many varieties, of which but one is the common “seeing in the mind’s eye.” Mental images can be conjured into and out of existence at will, but they often come and go unbidden, such as sexual images, to use an example from Colin McGinn (2004: 14). They are will-susceptible in that even though their arrival may sometimes be unbidden, we bear a limited degree of responsibility for willing their persistence and their course. In contrast, we are usually without choice in the case of both visual hallucinations and “dream-seeing,” both of which also involve some measure of weakening of one’s sense of reality. Visual hallucinations can be experienced with belief (e.g. Macbeth’s hallucination of Banquo), with doubt (e.g. Macbeth’s hallucination of a dagger), or with the knowledge that they are illusory (e.g. the first stages of mescaline intoxication). On O’Shaughnessy’s view, visual hallucinations and perceptual imaginings generally are imaginings rather than perceptions or seeings. An alcoholic’s “seeing” pink elephants, for example, is a visual imagining; it is an apparent visual experience that is the seeing of nothing rather than a real visual experience with a real presence in the visual field (as when we see pink elephants in a picture).
With this overview of imagination in hand, I turn now to various ways in which imagination has been said to play a role in our engagement with music, from our basic perception of music to the construction of musical culture.
There is an ongoing debate over whether music perception is (ineliminably) informed by spatial concepts applied metaphorically or imaginatively to sound. Roger Scruton claims that metaphors involve a deliberate transfer of a term or concept from a central context to something known not to exemplify it. In this way metaphors bring dissimilar things together in a highly imaginative fusion (Scruton 1997: 80–96). Metaphors are indispensable, holds Scruton, when how the world seems depends upon our imagination being actively involved with it, and this is the case with musical experience. In describing music, Scruton suggests, metaphors cannot be eliminated, for they define the intentional object of musical experience. For example, sounds do not literally rise and fall, but we often hear music move in this way. Moreover, Scruton claims, musical motion and other musical qualities are aspects or tertiary qualities (which, following Locke, are powers of objects to affect other objects, such as the power of fire to melt wax). These musical qualities, Scruton holds, are only perceived by rational beings via certain exercises of the imagination involving the metaphorical transfer of concepts from other contexts, and so we hear music under indispensable metaphorical descriptions. In hearing sounds, Scruton suggests, we may thus be on the listen-out for imaginative perceptions, hearing sounds and also simultaneously hearing the life and movement in them that is music, situated in an imagined space and organized in terms of such spatial concepts as “up” and “down,” “high” and “low,” “rising” and “falling,” and so on.
Malcolm Budd believes an alternative to Scruton’s account of the experience of hearing sounds as music can be offered that does without metaphors and the spatial and other concepts Scruton appeals to (Budd 2003: 211). Budd suggests that one can hear the distinctive timbral character of a note without appealing to a metaphorical description transferred from another domain (Budd 2003: 213–14). Turning next to pitch and melody, Budd rejects as untenable Scruton’s claim that without reference to space, tones would no longer be heard as moving away from or toward each other. Continuing to chords, Budd argues that if melody cannot tenably be explained in terms of sounds being heard under spatial concepts, as Scruton thinks, then it seems unwarranted that we hear tones sounding simultaneously (as chords) in terms of tones heard imaginatively as arranged spatially. Finally, as for rhythm, given that Scruton here bases his view on beat as being comparable to the heartbeat, Budd claims that the idea should not be one involving spatial movement but rather of something contracting and dilating, as in the case of the systole and the diastole.
Budd’s own positive suggestions on these matters are as follows. Arguing that the literal/metaphorical distinction may obscure things, Budd refrains from claiming (like Stephen Davies (1994: 235–6)) that it is literally true that melodies move up and down. He suggests instead that melodic movement from tone to tone is merely temporal, not spatial, given that relations between tones are due to their positions on the pitch continuum, which is not itself a spatial dimension. “Movement,” Budd thinks, does not only mean change in spatial location, but can also mean change along a non-spatial continuum or with reference to a particular variable (Budd 2003: 219–20). As for rhythm, to hear it, Budd claims, may involve imagining the pulsations of life (Budd 2003: 221).
Scruton responds to Budd’s criticisms first by trying to clarify what it means for an experience to “involve” a metaphor (Scruton 2004: 185–6). While admitting that we may be up against a sort of bedrock in this dispute, he suggests that seeing a dog, for instance, involves the concept of a dog applied in judgment, whereas seeing a dog in a picture involves the concept of a dog applied in an “unasserted thought” and thus figuratively. It is in a manner similar to the latter, claims Scruton, that we apply the concept of movement to pitches in hearing a melody, since pitches cannot literally move. Scruton also disagrees with Budd’s claim that spatial metaphors can be dispensed with in hearing music, and claims in opposition that we must hear music in terms of up and down, toward and away, mirroring, inversion, forward, backward, same direction, and so on, to make sense of it. Finally, Scruton contends that Budd’s suggestion that musical movement is temporal rather than spatial is itself metaphorical, and is the same metaphor of movement that Scruton is trying to explicate (Scruton 2004: 187). Scruton grants that merely temporal Gestalts may be broken down preconceptually into temporal chunks experienced as unified wholes without appeal to movement, but thinks that this level lies below the experience of music.
In this debate, Budd seems right to object to Scruton with regard to timbres and musical movement. For, contra Scruton, the distinctiveness of a timbre might be heard under very different metaphorical descriptions or under none at all; for example, the literally shrill timbre of an oboe holding a high note might be heard as such even by little children incapable of understanding metaphors. And one can hear melodic or musical movement without appealing to Scruton’s spatial metaphors. For example, a melody can be heard as moving from the leading note to the tonic in the familiar musicological terms of melodic tension and resolution (or melodic drive or yearning) that we literally hear in the music, or in some such terms that describe the experience without essentially referring to spatial features; musically untrained listeners unfamiliar with notions of musical space might be especially inclined to do so, or else they might hear music as moving from the “unpleasant” to the “pleasant.” For example, the supertonic and the leading note have a melodic tendency to go to the tonic, the subdominant to the mediant or dominant, and the submediant to the dominant. There are also notes of emphasis, such as the tonic in tonal music, the finalis in modal music, or the vadi (or main note) of Indian ragas. And there are notes of secondary emphasis such as the dominant in tonal music, the confinalis in modal music, or the samvadi (often a fifth higher than the vadi) of Indian ragas. Similarly, there are notes or points of melodic tension and repose. Such features might be especially important in the experience of a lot of essentially monophonic music, such as Gregorian chant and many non-Western musics, where melody is not just an important element but virtually all there is to the music, barring such things as background drones, pitched rhythmic accompaniment, and the like.
It might also be asked, against Scruton, whether ordinary language metaphors at the very least point to or suggest objective resemblances (or lack thereof in the case of negative metaphors such as “No man is an island,” “Life is not a bed of roses,” etc.) in certain respects between two or more otherwise very different things, in virtue of which we are prompted to imagine (or not imagine, in the case of negative metaphors) one thing as, or in terms of, another (Trivedi 2008). If such a resemblance-plus-imagination conception of metaphors is right, and metaphors can in principle be paraphrased, then any allegedly metaphorical description of musical motion, expressiveness, and so on, might be explained away via paraphrase in a way that involves resemblance and imagination and, contra Scruton, dispenses with the metaphor. It is also possible that musical experience may be organized by concepts that do not apply literally but might only be imagined (willy-nilly, readily, and immediately, and in ways that need not be highly foregrounded) to apply to sounds as we hear them, and in a way that need not, contra Scruton, invoke or involve metaphors at all.
To turn now to musical expressiveness, analogs of many of O’Shaughnessy’s claims about imagination discussed earlier would seem to apply well to the experience of music, especially to that of hearing musical expressiveness. In hearing absolute or purely instrumental music – music without words or an associated story or program – as sad, happy, anguished, tranquil, and so on, it is clear we are hearing something that is not literally or really true of the music, which after all is without life and consciousness, and so cannot itself have such mental states. It seems plausible, then, that music is not literally sad, happy, etc., but is rather only imagined to be so (Levinson 1996; Trivedi 2006). If that is right, then it is possible that music may be imaginatively heard as sad in a variety of ways, given that we imagine things in many ways, as outlined above, and that we may often imagine things without being aware of it. As Stephen Davies puts it, “what goes on in people’s heads as they listen attentively to music and . . . its expressive character is very varied” (Davies 2006: 190).
One of the many kinds of imagining involved when we hear music as sad may be our animating the music itself (Trivedi 2001), imaginatively projecting life and life-like qualities, including mental states, onto it and thus imagining that the music itself – not something else, such as the composer, performer, or listener, or an imagined persona in the music – is sad. Our animating the music when we hear it as sad involves imaginative perception or imaginative hearing, in something like the manner O’Shaughnessy and Scruton have in mind. We really hear musical sounds in hearing musical expressiveness, and so there is aural perception going on, fundamentally. But at the same time, there is also imagining going on as we imagine readily, immediately, and willy-nilly of these sounds, in ways we are not always conscious of, that they are sad, happy, etc. Furthermore, as in the case O’Shaughnessy deals with, what prompts imaginings in the case of musical expressiveness may be resemblances of various sorts between the music and something to do with mental states, either their typical vocal or behavioral expression or their affective feel. In the very midst of hearing musical sounds, there is thus a non-perceptual or non-audible “going beyond” as we imaginatively hear mental states in the music. In accordance with Scruton’s claim that the “literal perception and the imaginative perception can cohabit the same experience, since they do not compete” (2004: 184), we literally hear or perceive musical sounds unfolding in time and at the same time also imaginatively hear mental states in them, as part of the same experience.
O’Shaughnessy’s discussion of the distinct phenomenon of perceptual imagining might also relate well to a different way of imaginatively hearing musical sadness, etc. One kind of imagining involved in hearing music as sad, say, may be when we imagine an indefinite agent in the music, the music’s persona – someone or something, we know not what exactly – expressing its mental states via the music, its gestures, development, and so on (Levinson 1996, 2006; Robinson 2005: pt. 4). Imagining a persona may involve a kind of indeterminate mental imaging, not a visual imaging but an auditory imaging. Along with the kind of visual imaging or “seeing in the mind’s eye” that O’Shaughnessy describes, it is also possible, with the help of memory, to form mental images associated with the other senses besides sight so that one might form an auditory image of the distinctive timbre of a trumpet, an olfactory image of the smell of a rose, a gustatory image of the taste of a fine wine, or a tactile image of the prick of a cactus. To be sure, many of these mental images are faint and not very precise or determinate, which also holds for the imagined, indeterminate musical persona. Alternatively, one might view hearing musical expressiveness in terms of a persona as involving a kind of propositional imagining – that there is some agent expressing itself musically – though a possible problem here may be that propositional imagining seems to be both more determinate than and not as immediate or direct as hearing musical expressiveness in terms of a persona, which happens readily and immediately and is indeterminate; one hears the sadness in the music first – someone or something is crying or wailing in the music – and then forms the belief that the music is an agent’s expression. Moreover, as with the visual images O’Shaughnessy discusses, non-visual mental images can be conjured into and out of existence and guided at will, but they often come and go unbidden. In the particular case of imagining an indeterminate musical persona, we may form this kind of auditory image without being aware of doing so, and yet the unbidden image of a musical persona may be terminated at will after we realize we are engaged in imagining it.
In an important book, Nicholas Cook has suggested that sonata form, large-scale Schenkerian tonal structures, thematic unity, serial transformations, and other such staples of music theory are not directly audible, but are rather ways of imagining sound as music (“a repertoire of means for imagining music” (1990: 4)) that constitute musical culture – “a tradition of imagining sounds as music” (1990: 223). Contra Cook, however, many music theorists would contend that their aim is to understand how music actually works rather than merely create fictive or imaginative accounts of music that do not correspond to listeners’ auditory experience (Huron 1995). Indeed, though Cook rejects such claims, it has not infrequently been held that listeners may aurally apprehend sonata forms, serial transformations, and the like not directly but rather indirectly or subconsciously, thus contributing to coherent and unified musical experiences that may consequently please and satisfy (Réti 1961; Schoenberg 1978).
Of particular interest to our topic of music and imagination, leaving Cook’s main thesis aside, is his rich discussion of the different aspects of musical imagination. Cook recalls Jean-Paul Sartre’s example of imagining a thimble, wherein our image synthesizes within a single awareness the front and back, inside and outside of the thimble, even though in real life we would have to alternate between different viewpoints to see all of the front and back, the inside and outside of the thimble, and could not see them all wholly at the same time (Sartre 1972: 105). Analogously, Cook suggests that both musically trained and untrained listeners can imagine experiences of musical works in ways where all that is heard sequentially is integrated into a single, heightened experience that captures all features of the music, even though there is something illusory about this (Cook 1990: 89). Likewise, Cook follows Sartre’s example of imagining the Pantheon where our image is simply “many-columned” rather than one that has a determinate number of columns (Sartre 1972: 100–1), and suggests that we may similarly simply imagine the sound of Dietrich Fischer-Dieskau’s voice, say, in at least a partly generic way (imagining the mellowness of his voice, the emphasis of his articulation, etc.), without imagining the specifics of whether he sings loudly or softly, what syllable he sings, whether he sings the beginning of a note or its middle or end, and so on (Cook 1990: 90). Similarly, in trying to recall a familiar musical work, Cook claims we might form generic images of harmonic gracefulness and orchestral luxuriance rather than specific sound-images with these properties (Cook 1990: 92). All these cases, Cook claims, following Sartre, involve “the illusion of immanence,” that is, the illusion that is imagined is there before one.
Cook also suggests that a lot of imagery used by musicians in producing or playing music is kinesthetic, or even to some degree visual. For instance, imagining music as fingered a certain way, or writing in a certain fingering as imagined, is one of the ways in which musicians imagine or represent the music they play (Cook 1990: 74–85). Likewise, in trying to recall one musical work while hearing another very different and structurally incompatible musical work being played on the radio, though the work being heard interferes with auditory recall, nevertheless a skilled keyboard player might recall the other work by “playing” it on a silent keyboard, consciously focusing on the movements of her fingers, hands, and arms. Alternatively, a work might be recalled via visual imagery of its score – as when a pianist plays a work from memory and remembers what comes next by “seeing” it halfway down the next page – or visual imagery of the keyboard. The imagery of the voice can also help sometimes in imagining a musical work. For example, reading a score in a library where one cannot sing aloud and is without a piano, one might sense the virtual or even actual tensing of the throat as the vocal line hits a high note or plumbs a low note, and thereby grasp something of the melody’s expressive character. Sotto voce singing while performed by jazz musicians, the kora players of West Africa, or the great classical pianist Glenn Gould provides a similar sort of security that comes from vocal awareness. There are, then, according to Cook, many sorts of images besides the auditory in terms of which musical works may be represented or imagined – kinesthetic, visual, notational, vocal, etc. – and musicians may first analyze or deconstruct musical works in these different ways before reconstructing them as wholes. Finally, Cook suggests that a composer may conceive or imagine the basic framework of a musical work before starting to write the score. Then the composer elaborates the framework and ties together all sorts of details, just as an experienced public speaker may have the framework (the basic points, etc.) and some specific details (illustrations, jokes, etc.) of her lecture worked out in her head before elaborating the framework and tying the details together in the course of writing her lecture.
See also Analytic philosophy and music (Chapter 27), Composition (Chapter 47), Improvisation (Chapter 6), Music and language (Chapter 10), Music, philosophy, and cognitive science (Chapter 54), Music theory and philosophy (Chapter 46), Psychology of music (Chapter 55), Resemblance theories (Chapter 21), and Rhythm, melody, and harmony (Chapter 3).
Budd, M. (2003) “Musical Movement and Aesthetic Metaphors,” British Journal of Aesthetics 43: 209–23.
Cook, N. (1990) Music, Imagination, and Culture, Oxford: Clarendon Press.
Currie, G. (2004) Arts and Minds, Oxford: Clarendon Press.
Currie, G. and Ravenscroft, I. (2002) Recreative Minds, Oxford: Clarendon Press.
Davies, S. (1994) Musical Meaning and Expression, Ithaca: Cornell University Press.
—— (2006) “Artistic Expression and the Hard Case of Pure Music,” in M. Kieran (ed.) Contemporary Debates in Aesthetics and the Philosophy of Art, Oxford: Blackwell, pp. 179–91.
Harris, P. (2000) The Work of the Imagination, Malden: Blackwell.
Huron, D. (1995) Review of Cook 1990, Music Perception 12: 473–81.
Kieran, M. and Lopes, D. (eds) (2003) Imagination, Philosophy, and the Arts, New York: Routledge.
Levinson, J. (1992) “Composition, Musical,” in D. Cooper (ed.) A Companion to Aesthetics, Malden: Blackwell, pp. 82–5.
—— (1996) “Musical Expressiveness,” in The Pleasures of Aesthetics, Ithaca: Cornell University Press, pp. 90–125.
—— (2006) “Musical Expressiveness as Hearability-as-Expression,” in M. Kieran (ed.), Contemporary Debates in Aesthetics and the Philosophy of Art, Oxford: Blackwell, pp. 192–204.
McGinn, C. (2004) Mindsight, Cambridge: Harvard University Press.
O’Shaughnessy, B. (2002) Consciousness and the World, Oxford: Oxford University Press.
Réti, R. (1961) The Thematic Process in Music, New York: Faber.
Robinson, J. (2005) Deeper than Reason, Oxford: Clarendon Press.
Ryle, G. (1949) The Concept of Mind, London: Hutchinson.
Sartre, J. (1972) The Psychology of the Imagination, London: Routledge.
Schoenberg, A. (1978) Theory of Harmony, Berkeley: University of California Press.
Scruton, R. (1997) The Aesthetics of Music, New York: Oxford University Press.
—— (2004) “Musical Movement: A Reply to Budd,” British Journal of Aesthetics 44: 184–7.
Trivedi, S. (2001) “Expressiveness as a Property of the Music Itself,” Journal of Aesthetics and Art Criticism, 59: 411–20.
—— (2006) “Imagination, Music, and the Emotions,” Revue Internationale de Philosophie 60: 415–35.
—— (2008) “Metaphors and Musical Expressiveness,” in K. Stock and K. Thomson-Jones (eds) New Waves in Aesthetics, New York: Palgrave Macmillan, pp. 41–57.
Walton, K. (1990) Mimesis as Make-Believe, Cambridge: Harvard University Press.