Expression and truth are traditional opposites in Western thought: expression supposedly refers to states of mind, truth to states of affairs. Expression and Truth rejects this opposition and proposes fluid new models of expression, truth, and knowledge with broad application to the humanities. These models derive from five theses that connect expression to description, cognition, the presence and absence of speech, and the conjunction of address and reply. The theses are linked by a concentration on musical expression, regarded as the ideal case of expression in general, and by fresh readings of Ludwig Wittgenstein’s scattered but important remarks about music. The result is a new conception of expression as a primary means of knowing, acting on, and forming the world.
“Recent years have seen the return of the claim that music’s power resides in its ineffability. In Expression and Truth, Lawrence Kramer presents his most elaborate response to this claim. Drawing on philosophers such as Wittgenstein and on close analyses of nineteenth-century compositions, Kramer demonstrates how music operates as a medium for articulating cultural meanings and that music matters too profoundly to be cordoned off from the kinds of critical readings typically brought to the other arts. A tour-de-force by one of musicology’s most influential thinkers.”—Susan McClary, Desire and Pleasure in Seventeenth-Century Music.
Expression and Truth On the Music of Knowledge
Wittgenstein, Music, and the Aroma of Coffee
Expression is Description.
What does music express? The question is an old chestnut, and I raise it here not because I propose to answer it in some dramatic new way, but precisely because I don't. When asked concretely what a particular piece of music expresses here or there, we usually mumble out some vague, relatively stereotyped statement, from which we customarily conclude that we really can't say what music expresses. We often follow up by saying that this inexpressible expressiveness is one of the things we like best about music. In what follows I will be defending the first half of this scenario and dismissing the second. The vague statements are all right, in much the same way that Ludwig Wittgenstein famously said that ordinary language is all right. Wittgenstein, in fact, will be my chief interlocutor here and throughout this book, a kind of duet partner, alternately playing primo and secundo.
But as for music's ineffability, for that is the issue at stake when we mumble, this notion represents an error very much like the notion that we cannot know other minds (something else that exercised Wittgenstein, who could never quite make up his own mind about it). As J. L. Austin showed in a classic essay, it all depends on what you mean by "know."1 Can I form a reasonable estimate of what someone else is thinking? Of course I can. Can I think exactly the same thought, amid the same sensations, as if I were myself the other person? Of course I can't. If I blurt out the comment that a certain musical passage is thoughtful, say, or poignant, my comment is obviously inadequate only if I am trying to reproduce in words the exact experience of listening to the music. But of course I'm not-unless, perhaps, I'm Proust, who had the advantage of writing at length about music that doesn't exist.
The question of musical expression may nowadays seem less urgent than it once did. Recent thinking on what music is "about" has concentrated more on its social than on its expressive force.2 This tendency connects with the desire for a historically grounded understanding of musical meaning and it breaks with the tradition of treating feelings, the presumed substance of expression, as universals unaffected by history. My own work promotes this tendency, which I do not propose to curb or divert. Yet the question of musical expression deserves revival, because, as I hope to show before I'm through, the stakes underlying it have significant broader implications both for and beyond the way we understand music. Only if we have an adequate understanding of musical expression can we begin to understand the wider role of music in acoustic experience and the auditory culture that draws on it.
But let's raise the stakes: only if we have an adequate understanding of musical expression can we begin to understand expression in general and the complex ties between the two terms that give this book its title: expression and truth.
The first performance of Brahms's Clarinet Quintet was held in the grand Viennese home in which Wittgenstein grew up. Brahms was a frequent visitor at the Palais Wittgenstein, as were Mahler, Clara Schumann, and Bruno Walter. Ludwig's early years were drenched in the music prized by the Viennese classical tradition. This music was second nature to him; his love of it went without saying. Perhaps too much without saying: for he rarely wrote about it. It seems likely that he kept mute because the music spoke too eloquently. It meant too much to be talked about. Wittgenstein's attitude towards it was ascetic, almost renunciatory, in keeping with the monastic discipline that ruled his personal life. Words and pictures, his constant preoccupations, did not pose the dangers, the temptations of music: the sensuous and emotional immediacy, the power over memory, the cognitive pliability. Yet Wittgenstein could not remain altogether silent about it, either. His later writings contain a scattered handful of remarks that can help light up new aspects of music as a phenomenon. This wording, it will soon appear, is not casual. The remarks are, though; they are "about" music only indirectly. What they offer is a chance, not to rethink Wittgenstein's thoughts on the topic, but to think about it from the angle of reflection that he discovered and came to embody. It's the angle, not necessarily the thoughts issued from it, which I find most valuable.
The Wittgenstein I have in mind is not a philosopher who makes hypotheses but an aphorist who disturbs our conventional habits of thought, an observer who exposes the prismatic strangeness underlying ordinary life and takes the results at face value. This is the Wittgenstein who asks in all seriousness why his right hand can't give his left hand money (PI 94) and ponders whether a stone in pain could be said to have a soul (PI 97).3 He is a thinker who affirms the authority of experience by refusing to submit experience to higher authority. He is a writer who sharpens description where others call for explanation. And he is in some ways not entirely himself: not, at any rate, the therapeutic figure who dispels false problems by appealing to the grammar of our language games but an incurably fascinated figure who yields to the countervailing enchantments of life in all its perplexity to voice a grammar of intimate estrangement, the grammar of no grammar. He disenchants metaphysics, yes, but he does it to re-enchant the world.
I don't know whether I have read or invented this Wittgenstein, whose original is so iconic and much-appropriated a figure as to make the question moot: despite the fact that Wittgenstein really existed, we have to invent him.4 But invention is also a musical form, and one that is not a bad metaphor for Wittgenstein's practice, which typically picks up a theme and subjects it to a series of variations, embellishments, inversions, and counterpoints. At times it even seems that music is Wittgenstein's intuitive model of intelligible utterance. He treats it as irreducible combination of mystery and clarity to which language should learn to lay claim: "Speech without thought and not without thought should be compared to playing a piece of music without thought and not without thought." (PI 109).5
Which is not to say, all too hastily, that we should prefer the route of thought. The two routes are alternatives, not contraries; each leads to places we might want to go. Speaking without thought may become the discovery of what has not yet been thought, what is hidden from thought or forbidden it; playing a piece of music without thought (as he often does, Wittgenstein is thinking of the piano, the center of musical life in the culture of his youth) may be the best way to tap a fine-tuned expressive instinct, to enter into the spirit of a mood, attitude, or feeling that thought might retard or subvert. We can always do our thinking afterwards.
As now: for there are questions that come crowding in here. Let's say that when we play or speak with thought, we articulate our thought; when we play or speak without thought, we express ourselves. This distinction begins to break down almost as soon as it's made; slowing the process down a little may prove to be a good idea. Ask about this in the manner of Wittgenstein: when you're playing the piece of music, say a Chopin prelude (Wittgenstein will speak of "a reflective Chopin"), how do you know you're not thinking? Don't you have to think about that? Isn't the way you play a form of thinking? And if you exclaim afterwards, "I don't know how I did that!" or something of the sort, is that speaking with thought or without? Doesn't the that! presuppose a certain conception, the very thing about your playing that satisfied you, that seemed true to you or to the piece?
Questions like these can be multiplied endlessly. It's a matter of fact, of common experience, that we take some communicative actions as articulations and some as expressions, but there's always an element of decision in doing so, and perhaps an element of pretense or fantasy. Articulation is always also performative; it has an expressive dimension. Just so, expression is always also articulate; it entails substantive claims. Austin said of speech acts that they are all both constative and performative. It's easy to say that nowadays; it's become a common speech act in its own right. The hard thing is to keep it in mind and let it affect one's practice.
As with this: there is a habitual asymmetry between articulation and expression that we should both think about and play on. It's important, not just because of what it says (articulates? expresses?) about language and music but because it deeply affects the way we experience the world.
We usually grant authority to what we think independently of what we feel about it, at least insofar as we try to honor the truth. Ideally, when we say what we think we say what we think is true. But we don't grant the same authority to what we feel (intuit, imagine, fantasize), which we regard as unlinked from the truth precisely insofar as we merely express our feeling. We require that the truth come at the feeling from the outside, or else we accept that it can't and value or devalue the feeling on its own terms, which are inevitably the lesser terms.
This is a mistake. Our expressive acts are as much (and as little) capable of describing the world truthfully as our articulations, and like the latter they also contribute to what they describe. Not that they fabricate: if they succeed, what they do is make something apparent. In making it apparent, they also make its interpretation possible. To say so is to grant them a power of descriptive realism denied to them by the blind elevation of articulation over expression. To take one position or the other-and unlike the initial opposition of articulation and expression, this either/or is for real--is to enrich our experience or impoverish it.
Experience is the key: the locale where both concepts and feelings are lived out, lived by, lived through. The asymmetry has a further dimension that comes closer to the metaphysical bias that continually devalues experience. (By this I want also to say: continually devalues reason, in the most ideal post-Enlightenment sense. But this cannot be said quite yet, except indirectly, via expression.) Both speaking and playing a piece of music are acoustic events. Their relationship to the truth is mediated by the ear, not the eye; they work by matching a fluctuating contour of pitches and rhythms, the substrate of voice and bodily activity, with ideas or feelings. Good matches make for strong meanings. They convey power and authority from the speaker or performer to the listener. But some go further.
What is said can also be written. Articulation both demands and provides the possibility of transmission; in principle the articulate is articulate permanently as long as it can find a storage medium. What is expressed cannot be written. Expression consumes itself in the moment. It can be documented and recorded but not repeated; information storage is a record of its loss.
As with the original distinction between these terms, this one begins to collapse as soon as it is made, but the exact character of its collapse is a source of great significance. Re-articulation demands re-expression. To say something over is to play it a new way, apply it to new circumstances, activate the performative element always latent in it. And although there is no such thing, exactly, as re-expression, recorded expressions can be reanimated by grasping their connection to what has been said through them, while textual forms that invite expressive realization, such as plays or musical scores, are of course always ready to be performed anew, in which case what has been expressed through them will change. Experience thrives on neither the acoustic nor the visual, the articulate nor the expressive, the permanent nor the temporary, but only on the constant spillage or collapse of each into the other. The traditional subordination of the fluid to the more rigid terms only leaves us high and dry. Thought holds the heights in imperial solitude and demands conformity to the letter of its law; experience becomes arid and solaces itself with what it takes for mirages and oases.
We can pretend otherwise. Many of us fervently believe otherwise; our time even seems to be suffering an epidemic of such belief. But the terms of the either/or will not let us go. We must choose or accept delusion. And once we choose, we have resolved nothing, merely set ourselves a task that has to be renewed constantly because it is so easy to forget in practice the basic principle that Wittgenstein expressed in the saying "Words are deeds," together with its corollary, which he left out, that doing speaks.6
So, in the spirit of the Wittgenstein I have conjured here, and what I think of as the descriptive realism that impels his incessant thought-experiments and makes itself known in quasi-musical clusters of short paragraphs with the tight weave of prose poetry, I want to consider the experience of musical expression, an experience that is both musical and, just because it is musical, more than musical. In the spirit of descriptive realism, any understanding won from this effort should be expressly regarded, not as a thesis offered in the transparency of thought, but as a venture made amid the opacities of language. We are already caught in the meshes of the word, tangled in musical metaphors just as Wittgenstein's "speech without thought," gedankenloses Sprechen, is caught between meaning careless and unreflective speech, and there is, happily, no way out.
"Soulful expression in music," says Wittgenstein, "--it can't be recognized by rules."7 How can it be recognized, then? How do we know soulfulness when we hear it? Wittgenstein says we don't have to know how we know: "If a theme, a phrase, suddenly says something to you, you don't need to be able to explain. It's suddenly just this gesture that's accessible to you." You just suddenly get it.
This is not to say that such getting is not also a giving; to recognize a meaning is always in part to endow with meaning. But all this, as Wittgenstein might say, is implied in the grammar of "getting it." The metaphor of music suddenly speaking in a way that sparks this eureka! experience helps explain the lack of any need for explanation. "The speech of music. Don't forget that a poem, although constructed in the language of communication, is not used in the language-game of communication" (Z, 28).8 Like the words of a poem, a musical phrase can "go through and through us" (uns durch und durch gehen), a sensation framed in part by the way we "let our thoughts roam this way and that in the familiar surroundings of the words" (Z, 28).9 The enhanced understanding does not correspond to the familiarity, but to the roaming. The familiarity relieves us of the need to explain (I don't have to explain how I understand this sentence) while the roaming takes the familiar outside itself, expands into the not-yet known.
Of course to have this experience at all I must know how to listen or play, just as I must know the language of the poem. If, need aside, an explanation of soulful expression were still required, its basis would be acculturation: "Someone who is brought up in a certain culture--then reacts to music in such-and-such a way, to him you can teach the use of the words `expressive playing.'" Such a person would know how to imagine an expressive performance of the music, or be able to describe the expressivity of a particular performance, or have the capacity to reflect on the expressive content in an informed way. For such a person--and any listener must be such a person--the music has no identity, hardly has an existence, apart from the possibility of its being played or heard expressively. Expression, which is to some degree coextensive with the "language game" of music, requires extension, otherness, a going-beyond (what we construe as) the notes.
Music is mute. It not only does not speak, but cannot be made to speak. (So we're told, anyway; perhaps it is not impossible, only frowned upon). This concept is a historical consequence of the rise to preeminence of instrumental concert music in post-Enlightenment Europe. In a sense it is a false universal, since most of the world's music was and is vocal music. But once the idea of a speechless music, music as speechlessness incarnate, is historically established, once the possibility has been discovered and elaborated, music acquires the power to reduce any and all sung words to mere phonetic substance, even if the listener still understands them, which is by no means always the case. Language dissolves into the music itself the moment there is such a thing. The old canard that music expresses what words cannot finds no serious resistance from vocal music: rather the reverse. To give music back its lost voice is to deny and defy the regime of music's muteness, the regime that strikes it dumb.
Music has the force of verbal expression or description without the substance. It is like a seemingly grammatical sentence that says nothing: "Colorless green ideas sleep furiously" is the standard example. Like a figure on a staircase drawn by Escher, you can never get to where it leads. Why do we take this strange thing so much for granted? And what would happen if we didn't?
Wittgenstein wondered, too:
Couldn't we think that a man who has never known music and who comes to us and hears someone playing a reflective Chopin, that he would be convinced that this is a language and we just want to keep the meaning secret from him?
In verbal language there is a strongly musical element. (A sigh, the intonation of a question, of an announcement, of longing, all the numberless gestures [Gesten] of intonation.) (Z 29)
Wittgenstein's tone-deaf guest doesn't know how to recognize music, and just for that reason he perceives something true about it that is normally glossed over. The music is hiding something; it has a hard edge. Those who know music have learned how to treat this hiding not as a lack but a gift. For them--"us"--music reveals and conceals its meaning at the same time, so that we get an inkling of the meaning in the moment it eludes us, which is every moment. And knowing music, we know that the little semantic vortex that results is supposed to be giddy with pleasure.
Wittgenstein links this experience to the sphere of inflection rather than articulation: music flows in and out of the continuously fluctuating tonality of speech, here figured as a strongly physical as well as a strongly musical medium, one in which, or that in which, knowledge (sought in a question, found in an announcement) and desire (even the unlimited desire of Romantic longing) blend into a continuum.
But there is more to the music than this, more in it than this. Wittgenstein does not speak here about just any music, but about music by Chopin, and more particularly about reflective music by Chopin--even such music "as" Chopin, as it were a piece of him, not just of "his," in Wittgenstein's elliptical phrasing: "a reflective Chopin" [nachdenklichen Chopin]. (The phrasing plays us, plays on us, precisely because we know what it means without having to think about it.) Unlike the tone-deaf guest, those of us who know music hear Chopin when it's Chopin that someone plays. We hear the music's gestures of intonation--a question, an announcement, longing--knit together by a process of reflection that lies transparently open before us. "Chopin" is the name we give to this process in its singularity, and by giving it this name we make possible a full and open hearing that would otherwise elude us. This "Chopin" is a fabrication, but its consequence is the music's reality.
Wittgenstein's presentation of "someone" playing a reflective Chopin is remarkably close to Jane Welsh Carlyle's impression of Chopin himself doing so in 1843: "[Chopin's music] is not a specimen of art offered to the general admiration. . . It is rather a reflection of part of his soul, and a fragment of his life lavished on those who have ears to hear and a heart to understand."10 Wittgenstein's tone-deaf guest hears only the specimen of art, which, oddly enough, lacks the signature of the artist: Chopin without "Chopin." Because the tone-deaf guest does not know music by ear or by heart, because he does not, in Wittgensteinian terms, know how to do anything with or say anything about what he hears, he cannot make the leap to the fragment of life lavished on him. So he responds to the music's Cheshire-cat purr--its message with a vanishing meaning like a smile with a vanishing face--with resentment. And in a way he is right to do so. Chopin's reflective music, the paradigm of musical reflectiveness in general, really is a kind of secret language, a paradigm for the secretiveness of language in general. But the secret kept by language is always an open secret, the action of which, disclosed without disclosure, is itself a form of meaning.
Chopin's role in the history of feeling thus blends with Wittgenstein's inquiry into the lived character of language. It blends, indeed, with the feeling of that inquiry's own lived character, its own musical gesture of intonation, much as it might have done in the Wittgenstein mansion in fin-de-siËcle Vienna where someone would so often come and hear someone playing.
One of Wittgenstein's key preoccupations was with what it meant to perceive an object "as" something. The question was initially a matter of simple perception, famously illustrated by the duck-rabbit, a line drawing that could be seen as either animal but not both at once. But from simple perception the question branched out to embrace fiction, metaphor, and what might broadly be called perceived conception. What goes on, Wittgenstein wanted to know, when I perceive a chair as a fort? What about when I perceive a smile now as an invitation, now as a threat, now as true, now as false? What happens when I hear a melody as a lament?
These questions were not looking for explanations of subjective phenomena; a leading theme of Wittgenstein's later writings is the futility of just that. ("All explanation must go," he wrote [Alle Erkl‰rung muss fort, PI 47], "and only description take its place.") The questions were meant to elicit descriptions of human actions and the forms of life embodied by them. How, exactly, do I act when I perceive an object "as" something? How does the perception act on me? What do I want from the object and it from me? What happens?
Wittgenstein responds by talking about changes in "aspect." When I see a smile as a threat rather than a greeting, a different aspect of the smile comes to the fore: it is, so to speak, switched on while another aspect switches off. The change of circumstance or interpretation is realized by the "lighting up"--Aufleuchtung--of an aspect. The new aspect is not discernible in itself. We see by it, but do not see it. Nor does it correspond to a physical or material change, or even to a change in appearance. "But what is different?" Wittgenstein asks, "My impression? My point of view?-Can I say? I describe the change like a perception quite as if the object had changed before my eyes" (PI 195). There is nothing above or beyond that description, which is neither literal nor metaphorical but merely truthful. "The expression of a change of aspect is the expression of a new perception, simultaneous with the expression of an unchanged perception" (PI 196). I see nothing new in the object; I just see the object in a new way. Yet there is nothing paradoxical about this experience. It is the most familiar, most natural thing in the world. If it baffles us, that is only because we allow ourselves to forget that its remarkable qualities are part of the fabric of our ordinary experience and need to be neither explained or demystified, but only investigated.
When does that investigation lead us to music? What aspect of music does it end by lighting up?
Wittgenstein habitually identifies perceiving-as with seeing-as, as if unconsciously swayed by the general tendency in Western thought to equate knowledge with vision. Yet he does observe, even if he does not quite acknowledge, a difference between seeing-as and hearing-as. His examples suggest that seeing-as is usually instantaneous; that it usually jumps abruptly from seeing an object now this way, now that; and that what we see is usually seen as unambiguously present. His definition of aspect-change embraces all these tendencies: "I contemplate a face, suddenly I notice its similarity to another. I see that it hasn't changed; and still see it otherwise" (PI 193). Elsewhere, still more forcefully: "We don't see facial contortions and now infer (like a doctor making a diagnosis) joy, sadness, boredom. We describe a face immediately as sad, radiantly happy, bored, even when we aren't able to give another description of the facial features" (Z 41). Breaks in this visual immediacy have a threatening quality, as if they endangered the texture of phenomenality itself. The examples given are few and disconcerting: a familiar object made unrecognizable by unfamiliar lighting or position, the inability to recognize someone after a lapse or years or in a crowd (PI 197).
Hearing-as is altogether different. Wittgenstein's examples, most of them involving music, suggest that hearing-as usually involves a deferral; that it usually marks a transition from not hearing-as, that is, from nonsense to sense rather than from sense to sense; and that what is heard is only ambiguously present in--but this is my term, not his--the music's sounding presence. (In, as we used to say, "the music itself." If we still express ourselves this way, it is because this sort of "itself" is a necessary fiction, or at least one that has proved to be unavoidable. In truth there "is" no music itself, no bounded self to music, except as a moment in our passage from one other to another.11)
Let's take these things up one at a time.
1. For Wittgenstein, hearing-as in music is modeled on performing-as; musical expression is to the ear what expressive performance is to the hands (the implied medium is usually the piano). This process takes time. In principle it involves sorting through alternatives until the right one presents itself:
I have a theme repeated to me and each time played in a slower tempo. Finally I say "Now it's right," or "Now at last it's a march" or "Now at last it's a dance." (PI 206)
The right aspect lights up in an abruptly enhanced present moment: the italicized Now. The Now of seeing-as is typically much less marked, less sharply distinguished from the time before it.
2. The time before hearing-as is expressively indefinite. Before one hears the repeated theme as a march or a dance, one doesn't hear it as much of anything. Wittgenstein imagines someone deficient in seeing-as whom he calls aspect-blind; we might suggest that a period of being aspect-deaf is built in to the experience of music.
Insofar as hearing-as moves from the sound of some mute pre-intelligible substance to the music of an aspect, as if the aspect were lit up in the auditory dark, hearing-as is the Ur-form of perceiving-as. Before it is heard as something, music embodies the "it" of the phrase "it makes sense" apart from any sense it might make. This separation, however, is not absolute, which is why, luckily for us, music is never "absolute," either. Although, as Wittgenstein says, one can hear a musical phrase apart from the special feeling or sentiment (Empfindung) that it gives us, what one hears in that case is not yet music. One can't hear the music itself until one hears it as something else. Only from the perspective of that "as" can one even speak of the music itself. This is not a purely conceptual formulation, but a description of what the experience is like. Where visual uncertainty brings anxiety, the uncertainty of the music itself brings pleasure, or, more precisely, repeats the pleasure that follows it. The music itself is always that which will have become the music-as.
Conversely, the music-as is that which may have been, and may yet become, the "it" of the music itself. When, as sometimes happens, I hear music as something right off, the possibility of losing the -as remains as part of the horizon of my listening. The music may even seek to install that loss the better to induce the pleasure of music in between the it- and the -as, and the richer to make the -as when it comes to me and I to it. This primordial play of pronouns has a meaning and a music of its own.
3. The content of musical expression is neither in the music nor outside of it. Unlike seeing-as, hearing-as does not give its outcome a place. If I see a face as sad, I know how to locate the sadness in the person whose face it is. If I hear a melody as sad, the sadness "belongs" only to the melody. "Sad" is simply what the melody is. Yet there is no lack here, not even an absence; although we have no term for it, the non-locality of musical expression is a positive quality. It becomes even more so when one goes beyond blunt terms like "sad" and enters the arena of modern musical hermeneutics, where one might hear a melody, or whatever, as, say, nostalgic for the utopia of a form that its own construction renders ideologically remote.
Wittgenstein plays out the non-locality of musical expression through the ambiguities of the loaded word Empfindung, which can mean "sensation," "feeling," "perception," and "sentiment," and in the case of the last carries longstanding connotations of refinement and emotional quickness:
Just think of the expression: "I heard a lamenting [klagende] melody"! And now the question: "Did he hear the lament?"
And if I answer: "No, he didn't hear it; he only got a feeling of it" [er empfindet es]--what does that accomplish? One certainly can't pretend even once to name a sense organ for this "feeling" [Man kann ja nicht einmal ein Sinnesorgan dieser "Empfindung" angeben].
Some might now reply: "Of course I hear it!"-- Some: "I don't really hear it."
But conceptual distinctions can be established. (PI 209)
The last statement notwithstanding, no conceptual distinctions are established, either here or elsewhere. On the contrary, it is the futility of such distinctions, rendered here in the Janus-faced identity of hearing and not hearing, that gives us the feeling, die Empfindung, of hearing-as.
Music in the realm of Western listening must be described before it can be interpreted: in that respect it is like everything else. But there is also a respect in which music must be described even before it can be heard. It can be heard intelligibly only "under a description" in the philosophical sense of the term.12 (A "description," in this special sense, is the answer to the question "What are you doing?" The question, of course, does not actually have to be put into words, but it does have to be understood. It makes a difference whether I am snapping my fingers to call a dog, express impatience, or insult someone.) Apart from such descriptions, whether potential or actual, there is no "work" of music, no music itself, though there may be musical sounds in abundance. Music must always be reconstituted in order to be constituted in the first place. The case is suggestively similar with dreams, which have to be put into textual form not just in order to be interpreted, but in order to be remembered. Like unvoiced dreams, the music itself can only show itself and vanish.
The very things that are supposed to be the objects of description can suddenly turn out to be the means of description. Here is Wittgenstein noticing how metaphors of understanding pervade the common forms of musical description: "If I say, e.g., it's as if a conclusion were being drawn here [in this music], or as if here something were confirmed, or as if this were an answer to something earlier,--my understanding just so presupposes a familiarity with conclusions, confirmations, answers" (Z, 31).13 Strip away the "as if" clauses and the metaphors appear in their musical guises, forms that disguise their own metaphorical character: cadence, key, counterpoint. The statements that Wittgenstein invokes are speech acts that not only report but also enact understanding, and their relationship to the objects they describe is completely reversible. That is, I could perfectly well understand a passage of prose by saying, "It's as if a cadence came here, or as if a new key had been found."
Perhaps this is why Wittgenstein, reworking his figures of understanding, elsewhere treats understanding speech and understanding melody as essentially the same activity:
Understanding a sentence in speech is more closely related to understanding a theme in music than one might suspect. What I mean, though, is this: that the understanding of a spoken sentence lies nearer than one thinks to what one ordinarily terms the understanding of a musical theme. Why should dynamics and tempo move in just this line? One would like to say: "Because I know what it all means." But what does it mean? I wouldn't know how to say. To "explain" I could only compare it with something else that had the same rhythm (I mean the same line). (One says: "Don't you see, it's as if a conclusion were being drawn," or "That's like a parenthesis," etc.) (PI 143)14
The meaning comes from anywhere but the sentence or the theme. It comes from a rhythm, a line of action traced in time, that becomes fully perceptible only when I make, that is, perform, a certain kind of description, a persuasive comparison. The sentence and the theme mean precisely nothing. In fact, these supposedly solid and familiar entities may not even make sense as phenomena. How exactly does one experience the sentence apart from its use, a theme apart from its expression? Why do we even want to, or think we ought to?
Wittgenstein points to one answer in a suggestive passage that draws an implicit parallel between the sound of music and something supposedly more humble, but no less hard to grasp in its very definiteness:
Describe the aroma of coffee!--Why doesn't it work? Are the words for it lacking? And for what are they lacking?--But how do we get the idea that such a description must after all be possible? Have you ever felt the lack of such a description? Have you tried to describe the aroma and not succeeded?
(I would like to say: "These notes say something glorious, but I don't know what. These notes are a powerful gesture, but I can't set something explanatory by their side." A grave nod. James: "The words are lacking." Why don't we supply them, then? What would have to be the case for us to do it?) (PI, 159)15
The difference between the music and the coffee lies in that somewhat satirical grave nod. The questions about the coffee are again rhetorical: because we don't have a standard language game for describing the aroma of coffee, merely one for remarking on it, few of us have ever felt the lack of such a description. But there is a language game for describing music, or rather a cluster of such games, and they do sometimes fail us. We might want to say, in a Wittgensteinian spirit, that there is an ordinary language of musical description that usually suffices us: "It's as if a conclusion were being drawn," "That way of playing it is expressive, seelenvoll." But there is also an extraordinary language that constantly invokes the trope of its own inadequacy, that registers the failure of the performative comparisons on which our ordinary musical understanding rests. This language typically takes away in one breath what it gives in another: "These notes say something glorious but I don't know what." Music regarded as ineffable is the byproduct of this second language game, and we value it, perhaps, because precisely this kind of failure of words is a traditional trope for the experience of the sublime, the revelatory, the transcendental. We act this out when, accosted by perceptual excess, we sigh or catch our breath, murmur or cry out-when we call on the auditory to make our wonder audible. There is, we might want to say, a language game for the performance of the transcendental in which music plays a central role.
This game, moreover, is not only played in, or with, the present. Historical acts of musical understanding often look misguided or irresponsibly "subjective" to later interpreters, but reveal their rationality when placed in historical context. What were once successful ordinary descriptions evolve into failed extraordinary ones. Most of Wittgenstein's own descriptions seem old-fashioned at the distance of half a century. Even something as simple, as seemingly neutral, as calling a passage "glorious" can sound a little embarrassing in the ironic, pastiche-laden postmodern world. Most descriptions, most interpretations of any kind, can expect at best to be destined to this kind of obsolescence. But they do not on that account become invalid or illegitimate, nor stand exposed as mere semantic barnacles encrusting the pristine surface of the music itself. The process of understanding evoked by Wittgenstein involves a sort of situational body heat based on the urgencies of time, place, and circumstance. When the performative urgency cools, meaning hardens and shrinks, as if turning to stone. But it can be brought back to life in a vivid enough retrospect.
Consider the constant nineteenth-century habit of assigning emotions expressed in music to the person of the composer, especially if the emotions are painful. The animating context is the double one of the era's inability to conceive of subjectivity apart from an actual subject, except in the uncanny form of the automaton, and the idealization of suffering as a sign of spiritual distinction. Hence Heine's remarks on Meyerbeer: "How is it that an artist . . . destined far more than any mortal artist for good fortune--how is it that he has nonetheless felt that enormous pain that sobs and sighs out to us from his music? For what he has not felt himself, the musician cannot express so powerfully, so shockingly . . . . The artist is that child of whom the fairy tale relates that its tears become pure pearls."16 For Heine, pain cannot be expressed powerfully unless it is sincere, so it is safe to assume its sincerity on the basis of its power. The same expressive power and power of expression is the magical means by which the pain is embodied and revalued, by which tears become pearls. We know, all appearances to the contrary, that Meyerbeer must have suffered enormously because the music itself says so by the sobbing and sighing that suddenly speaks to Heine. What makes such past commentary as his viable is not only our ability to recover the historical moment of its viability, but its real relationship to the music, which it understands--correctly--as courting description as a person might court love or understanding.
During his tour of England and Scotland in 1848, deeply depressed and correctly convinced that his own death was near, Chopin repeatedly played "a funeral march"--presumably the famous March from his Op. 35 Sonata--when asked to perform in private.17 Evidently he needed to hear something in this music, needed to hear it as something: perhaps as the survival of his sensibility after his body's death, perhaps as the tragic dignity of his passing, or perhaps as something darker and more tortured, the self-consuming substance of his depression. On one occasion, whatever it was he heard become appallingly clear, though only to him. Or so we're told: the incident is reported by a collector of Chopiniana with a good reputation, but the original source, a letter, is supposed to have been lost.18 But the story makes sense even if it is only a fiction about Chopin, which in a way it remains even if it really happened.
This is what Chopin is said to have said: "I had played the Allegro and the Scherzo [of the Bb Minor Sonata] more or less correctly. I was about to attack the March when suddenly I saw arising from the body of my piano those cursed creatures which had appeared to me one lugubrious night [in Majorca]. I had to leave for one instant to pull myself together, after which I continued without saying anything."19 Chopin had been desperately ill in Majorca some ten years before and described the small "cell" where he slept as "shaped like a tall coffin."20 The "cursed creatures" he encountered again as revenants seem to be harbingers of death, bearers of portent like the banshee or incubus. As they arise, they identify the body of the piano with Chopin's own body. It is "my piano," he says, and whose body could be more identified with the piano than his? Whereupon the creatures turn these bodies into vessels of death, the coffins, graves, crypts of a gothic churchyard. Displacing the music but also expressing its spirit, the cursed creatures intrude a terrified silence on the music that is doubly mortal for Chopin, the silence of his death as both man and musician, the silence, the horror, of an ear that can no longer hear, a hand that can no longer play.
Chopin hears the music as what these cursed creatures embody; the creatures are tropes, figures, allegories, gargoyles, for what the music becomes--not expresses, but becomes--in this hearing. But the shape of this becoming, in the form of the creatures, is separated from the music, whose onset the creatures block. On this one occasion, Chopin hesitated to release the power of hearing-as; he acted to protect the music from becoming the advancing march of his death. But in so doing he lost something he did not yet know he had, the power of the music to protect him from the very creatures whose silence it reversed. To hear the music as his death was bad; not to hear it as his death was worse. Just to hear the music itself, without the death it harbored, was to hear nothing, a sound more void than silence. So Chopin withdrew a moment to pull himself together: to compose himself, so to speak. And this in turn enabled him to consign his own voice to the music, both in the immediate present and the posthumous future, so that he could keep silent about what had happened to him: so that he could silence the silence.
Or did he just tell himself not to be ridiculous, to forget about things that go bump in the night and just go back and play the music "more or less correctly"? If so, was he coming to his senses or denying them? Would it even have been possible for him, above all, to hear this music, above all, without hearing it as something burdened with mortality? At what point in this whole fantastic scene could Chopin have gotten (his) death off his hands?
Musical expression resists description. So we're told. It's commonplace to say that no verbal formula, no matter how rich, can fully represent what even the simplest piece of music expresses. The statement is not wrong, but the wrong conclusion--that the truly musical value remains unaccounted for, a secret that can only be lost to the imperial force of language--is consistently drawn from it. Music presents no resistance to language that is not already fully formed in language itself.
The immediate object of any interpretation is always a description of the proposed object rather than that object itself. The description thus has the difficult task of opening up interpretive possibilities without predetermining their outcome. No merely "objective" description, in the usual sense of the term--empirical, impartial, conceptually closed--can do this. A description that claims to stick just to the facts, and postulates its own adequacy in this fidelity, is completely unable to formulate meaning and instead actually distances and neutralizes its object. Only a description that grants its own approximateness, its own figurative, conjectural, constructive character, can hope to energize the object and bring it rewardingly close.
From a hermeneutic standpoint-and this can't be said too strongly or too often-- it is precisely the semantic gap between interpretation and the object interpreted that is constitutive of meaning. That gap must be preserved, not closed, in order to speak effectively of the artwork, or for that matter of any event or circumstance with a meaning in question. The act of description required by the object's inability to speak for itself is an encounter with otherness in the most positive sense of the term. The risk of exceeding the knowledge-protocols available to one as a historical subject--that is, the risk of being "subjective" in the sense of eccentric, too "personal"--is a necessary condition of effective description.
What would happen if we examined the commonplace about words and music according to the Wittgensteinian mandate: if we didn't take it for granted, but regarded it as something remarkable that no interpretation can fully represent a piece of music?
Imagine an interpretation that does tell us something we can hear in a piece, something that we want to hear, even as we always, necessarily, hear more. Let it be something simple, say that a piece by Chopin is reflective; it sounds like thinking, like a meditation or reverie. To the extent that the description "applies," i.e., to the extent that we accept it as a rough template through which to listen, the music simply becomes an embodied thinking, a reverie in the mind without the mind. We hear it differently than we would otherwise-if, for example, we heard it not in terms of "thoughtful" but of "thoughtless" expression. Any work of music can be heard under any number of descriptions of this kind, and offer in return any number of different aspects. Music does this with a distinct sensuous immediacy and a seductive, "touching" intimacy which set it apart, and perhaps are responsible for the idea that musical expression can be grasped only in musical experience. But this is an illusion, however pleasing it may be. The aspect-logic of music, its pliability under informal, relatively non-specific description, is anything but unique. Virtually everything that figures in our world beyond the level of sheer sensation, everything that goes into the making of what Wittgenstein called a form of life, behaves in exactly the same way. We may listen to our music as we drink our coffee, but the expressiveness of the one is not that same as the aroma of the other.
During Chopin's lifetime and for many years after his death in 1849, both he and his music were consistently described in terms suggesting disembodiment: they were ethereal, angelic, elfin, sylphlike, fairy forms.21 They seemed to occupy a border zone between spirit and matter to which they alone had access. Implicit in this way of hearing Chopin is a recognition that music outside the sphere of his exceptionality, music in general, is weighty with a bodily being that it may embrace or seek to surmount but can never forget for long. Another aspect of the Chopin legend, the emphasis on his frailty and sickliness, allied to a morbid sensibility, may bespeak the burden of that memory. But the burden creeps in even where Chopin's body is at its most diaphanous:
As frail as he was in body, was he delicate in style; a bit more, and he evaporated into the impalpable and imperceptible. . . . Only the divine pen that described the fantastic retinue of the dream fairy [Queen Mab, i.e, Shakespeare] could analyze the complicated, infinite, and yet light-as-lace tangle of that phrase charged with notes, in the folds of which the composer always enveloped his ideas.22
What is most striking about this obituary tribute is its slippage from the vaporous and impalpable to the soft materiality of tangled cloth, of lace and other "stuffs" as such fabrics are called, charged, weighed down, with acoustic matter, in the rich, delicate yet sensuous folds. We no longer listen to Chopin under this description. It would embarrass us to do so. But we could.
What do we speak about when we speak about music? Statements about expressive content are usually referred to the subjectivity of the speaker; statements about form or structure are usually referred to the music. The assumptions that underlie this distinction are effectively the same as those underlying the model of language associated with logical positivism in the early twentieth century. This is the very model or "picture" of the relation between language and truth said by the later Wittgenstein to have held his younger self captive (PI 48). The distinction is based on the ease with which we can verify the statements, and thus on the degree of certainty with which they can be credited. The expressive statements have the inferior status here; they are regarded as mere "pseudo-statements" about the music itself. The analytic statements are stronger stuff, or so the old picture would have us believe.23
Analytic statements may not give us the whole truth about music, which after all has an expressive content even if we can't specify it, but they are thought to give us the only truth available. In short, we or may not agree with Wagner's claim that the first theme of Beethoven's Coriolan Overture expresses the hero's untamable force and arrogance, but we cannot deny that this theme is recapitulated at the subdominant.24
The positivist view of language is, of course, long gone, but as applied to music it has great staying power; it can still seem like simple common sense when not stated too baldly. What disappears at the level of principle returns at the level of practice. Most musical analysts will now readily admit (or at least admit when pressed) that statements about form or structure do not represent music positivistically, "as it really is." Instead such statements constitute descriptions relative to an implicit or explicit, formal or informal theory of musical articulation. The statements are true to the theory more than to the music itself. Yet the positivist ideal is continually reinstated by the assumption that statements about form or structure are more fundamental and more reliable than statements about expressive content. Analytical descriptions are still widely taken to have priority over hermeneutic ones. Analytic statements are taken as propositions that specify musical realities to which all hermeneutic descriptions have to make both reference and deference. The recapitulation at the subdominant still trumps the force and arrogance.
But there is no warrant for this priority. It is no more than the effect of a longstanding language game that enwraps little truths--this theme is recapitulated at the subdominant--with the transcendental aura of a big truth. If both analytic and hermeneutic statements are forms of description, the priority of one over the other is a matter of practice, not principle. For E. T. A. Hoffmann in 1813, priority went to expression, not to structure, and it did so in a lopsided way: "That composer alone has mastered the secrets of harmony who knows how, by their means, to work upon the human soul; for him, numerical proportions, which to the dull grammarian are no more than cold, lifeless problems in arithmetic, become magical compounds from which to conjure up a magic world."25 The underlying metaphor here is alchemy, not logic; it treats structure not as the basis of expression but as the base metal from which expression is sublimated.
In short, it is no more true that the first theme of the Coriolan Overture recapitulates at the subdominant than that the theme represents the hero's force and arrogance. It is not made more true by the fact--if it is a fact--that explicit criteria are available for recognizing a subdominant but not for recognizing force and arrogance. In Wittgensteinian terms, such criteria are part of the language-game for specifying harmony, but not of the language-game for specifying expressiveness. The usual "positivizing" suppositions about such criteria, moreover, do not always hold up. Explicit criteria are not always reliable or unambiguous, nor is their absence always a source of dubious judgments. It is easy to imagine an analytic argument to the effect that the recapitulation of the Coriolan Overture is not "really" subdominant, but it is hard to imagine not hearing something like force and arrogance in the Overture's first theme. Either the force and arrogance are as much a part of the music as the subdominant recapitulation, or the music itself lies somewhere outside either one.
Both sides of this either/or are perhaps equally right. On one hand, the music can be understood as the product rather than the object of musical descriptions. The act of description specifies a form of perception and in so doing conjures up the "picture" of an object to be perceived. This is not to say, of course, that all descriptions are equally good. They manifestly are not. The first theme of the Coriolan Overture is not pastoral and its recapitulation is not at the flat supertonic. The fact that we can only hear under a description does not mean that we can hear under just any description. But what matters about a particular description is not its type, but its effect, its power to animate listening with pleasure and knowledge. For most listeners, the statements that do this best are more expressive than formal. Musical aesthetics has for too long labored under the illusion that such descriptions are the least faithful to the music itself.
On the other hand, no description, no matter how effective, has an exclusive claim on the music itself. For again: it is the very act of listening under a description that tends to produce an intuition of the music itself, a music that soars beyond the grasp of every particular description but not beyond the apprehension of the listener. This, too, is an illusion, but only in the affirmative sense of a trick of the senses--trompe d'oeill, trompe d'oreille: a perception one can't help having despite one's conceptual knowledge that it is false.
Wittgenstein's writings constantly try to mimic the rise and fall of intonation. His texts are full of exclamations, questions, and imaginary utterances, and especially of italicized words and phrases. They are full, too, of deictics, "shifters," of this and that turned into instruments of thought. This is both philosophy by intonation and the philosophy of intonation. It is thinking with the musicality of thought.
Is this just a matter of sensibility or style, or is there some deeper reason?
There is. If writing is, as it is traditionally understood to be, a sign for speech, then italics and a few punctuation marks-a small but hardy band of visual signals-are signs for intonation. What the italics and so on signify is a change of voice, which the reader must somehow hear even if, as is most likely, the reading is done silently. With Wittgenstein in particular this is quite literally true, since he tended to dictate his typescripts. The words that would be printed in italics would have to be underlined, so that the typist would perform the equivalent of the vocal emphasis by going over the same spaces twice, once with letters, once with the repeated underline. The speech emphasis would translate immediately as the repeated keystroke.
This intonational script of written language is often suppressed in "good" writing. This point of style, enforced by a small army of teachers and editors, is the symptom of a bias that it also helps produce. It positions language on the side of reason rather than mere feeling; it objectifies the linguistic act regardless of the act's content. It silences the vocal dimension of language on behalf of a generic voice of authority.
It is precisely this voice that Wittgenstein sets out to dismantle. His power to do so no doubt stems primarily from his critical undoing of its pretensions, but this work of critique would be incomplete, perhaps even impossible, without the language that voices it, the language that philosophizes expressively about the philosophy of expression. It is no accident that this mode of thinking repeatedly leads Wittgenstein to imagine situations in which whatever happens, be it trivial or momentous, is dependent on the expressiveness of speech. No expression, no event.
From which we conclude what?
First, that the effect of expression is as much cognitive as it is affective. Expression may sometimes follow or supplement the understanding of an utterance or situation, but it is sometimes the precondition for understanding either or both. Cognition without expression makes as little sense as expression without cognition.
Second, expression is essentially an auditory phenomenon. It occurs on the threshold between voice and speech. What appears as expression in any other register or medium is a lived metaphor for a change of intonation. Language, even spoken language, is not so much expressive in itself as enunciated in the medium of expression. If, as Austin would claim after Wittgenstein, the meaning of any utterance is bound to its context, and so may change completely as the context changes, then the expressive line of the utterance, its intonational movement, is the first element in any and every context. Expression is, as we might "express" it, the context of context, the lived, embodied principle of contextuality.
Musical expression becomes particularly significant in this connection. It does so in part because the "music" of an utterance is something between a synonym and a metaphor for its expressiveness. In part, though, and perhaps the larger part, music, in expressing one thing or another, also always expresses the condition of possibility for expression itself. Music is expression in its most concentrated form. We might almost say that the phrase "musical expression" is a tautology. Expression is just that, and it is both remarkable and perfectly ordinary that we always know what that is, even if we have some trouble saying so. When we understand what music expresses, we dissolve it into our comprehension. Musical meaning diffuses itself through us like milk through coffee.
To know music in any genuine sense, we must become adept at moving between expression and the unexpressed. Although Wittgenstein is not explicit on the topic, it would appear that he simply identifies expression with the possibility of a certain type of inexact but irresistible verbalization. Music is paradigmatic of the expressive because expressiveness occurs when we sense, and in particular when we hear, something being said to us without being able to say what. We may hear it in the tones of music or in a tone of voice; we may read it between the lines in the tone of a text or in the tone of our own voices when we are moved to speak. In response, we clumsily say whatever comes to our lips; we make a description and reject it in the same breath. The content of the statement matters less than the existence of this language game, which informs our sense of a significant world or, better, informs our world with an "atmosphere"-Wittgenstein's word-of significance, evokes the sense of sense itself.
But Wittgenstein's own language may be taken to imply exactly the opposite. This atmosphere is not the shroud of expression but its medium. Our descriptions are not the end and substance of expression, but the means by which we elicit certain aspects from the music and in that very act endow the music with a consistency that stands behind and supports the face, the aspect, that we present to it. This consistency is not a metaphysical phantom, but a concrete product of our listening practice, and one that manifests itself to us in a sensuous material form.
The movement between expression and the unexpressed corresponds to the sinuous flow (think here of the serpentine shape identified in the eighteenth century as the line of beauty) of conceptual transformation, of symbolic substitution, and, in the classic Lacanian formulation, of desire. This flow is perhaps best conceived as a temporal equivalent to the space of hybridity described by Homi Bhabha as the core condition of 21st-century cultural formation. The description draws out the implications of a liminal symbolic space that is also an everyday object: "The hither and thither of a stairwell, the temporal passage that it allows, prevents identities at either end of it from settling into primordial polarities. This interstitial passage between fixed identifications opens up the possibility of a cultural hybridity that entertains difference without an assumed or imposed hierarchy."26 The stairwell winds between levels, identities, disciplines; motion along it, and under it, behind stairs, affords a continuous shifting of perspective and orientation.
As a model for thought, the stairwell involves a continuous transformation of descriptive terms into metaphors, speech acts, texts that interpret, texts to be interpreted. In following that model, I have sought (and will seek) to emulate Wittgenstein without imitating him: to use his oddly angled observations as a means of reconceiving musical expression in ways that help both clarify and dissolve its mundane force; to draw on the procedures of defamiliarization, lapidary writing, and thought-experiment to inform and sustain a mode of inquiry based on the work and pleasure of thinking between, enjoying between, living between. For it is there, in the interspaces and the interludes, that we discover the space of our own animation, the space of the lifelikeness of life and its knowledges.
The cognitive form of the result, referred to earlier under the name of "descriptive realism," is the ultimate issue here, especially, but not exclusively, as it is reached-perhaps reached best--by the proximate issue, the experience of music. At stake is an entry in the ongoing debate about realist versus constructivist modes of understanding that is perhaps the dominant trope of knowledge in the academic world today and the primary source of the (in)famous conflict of the faculties. More broadly, the stakes are those of struggles over the status and sources of truth that seem increasingly to pervade social and political life on a global scale, but with a complexity that could only remain penumbral here. What I have sought to propose, via music, via Wittgenstein, is a small piece of an unaccomplished whole, a Wittgensteinian language-game, if you like. The speech-acts involved have the aim, not of producing realistic descriptions in the familiar empirical sense, but of recognizing how descriptions help produce realities that are, nonetheless, independent of every particular description. Such realities are fully capable of serving as points of reference against which the interpretive claims of the descriptions can also be understood as what J¸rgen Habermas calls validity claims.27 One such reality is what we call musical expression.
Descriptive realism is the understanding that description is what gives us access to reality, that meaning arises where and when this happens, and that meaning is not something affixed to a prior reality but something that suffuses and reshapes a reality of which it is a part. The world as we inhabit it has "intonations" like those that Wittgenstein focuses on in speech, and one mode of response to those intonations-which we partly hear, partly overhear, partly project from ourselves-is to sing, whistle, hum, play, improvise, compose. Musical expression answers the expression of a world, in much the same way that, according to Wittgenstein, to understand a sentence is to understand a form of life. One way to symbolize or paraphrase the understanding that results would be to say that meaning is the music of reality.
What would music be if we could hear it (not just hear the sounds, but hear the music) without hearing it as anything? Would we even recognize it as music, and, even if we did, would we enjoy it or be repelled by it? This question is more like a question about understanding a sentence in language than one might suspect. Is it possible for me to hear a sentence in a language I understand without understanding the sentence? If so, what would I hear? One answer would be: babble, meaningless noise, made doubly senseless by the latent presence of a sense I must actively suppress because, in truth, I understand it at the same time as I do not. I would hear the debris of an expression that could not take place. What I would not hear is the sentence, which has no existence apart from my hearing it as a sentence. The sentence is not an acoustic substance to which a meaning is affixed: no sense, no sentence.
Music follows this logic in reverse, and it does so repeatedly, in a perceptual process that is also a cultural ritual. As Wittgenstein's account suggests, hearing-as in music is often deferred, even if only a little: no meaning, no music, but the former must often catch up with the latter and we sometimes take pleasure in drawing out the delay. Before long, however, perception demands an aspect. We hear music as something, as expressing something, because there is no other way to keep hearing it. Music is not the rabbit hole down which meaning falls and disappears. It is not a mysterious substrate to which meaning is, always falsely or inadequately, affixed. Music is a sensuous form in which meaning is both inchoate and immanent. It is something that will always have been something more; it is the threshold of that becoming other on which it is also pleasant to linger. One way to define music is precisely as meaning in its material form.
"I think," writes Wittgenstein, "of quite a short phrase, only two bars. You say `There's really a lot in there!' But it is only so to speak an optical illusion if you think that what is in there goes on while you listen. (`It all depends who says it.') (Only in the stream of thought and life do words have meaning.) (Z, 31).28 So the phrase--two bars, two words--has meaning only in relation to a stream that flows through and away from it, and thereby permits the passing illusion that one can hear the phrase itself.
The perception that something is "in there," that there's really a lot in there, is the imaginary form of a readiness to make educated guesses about how such a phrase might be treated, what might be done with it, how one might describe what comes of it. It all depends on who is speaking, which for Wittgenstein means on what kind of language games the speaker is ready and able to play. Such games, such speakers, fall into a finite repertoire, even a fairly small one. (I am always conscious, these days, of how small the "classical music" game-my game-has become, which, of course, only increases my determination to play it.) After observing that soulful expression in music can't be recognized by rules, Wittgenstein wonders aloud, "Why can't we imagine it might be, by other beings?" Perhaps the best answer is that any such beings would be tone-deaf by definition. Only guessing-games will work here. The guesses called for do not actually need to be made. The unacted potential of making them is part of what keeps us absorbed when we listen. But the moment a guess is ventured, however offhandedly, a moment that always comes: the moment you detect expressive performance or the lack of it, the moment you represent what you've heard, even just as glorious, the music lights up. It gives us the light we hear by.