Jean-Paul RIOPELLE, L'hommage à Rosa Luxemburg, 1992 (détail)[ * ]

Musical Meaning: Reference and Symbol
Michael Woods
                  

I wish to present an argument which will establish not merely the intelligibility of the concept ‘meaning’ as it applies to music, but to argue that we can talk, without idiosyncrasy, of musical meaning and say in what that meaning consists. It will be conceded that meaning is something which is characteristically although, by no means uniquely, attributable to language, more specifically, to sentences in a language. Further, according to Michael Dummett,[1] “general agreement” seems to have been reached that the meaning of sentences consists in their truth conditions. And although music has often been held to be a language, and even to have a logic of its own, we would be reluctant to argue that music is a semantic system whose meaning consists in its truth conditions.

Perhaps that might count as a very good reason for arguing that music is not, therefore, a language, and I shall return to this point presently. But we can talk without embarrassment of understanding, or of not understanding, some musical works; and because I believe that to understand anything in this sense is to understand its meaning, we can legitimately talk of musical meaning and in what it consists. I would like to say that there is a conceptual and therefore significant aspect to musical apprehension, but not of the kind that has been traditionally presented.

This does not entail any semantic relation with truth conditions for, assuming that truth conditions themselves entail some notion of correspondence with entities in the world or with facts or states of affairs, music, being non-referential in any essential sense, cannot have meaning and truth.

Meaning and truth, at least as far as music is concerned, are separable; and even though music is sometimes said to have or reveal or express truth, it can hardly be said to assert it. And if, as I shall argue, music has meaning and therefore value, it does not express any truth which may be semantically construed.

In this way, I shall argue that although music is not a language, because no specifiable unit of a musical composition can be said to be declarative or even referential, it can still be said to have a meaning. And “meaning” is not being used in any attenuated sense, even less in any idiosyncratic sense; rather it is used with the same force in which we talk about the meaning of a given meaningful sentence, which has an intention, but no denotation.

I say this because meaning functions in a variety of ways, the range of which is broadly reflected in St. Augustine’s distinction between natural and non-natural, or designative, meanings. Meaning functions linguistically in a designative way whereby someone intends an utterance or sentence to convey the meaning he wants it to convey. Meaning functions culturally in this designative sense whereby, for example, a given symbol comes to have the meaning that it has in virtue of the historical circumstances in which it has arisen (the cross, for example, the crescent, or the hammer and sickle). Examples of natural meaning might be expressed in such locutions as “that red sky means good weather” or “He’s here tonight – that means trouble!” But the essential question concerning anything claimed to have a meaning is not whether it does or does not have meaning as such, but in what way its meaning functions.

Meaning functions in varying ways: dreams, headaches, nothingness, a gibbous moon, a two-fingered gesture, all mean something, but in ways not necessarily assimilable to sense/reference distinctions, those common to discussions of linguistic meaning.

Perhaps the problem may be viewed in this light: Schumann (in some accounts it was Beethoven, suggesting that the story is apocryphal, but no matter) was reportedly asked after the performance of one of his compositions what the piece meant; he responded by sitting down and playing it again. The question is: what are the ways in which the music may have been said to mean something? The anecdote suggests two widely varying interpretations between which polarities the controversy has continued for centuries. Namely, that music has a meaning which is constituted in its having a significance beyond itself; or alternatively, the meaning of music is immanent to it. Schumann (or Beethoven) was implicitly expressing the latter view by demonstrating that the meaning of the piece lay in the piece itself and was not to be revealed in a discursive explanation of its supposed provenance in some realm of extra-musical reality.

My “use,” in the following argument, of the accounts of Deryck Cooke and Susanne Langer is intended more as a discursive procedure than a critique of their theories. I am aware that there have been more recent contributions to the aesthetics of music, specifically Peter Kivy, whom I shadow here when he describes musical descriptions as being “either nonsense, or subjective reverie,” or Roger Scruton, who, for example, describes the music of Philip Glass as “nothing but figures … endless daisy chains.”[2] The central question is whether musical meaning is expressible and therefore describable in terms of what it, itself, is not; that is the problem I address here. If music can be said to express a truth, does it lie outside of itself or is it an interminable succession of tonal daisy chains without any significance?[3]

This “extra-musical reality” is usually identified by philosophers as the realm of feelings and emotions. The relationship between music and the life of feelings or emotions has been defined in a number of ways, but among the most significant of these are Deryck Cooke’s contention that music is a language of the emotions, and Susanne Langer’s theory that music is a symbol of the emotional life. I shall briefly examine a particular aspect of the claims of each, namely that there is an identifiable abstraction of the musical experience, whether of composer, listener or performer, which can be broadly regarded as the correlate of this experience and the source of its meaning.

Cooke’s The Language of Music[4] seeks to establish a referential, and hence semantic, relation between melodic shapes and other features of musical expression, and the specific feelings they are said to represent over five centuries of Western music. From settings of a range of texts by many composers, he derives a supposedly conventional meaning, attached to musical phrases in virtue of their more-or-less fixed correspondence with particular emotions. Each distinguishable unit of a musical work, he argues, has an emotional meaning corresponding with the composer’s, making music, therefore, a language consisting of separable expressions, each with conventionally fixed verbal connotations. As he says, “Music is a language, not just in a vague, general sense, but in the detailed sense that we can identify idioms and draw up a list of meanings.” Or again: “Music is, properly speaking, a language of the emotion, akin to speech.” [5]

On the basis of an examination of what he calls “the elements of musical expression” (the thirds, the sixths, the sevenths and so on), Cooke argues that there are sixteen basic melodic shapes which are inherently and consistently related to some emotion: the descending 5-3-1 minor progression is said to refer to “a passive falling away from the joy of life”; the ascending 1-3-5 major corresponds to outgoing emotion, an active, assertive emotion of joy as is expressed, for example, in Strauss’s “Blue Danube Waltz”.[6] The sixteen musical elements are said to have the following characteristics: obsessive, outward-looking, assertive, despairing, anguished, reassuring, joyful, sorrowful, happy, consoling, gloomy, inward-looking, exuberant, innocent, painful and passive.

I will not go into the question of whether the emotions listed here are separable or even distinguishable, or whether a principle for distinguishing them can be established. And following from this, I will leave open the question of whether we can say that a musical expression can be said to specify a feeling of despair as opposed to anguish, or gloom, or pain; or whether joy or happiness or exuberance are being specified. But an empirical claim that such specifiable emotions correspond with, and are therefore specifiable in virtue of, musical expressions will not be readily accepted, mainly because emotions are in no way separably identifiable objects of musical expression, although they themselves may be said to have objects. Claiming fixed emotional referents for the musical expressions with which they are supposed to correspond is a position which, I shall argue, is untenable.

The claim that every musical expression has a fixed emotional referent suggests that an interpretation of a piece that fails to identify correctly the “intended” emotional referents fails for this reason. Disregarding the fact that any performance is ipso facto a unique interpretation of a musical work, Cooke points to the ideal interpretation, i.e., an interpretation in which the intentions of the composer are fully recognized by the listener.

Presumably, disagreement about musical interpretation will be resolved by an analysis of his work (assuming, in the first place, that that reveals his intentions in this sense), and by the correct identification of the emotions it supposedly expresses. The notion of musical interpretation, therefore, as an evaluative “reading” of a piece, becomes supererogatory; for now we are left with the only valid interpretation, one which turns out not to be an interpretation at all, but the correct exposition. The meaning of a piece of music can thus be established empirically so that logically speaking, and, in effect, a musical composition is independent of its hearer, a claim similar to the notion that colour is independent of sight.

Moreover, such an analysis would make serious demands on the capacity of the listener to ascertain (correctly) the intentions of the composer and the state of mind he resolved to express. The ‘form’ of the music has now become almost an irrelevance, a crypto-grammatical code, which can be unlocked by reference to a musicological dictionary of the emotions. In this way, musical meaning becomes synonymous with discursive exegesis, so that any music, which does not extend beyond itself, and is therefore exhausted in its own particularity, is without (extensive) significance.

Any music which is not susceptible to Cooke’s standard associative procedure, including the greater part of contemporary and perhaps non-Western music, will thus have to be judged as having no meaning and as valueless. Musical works, which do not conventionally refer to fixed emotions, do not fit his frame of reference, and are not, therefore, examples of musical language, but just tonal pretensions, reducible to a colligation of sounds. If they don’t have any reference, or none, at least, that he can recognize, and if we admit this characterization, then we will have to accept the conclusion that they are not music, in much the same way that if one defines painting as a representational art, then almost all of contemporary and modern art since Matisse is not painting. But it is obvious that this is not defining in an essential sense but defining away. And the problem lies with defining music in the first place. The identification of what are deemed to be the essential characteristics of musical meaning in a nexus of discrete emotions—each (sixteen) of them being linguistically specifiable―raises further questions concerning reference and identity. Are the referents of such expressions as ‘anguish’ or ‘despair’ uniquely distinguishable particulars being brought under empirical conceptual rules, which, in turn, are being identified by discrete musical phrases or tones, and which enable us to apprehend these particulars? And so on. The possibility of identifying such particulars would presumably enable us to enumerate them, so that we can count them as being neither more nor less than sixteen in number.

I don’t think such a characterization, and following from this, the enumeration of feelings, accords either with our experience of affective life as a continuum, as a series of discrete states, or with our experience of music as an expression or embodiment of that life. As well as everything else, definition is also a procedure of exclusion and Cooke appears to presume that the scope of the life of feeling, which is supposedly being identified in music, is precisely as inclusive and exhaustive as he describes it, as well as being separate from the being who lives it. This would prohibit the possibility of his musical syntax expanding or refining or developing the life of feeling through invention or juxtaposition, or reshaping the world of tonality so that, in effect, musical form is petrified in its referential sameness. It is precisely in its discomposing of this emotionally calcified sameness that music finds its significance. As Nietzsche said of art, it “ … continuously muddles the rubrics and the compartments of concepts, presenting new transcriptions, metaphors and metonymies: it continuously reveals the desire to give the subsisting world of waking man a figure so multicolored, irregular, devoid of consequences, incoherent, exciting and eternally new …” [7]

Because we could not seriously accept the consequences which flow from the argument that only languages have meaning, that music is a language and thus has meaning (one of the consequences being the absurdity, as I have suggested, that most contemporary music is not music or that the development of musical form is impossible), we will have to call into question the claim that only languages have meaning and that music is a language in that denotative sense. The argument is not that some music is meaningless but that, if it has a meaning, then this meaning does not inhere in some conventional referential relation with linguistically specifiable emotions. Cooke implies that if music meant nothing beyond itself, it would be meaningless, incapable of expressing anything. But this no more follows than saying that a mathematical equation is meaningless because it refers to nothing beyond itself or that the terms of a mathematical equation must be capable of being linguistically described or, indeed, to put another converse slant on it, that stretches of discourse would be meaningless because they are not musically expressible.

But we do not understand music as we understand a language or at least language as a means of correspondence, principally because understanding a language involves understanding the conditions under which a (meaningful) proposition in that language is true. Music, or a musical expression, cannot be declarative, or propositional in that sense, and there is nothing in music which has any correspondence-to-fact relation which language is sometimes said to have and in virtue of which its propositions are true.  

In claiming for music a referential capacity, indeed in claiming that music, in order to be music, must have a referential capacity, Cooke implies a dual relation between music and what it refers to, its form vis-à-vis its content.  Further, Cooke’s   argument that music derives its significance from what it refers to suggests that the emotions assume a logical priority over their means of expression; this fact indicates that music is merely a means of reproducing or evoking an emotional experience which gives it its value. The experience of music, therefore, is reduced to something else so that our reason for listening to music will be that it enables us to experience a particular emotion, or a series of discrete emotions. And the difficulty with this is that if the point of listening to music is to attend to the emotions thereby set off by association, and we are then in some external way using the music to evoke these emotions upon which we are now concentrating (assuming that were possible at all), then we are no longer hearing it in the sense of attending to it, which is to say, listening to it, a disposition one would normally construe as being the indispensable condition for musical understanding in the first place.

If there are serious difficulties, as I think there are, in the notion of music as a language, perhaps it can more plausibly be said to be a symbol of emotional life where the referential relation, if there is one, is not conventional, but morphological. Music, in other words, is not a system of expressions which refer to the emotions, but a symbolic projection of emotional life itself. This is the position which has been held by Susanne Langer in a number of writings.

For Langer, all art, and therefore music, is a symbol of human feeling, and a symbol is “any device whereby we are enabled to make an abstraction”.[8] It is a vehicle for the conception of the object in abstracto.  Similarly, C. G. Jung writes: “By a symbol I do not mean an allegory or a sign, but an image that describes in the best possible way the dimly discerned nature of the spirit. A symbol does not define or explain; it points beyond itself.” [9]

A necessary condition for something to be a presentational symbol of something else is that it has a similar structure, and this structure is projected onto something in accordance with a rule, thus making it a symbol. In this way, for example, a picture may be a symbol of what it represents insofar as it leads the viewer to conceive of what it represents in virtue of its possessing an arrangement of elements analogous to, or as a projection of, the salient visual elements in the object represented by it. It is the same with maps, blueprints and musical scores.

Langer describes two kinds of symbol: discursive and presentational. Language is a discursive symbolic system which has a vocabulary and a syntax out of which can be constructed, according to syntactical rules, new units with relatively fixed meaning, carrying with them the conditions for the possibility of truth. We have seen some of the difficulties in the theory that music is a language, difficulties which Langer principally avoids because for her, music is not a discursive, but a presentational or non-linguistic symbol. This is to be distinguished from the symbolism of language inasmuch as it does not symbolize by means of discrete, fixed units of meaning, as with Cooke. Rather, the elements of a presentational symbol are understood only through the meaning of the whole symbol.

Thus a musical composition, indeed any artwork, is an expressive form and what that form expresses is human feeling. A form, for Langer, is a symbol of what it represents: an articulation, a unity resulting from the relation of mutually dependent constituents, projected from something other than the symbol, namely feeling, and endowed with the same structure. Music, as a symbol of feeling, has this relationship, according to Langer. Its function is not, as with Cooke, to express the composer’s knowledge of emotional life, but to provide the listener with an insight into the life of feeling itself.

Following from this, the structure of music is analogous to, or morphological with, the structure of the life of feeling which it projects. The dynamic structure of a musical work (its mode of development) and the ‘form’ in which the feeling is experienced can resemble each other in their “patterns of motion and rest, of tension and release, of agreement and disagreement, preparation, fulfillment, excitation, sudden change, etc.”[10]

Music is, in this way, a presentational symbol of emotional life. If Cooke maintained that the meaning of a piece of music is linguistically specifiable, Langer argues that its meaning lies, not in whether its discrete parts are specifiable in this way, but in its capacity to present or show the life of feeling in general for our contemplation. Such a form is the symbol which makes this life of feeling apprehensible: it is precisely the meaning of the music with which it is identical and which she describes as a “wordlessly presented conception of what life feels like.”[11] Feeling is less like what the artist feels than the ‘felt world’ which is presented.

We have seen a number of difficulties in the conception of music as a referential system, i.e., something whose meaning is derived extraneously or, in other words, from outside the music itself. The question now is whether the theory of music as an art symbol avoids these difficulties, and it seems that, at least at first sight, it cannot, for it is in the nature of a symbol to refer, or as Jung says, to “point beyond itself.” Langer attempts to avoid this implication:

The Art Symbol … is the expressive form. It is not a symbol in the full familiar sense, nor does it convey something beyond itself. Therefore it cannot strictly be said to have a meaning; what it does have is import … it formulates and objectifies experience for intellectual perception, or intuition, but it does not abstract a concept for discursive thought. [12]

For Langer, therefore, music does not point us to a meaning beyond its own presence: what is expressed cannot be grasped apart from the form that expresses it. Nevertheless, she proceeds to designate the Art Symbol as an analogue:

The tonal structures we call ‘music’ bear a close logical similarity to the forms of human feeling – forms of growth and attenuation, flowing and slowing, conflict and resolution, speed, arrest, terrific excitement, calm, or subtle activation and dreamy lapses – not joy and sorrow, perhaps, but the poignancy of either and both – the greatness and brevity and eternal passing of everything vitally felt. Such is the pattern, or logical form, of sentience; and the pattern of music is that same form worked out in pure, measured sound and silence. Music is a tonal analogue of emotive life.[13]

Some decisive questions enter into Langer’s interpolations and equations. First, there is the notion of a musical composition as a “wordless conception”. It is unclear whether this refers to the music itself or our response to it. Nevertheless, everything in it is to be understood in such a way that the music is presenting a form of life wordlessly and not in such a way that something, apart from this, is being thought or something else is being said. Musical meaning does not have extension, but nevertheless provides for us a concept (the symbol or form of feeling) for our contemplation. Its objectivity, symbolically represented, lies in its properly understood contemplative apprehension.

The tonal structure of the musical piece is said to “bear a close logical similarity to the forms of human feeling,”[14] and the question is whether the life of feeling can be said to have anything so definite or recognizable as a ‘form’ or ‘structure’ at all. Is it to be construed as a particular aspect of the life of the human being, and, if it is, how does it arise in the first place, and is its presentation in a work of art the only way in which this form or structure can be articulated? Is it being suggested that the work of art, as a unity, gives active formation to an inchoate manifold of feelings or sensations and symbolically transforms them into a unified and dynamic whole which “resembles” the formal structure of the composition? Is this symbiosis adventitious or is it more fundamentally determined so that there is a basic and essential connectedness between them? Finally, where is the living human being, the composer, listener, performer, in this musically-induced apprehension of the sentient form of life?

The question here is whether the Art Symbol is referential or not, and it is difficult to avoid the implication, in spite of Langer’s efforts to get around it, that music is being defined as something other than itself: the symbol articulates that which is not itself. Even with the special definition which Langer provides for it, “symbol” logically separates the source of meaning from its expression and that meaning is (explicitly) a concept, not, or, at least, not primarily, something felt or perceived. It is difficult to reconcile the dichotomy in which Langer is caught, namely, that the artwork, and in this case music, is self-referential and, at the same time, symbolizes, or gives the appearance of actual felt life, however that might be considered as being a part of a lived human life.

And even here there are further difficulties because, anxious as she is to avoid the conclusion that music is symptomatic of the feelings which the composer has, it is symbolic, a “symbolic expression of the forms of sentience as he understands them.”[15] Music expresses what he knows about inner life so that even the assertion that music bears some necessary relation to experience is replaced by the assertion that music bears some necessary relation to thought about that experience.

The same fundamental problems besetting the theory of music as language are now apparent in the theory of music as art symbol, inasmuch as we don’t experience music as referring to something other than itself; its meaning is sui generis, except in a way which is appropriated by a sentient composer, listener or performer. When we say that we determine, understand or perform a piece of music, we don’t mean that it is in virtue of some extra-musical reality to which it is supposed to refer: we mean that we have an essential relation to the disposition that we experience as sentient and intelligent participants. There is a vagueness in some apparently central notions: “feeling,” “form of life,” “symbol”. I doubt if a critique of pure reason can extend this far. In spite of the disclaimer that the “articulate but non-discursive form (has) import without conventional reference, and therefore (presents) itself not as a symbol in the ordinary sense …” (my italics)[16]—whatever the ‘ordinary sense’ may be―, the interpretation of the essence of musical meaning appears random.  A symbol, for example, cannot be symbolic of itself, but rather of something, a fortiori, which cannot be itself, and, therefore, is extraneous to it, even if it is another symbol. The possible inner connection between the musical composition, its symbolic nature, and the form of sentient life which it mimics is not exposed; it is left hovering between elements whose functions accidentally coincide, and are declared not to be functions at all, but “qualities”. Nothing appears to bring them together—there is no schematism here―except the (definitional) assertion that they are aspects of the same (conceptual) apprehension.

Perhaps Langer’s “wordless conception” provides a clue: what kind of conceptualization, if any, takes place in music? What is the nature of thinking in the musical experience?  And what is the object in which we are immersed? The “object’” and the “experience” here referred to can be construed as concepts that express the polarities within which this complex field is the subject of investigation; but these polarities are also, in general, the respective focuses of musicology and aesthetics. The first takes the composition as an empirically given object for investigation and the second takes as its point of departure an investigation of music’s capacity to express and evoke emotions, thus falling, broadly, on the experiential side. Musicology is characterized by technical analysis, aesthetics by philosophical theorizing. Although this rough dichotomy might suggest that both polarities are fixed in their isolation because of fundamentally different methodologies and expectations, I am using it not to give an account of any supposed mutual distrust, or of the details of professional academic jurisdiction, but rather to bring out their symbiotic relationship. This will lead me to give an outline of the conditions necessary for musical experience; these conditions are presupposed by musicological analyses of music as object, and epistemological analyses of the expression and evocation of emotions in music.

I shall take “music” to signify, minimally, the ordered arrangement of tones and their intervallic silences, whose meaning can be said to be embodied or immanent rather than denotative or symbolic. Tones must be distinguished from sounds, considered as physical objects, although sounds can be utilized in a musical composition, thereby becoming tones. The distinction between sounds and tones is not one which is rigid or categorical, but one which is dependent upon the musical attitude of the listener. In much the same way as an everyday functional object can become a work of art, or at least, an object of specifically aesthetic interest, by being put in an art gallery (or as in Picasso’s Bull, by being simply an arrangement of bicycle seat and handlebars), so a familiar sound can become a tone by being employed in a composition and being heard as part of the composition.

Essentially, and in its most fundamentally physical characteristics, a musical composition is a fixing of a state of auditory material through duration in terms of sonorous matter (amplitude, frequency and duration), all of which is physically calculable. This quantitative identity is a necessary but not sufficient condition for musical apprehension. The transposition of these sonic forms from acoustics into aesthetic cognition is precisely the locus of musical meaning, which is precisely the imagination.

The difference between a tone and a sound lies in the use which the hearer makes of it, not in any physical characteristics which such objects may be said to possess. Sounds are heard as signs: a buzzing around your ear in the dead of night indicates that there are mosquitoes about: a footfall on the stairs is a sign of someone present. Even an unidentified sound is heard as signifying something not yet known. But tones are not signs, nor any combination of them. They are struck in a nexus of melodic, harmonic and rhythmic retention and expectation, with the silences of intervallic meaning between them. They do not signify, although they may import, anything extraneous to themselves. If they can be said to have a meaning, it is immanent. And just as the sense of a sentence is not logically dependent upon any possible reference, so the sense (or better, meaning) of music is not dependent upon any extra-musical reality. Music is not significant in that sense.

Musical meaning is therefore not referential, but is immanent to the tones themselves; and these tones are taken up perceptively, interpretively and feelingly by the listener; they are not taken up as sounds. The point here is that, if there is musical meaning, it is non-referential. More specifically, the sound made by creaking chairs and clinking glasses are not musical in virtue of signifying the presence of physically interacting objects; and as long as I interpret them as being mere sounds in this way, I cannot hear them as harmonically, melodically or rhythmically related. This is not to suggest that such sounds cannot be so used; but as long as I associate them with the objects which produce them, they can have no musical import.

The sound of music, therefore, is taken to be music and is thus a complex situation in which an ordered arrangement of tones and silences, melodically, harmonically and rhythmically related, is experienced as an already synthesized entity. Our sensitivity, attention, even absorption, and general musical literacy will determine the intelligibility and quality of this experience through our capacity for retention and anticipation.

The trouble is there is more to it than that which is often suggested in such exaggerated descriptions of passionate involvement as “profound spiritual involvement” or “submission to a transcendent world”. Perhaps these descriptions are so exaggerated as not to be recognizably about music; but they do find their echoes in current theories (Cooke’s, for example) that music is expressive of, and referentially dependent on, the emotions. The distinction I am trying to bring out here can be illustrated by means of what Schoenberg said to the musicologist, Rudolf Kolisch, who had written an exhaustive analysis of set structure in the Third String Quartet: “You have gone to a great deal of trouble, and I don’t think I’d have had the patience to do it. But do you think anyone is better off for knowing it? … I have always been dead against: seeing how it is done; whereas I have always helped people to see: what it is!” [17]

Musicology, understood generally as the technical analysis of the musical composition as an empirically given object, tends to cover up the non-empirical, aesthetic status of music, namely, the musical experience. But in attempting to take the measure of this musical experience, philosophical theories have described this “what it is” as the referent of emotion-descriptions, or otherwise have given exaggeratedly passionate or metaphysical descriptions in attempting to account for it.

The musical experience is one which depends almost exclusively on a presentational rather than denotative element: the harmonic, melodic and dynamic structure is that which immediately, and more forcefully, affects the listener. Usually this outward structure of harmony―the quality of consonance and dissonance, its melodic-rhythmic idea—need not be understood in its full complexity; and yet, it is still possible to form an aesthetic judgement on it, one usually expressed in terms of its being sad or joyful, somber or energetic, noble or squalid. Such clichés and their application to the musical experience have a well-established tradition, apparently begun in earnest by Monteverdi, but seriously impugned by Hanslick in 1854 as having anything but a figurative significance. For him, such expressions were not to be taken as being denotative of the emotions which music was supposed to express: music was to be judged on the basis of its formal structure alone, not on its putative capacity to express or otherwise refer to anything beyond itself. Hence, an historical exemplification of the polarities which I mentioned at the beginning: on the one hand, we had those who felt the necessity to describe the (particularly affective) experience as primary, and those who wished to cleave to a formal description of the musical object.

Obviously this dichotomy is abstract and is used simply by way of bringing out the focus of two kinds of approach. In reality, it would be absurd to imagine the musical object and the musical experience as separable “things”. The distinction is nonetheless possible, for it is on the basis of the distinction that musicology embraces the formal investigation of the musical object, and musical aesthetics encompasses the philosophical investigation of the expressiveness and the effect of the musical object on the experiencing subject. This suggests a division of labour in which generalizations about the expressive and evocative powers of music are made by the philosophers, while musicologists offer a more detailed and penetrating analysis of its structure.

I am willing to concede that this, itself, is a generalization: not all musical aesthetics are crucially or centrally concerned with the capacity of music to express or evoke emotions, and most musicologists are more concerned with bio-bibliographical investigations than with critical analysis. But while the expression of emotion may not be of central concern to all musical aestheticians, it cannot be disregarded by any. Nor can an unwillingness by musicologists to practice critical analysis (perhaps on the grounds suggested by Schoenberg that it doesn’t get to what it - the music – is) be any justification for regarding it as being abstruse and without having any explanatory power. If “something is missing” (Brecht) at the end of the musicologist’s critical-analytic labours, “what it is” might be found in a closer collaboration of aesthetic theory and musical analysis.

In giving a brief outline of the necessary conditions of the musical experience, I return to Kant for assistance. This strikes me as one of the fundamental tasks of musical aesthetics, since it will ultimately have profound implications for musical judgement. Such an outline will involve a number of related problems, such as the mode of existence of the musical work and the process of establishing the inner causality which makes a musical work possible. Perhaps this philosophical, and therefore general, approach would need a critical musical analysis which would offer it the kind of concrete vindication that it obviously demands, thus giving particular justification to a general theory. But it seems to me that the guiding question in such a theory would be something like: “How is the unity of a musical composition possible?” However that question is understood, it is susceptible only of a philosophical answer, an answer which, nevertheless, can be adequately fleshed out by means of musicological analysis. I say that the question is susceptible only of a philosophical answer because musicological analysis is possible only if it is directed towards a unity that has already been determined as a unity. By “unity,” I mean unity of genre, style or of composition and its constituents, the composition being here the elemental form.

The musical composition is subject to determinative forms of unification; these are the necessary conditions of the musical experience. These determinative forms are a priori and are, therefore, not discoverable, but are rather exhibited in the musical composition. Being functions of unity, they are presupposed in all our experience of particular compositions such that the act of hearing is a synthetic act, although not merely a formally synthetic act. It also presupposes an affective or sensuous orientation towards the composition as a phenomenon. I shall attempt to explain what I mean by this.

I have suggested here that my interest in any given musical composition is not, as a musicologist’s might be, in the facts of its composition, whether historical, biographical or social, nor in the technique employed in its construction, but in the conditions which are necessary for the composition to be experienced as a composition. My interest, therefore, is not in the composition as an empirically given object, but in an experienced phenomenon, which is to say, in an appearance, whatever the empirical circumstances of its appearing may be.

The pure awareness of this appearance is not merely passive, not any more than perception itself is mere sense-perception, i.e., mere receptivity for tonal sense data. In the experience of the musical composition, original impression, retention and anticipation acquire a synthetic character of their own, in a temporal process of continual self-completion, independent of conscious activity. It is thus a precondition of analysis and evaluation. I wish to say that this precondition is time consciousness, through which musical identity and unity are established; it is also the source of the connective forms of coexistence and succession in objects of consciousness. It is the origin of what we know as “form,” for in it, entities are synthetically shaped for musical consciousness. I shall try to explain what I mean by this and venture to suggest that space and affectivity are equally fundamental in the constitution of the musical composition.

In broadly defining music as the ordered arrangement of tones and their intervallic silences, I made a distinction, at the beginning of this essay, between tones and sounds where sounds, unlike tones, are heard as signs. When we experience such an arrangement of tones as a composition, we do not experience it aurally as a discrete succession of points of sound, or if we did, we could not experience its rhythm, motion, form or direction. This means that a composition, for example, is not reducible to any supposed “basic material” of pitch frequencies, any more than a sentence is reducible to the words which go to make it up. Any theory which held music to be reducible to its pitch elements, in any case, would make it impossible to distinguish music in the sound from the sound itself, and could not account for retention, anticipation, feeling or values.

Therefore, what remains as a purely logical requirement is that any possible musical understanding must unfold temporal (and spatial) forms of some kind:  to perceive tones as musical, one requires a prior understanding of motion in time, beginning and ending, rest and tension, “toward” and “away from,” anticipation and fulfillment; one requires a form of life.

What is the “wordless conceptualization” that Langer spoke about? What is the essentially referential character of musical meaning that Cooke argued for? Is music a language?  What is good and bad music? What is musical thinking? The abstracted nature of how music works for us is often presented as an expression of an emotional life which (adventitiously) may or may not exist for some, may be thoughtless (as in tuneless whistling) or pointless amplification (as appears to be the case in contemporary rock music).

And yet, in describing music, we are of necessity driven to conceptualization: that is to say, we are compelled to say in language what it cannot say itself. We can talk of melodic completion, harmonic development, contrapuntal question and answer, the resolution of a seventh. But none of these bears any relation to how we talk. These descriptions convey, rather than describe, a non-specificity which a linguistic specificity can only approximate. Of course, we can say of a composer that “he is saying something here,” and this, without being able to say what he is saying; but now we are getting close to the unsayable. Maybe he is not saying anything at all.

“What are they doing?” you might ask of Picasso’s Saltimbanques. “It is a painting. Besides which, the acrobats are just giving the impression that they are involved in an act. As I said (I might say), it is just a painting.” Similarly, music gives the impression of being an “as-if,” so that we feel compelled to describe precisely what is supposed to be said, represented, expressed or embodied, “as if it were a real experience”.

Yet, unless we fall into silence, we feel compelled to give thought to our aesthetic experiences. Many of the difficulties which attend aesthetic theories, such as they are, are compounded by the confusion concerning precisely who is involved in the musical experience itself and what kind of thinking, if any, is taking place.

There appears to be a chasm between the experience of the music and a description of the experience, between the feeling and the thought. The experience of a performance, for example, is not merely an acoustic event, but has a cognitive finality: we know, or strive to know, what it means, such that sonorous material resolves itself into meaning both for composer and listener (and, of course, performer). At least to that extent, Cooke and Langer are right. Except that they do not distinguish musical thinking from conceptual thinking.

Any description of a musical experience presupposes a communal understanding, whatever disagreements may exist about its value, of what it is to know the form of such understanding or the nature of its agreement. Disagreements presuppose an understanding of the general nature of agreement, which is to say, dialogue, and an understanding of presumption as to procedure, acceptance and conclusion, if any. In general, we are familiar with the essential form of such discussions: so far as talking is concerned, we don’t generally embark on a voyage in which we are already lost: we know, in advance and in general terms, the trajectory and the terminus, whatever our success in achieving it.

In precisely the same way, music should not be understood as an inchoate storm of sound in a vacuum and seized upon by a naïve musical perception: it is taken up and apprehended as meaningful, the response to which is often expressed in terms of natural language. In this sense, music can be thought, but only in musical terms. This may give rise to the notion that thoughtless music is bad music. And this is because we often understand (and do not just feel) the creative mind or more particularly the human being that generates music in such a way that gives evidence of thinking in their creator. It is to this that we respond. How does this come about?

As far as the listener is concerned, musical passages may give the semblance of a thoughtful composer,  asking questions, searching for solutions, reaching conclusions in such a way that the music appears to have embarked on a reflective quest. And musical passages themselves may display or imply thought processes on the part of the composer so that we are directly and cognitively engaged in his manner of setting a scene, resolving conflict and attempting a solution. In this way, the object of cognition is the development of the composition itself; for example, we make an assessment of how a fugue theme has been devised in counterpoint, why a composer has decided on a C-sharp as opposed to a C-natural (thus setting up a tension to be realized further on in the piece) or why he has chosen to realize and sustain the greatest number of harmonic combinations within the narrowest tonal space possible.

This is the implied thinking we ascribe to the composer, and in so doing, we follow the music, its succession from chord to phrase to motive, from opening to development to conclusion. This is musical apprehension, a purposive, teleological process to which we may ascribe such terms as ‘coherence,’ ‘balance,’ and ‘intelligibility’.  It is the kind of thinking Langer might have been referring to when she spoke of the effect of music being a “wordless cognition.”

Following a piece of music in this way is to follow the manifestation of a musical mind in its progressive solution to formal problems. In the Finale of Beethoven’s Ninth Symphony, for example, we can describe the incongruity of introducing voices into a long instrumental, a decision which determines the movement’s form. We are presented with a brief, tumultuous, and dissonant introduction, followed by a review, and rejection by instrumental recitatives of, the themes of preceding movements. A brief suggestion of the “Ode to Joy” theme is followed by its exposition in four stanzas, the crescendo of which defies our expectations; we are seized by tumultuous, dissonant opening measures. Again, our grasp of coherence is defied when we are presented with a new theme with orchestra and chorus. A double fugue and a complex coda appear with matchless sublimity in a tonic A to which a strident F# has been added. This conflagration has been in preparation for over 90 measures and leaves the listener breathless; it is followed (just when the conclusion is thought to have been reached) by what seems to be a village brass band thumping out a Turkish march, resolving into a heroic solo tenor, an intricate double fugue, and the final choral statement of the principal theme.

No language adequately expresses this experience of uncertainty, wonder and sheer physicality of melodic starkness, this creation of a reflective clearing for the dissolution of (tonal) anxieties. We are persistently aware of a powerful musical intelligence at work; we also experience its evolution, which places demands on our perceptual capacities along with our emotional sympathies. This is musical thinking, a following of an event which involves retention, concentration and anticipation; a persistent forward and backward movement of thought in an attempt to seize and construct the musical “now”.

Essentially, then, musical apprehension is a thoughtful experience, one that implicitly, but not exclusively, draws on sentient life. The musical composition may be said to be the incarnation of thought where “thought” is considered to be the essence of the felt life. The knowledge gained through this experience is not propositional, and to that extent, music is not a language that is extensional and that needs to go beyond itself to root itself in meaning. Rather, the knowledge gained stems from the behavioural and experiential possibilities of a form of life which, when followed and understood, leads to a profound and articulable meaning that musical form contains within itself.

I accept that the concept or the experience of music upon which my argument is based may be considered to be too limited or even lacking in refinement: there are many other kinds of significant music which do not fall within the ambit of my considerations, as, for example, music which may be thought to be both declarative and referential. The music of the dance or something by Jimmie Hendrix or the Ramones or even advertising jingles could be problematic, so far as my argument is concerned. But the argument is essentially this: if truth is considered to be an equation between propositional expression and an (objective) state of affairs, then music is singularly untruthful and often explicitly so. And if we can speak at all about music and truth, or even untruthfulness and deception, then it must reside within the music itself. It’s as if someone were to ask on the birth of a child: “Who does she look like?” The only answer is: “She looks like herself.”

Music is its own revelation and its own (un)truth.



[1] Michael Dummet, “What is a Theory of Meaning?” Truth and Meaning: Essays in Semantics. Eds. Gareth Evans and John McDowell. Oxford: Oxford University Press, 1976.

[2] Peter Kivy, The Corded Shell: Reflections on Musical Experience. Princeton, New Jersey:  Princeton University Press, 1980.

[3] Roger Scruton, The Aesthetics of Music. Oxford: Oxford University Press, 1999.

[4] Cooke, Deryck, The Language of Music.London: Oxford University Press 1959.

[5] Ibid.

[6] Ibid.

[7] Nietzsche, Friedrich, “Ueber Wahrheit und Lüge im außermoralischen Sinne,” in Grossoktav-Ausgabe, Leipzig, 1895. Translated by J. Culler, 1981.

[8] Susanne Langer, Feeling and Form: A Theory of Art. London: Routledge and Kegan Paul, 1953.

[9] Jung, C.G., The Structure and Dynamics of the Psyche. Trans. R.F.C. Hull, 2nd edition, London: Routledge 2002, p. 45.

[10] Michael W. Lind, “Aesthetic ‘Sympathy’ and Expressive Qualities.” Ed. Michael Mitias. Aesthetic Quality and Aesthetic Experience. Amsterdam:Rodopi, 1988, p.44. 

[11]  Langer, op. cit.

[12]  Ibid.

[13]  Ibid.

[14]  Ibid.

[15]  Ibid.

[16]  Ibid.

[17] Erwin Stein, ed. Arnold Schoenberg: Letters. Trans. Eithne Wilkins and Ernst Kaiser.  New York: St. Martin’s Press, 1964. pp. 271-272.


Previous article /

Article précédent


Top / Début

Table of content / Table des matières
Next article /

Article suivant