Music and Representation

Richard Vella

This article first appeared in NMA 8 magazine.

"When Kant's trancendental ego and Hegel's Absolute Spirit are consigned to the dustbin of history, there can be no mistaking the consequent triumph of spontaneity. As a result, theory (the contemplation of truth) is replaced by interpretation (a perspectival fiction, masquerading as theory). All thinking is then hermeneutical. As Michel Foucault once expressed, `the death of interpretation is to believe that there are signs, signs which have a primary, original, real existence. . . . The life of interpretation, on the contrary, is to believe that there are only interpretations.'" (Stanley Rosen, Hermeneutics as Politics , p. 9)

In any attempt to understand music it is necessary to draw the distinction between empirical analysis and analysis via analogy and what models of analysis are being used at any particular point in time. Empirical analysis in music is an analytical method which is based on observation of fact. It attempts to draw relationships between various facts in order to pro¬duce a structural understanding of the music or sonic event. For example a particular set of notes may be the inversion of another set of notes; their connection being their intervallic order. An empirical analysis is useful in understanding the relationships between the musical parameters and how these parameters define form.

Hermeneutics addresses the problem concerning the relationship between empirical analysis and analogy. It enables the logic of one system to become analagous (representational) to the logic of another system when clearly both systems have completely different logics. In other words, how does one system of signs (in our case: music) become representational of another system of signs (numbers, ideas, programmatic, film images, texts, etc) and still retain its autonomy as an independent system? In the case of music, what processes occur such that we still call it music?


". . .Such is credulity. Not that the incrudulous person doesn't believe in anything. It's just that he doesn't believe in anything. Or he believes in one thing at a time. He believes a second thing only if it follows from the first thing. He is nearsighted and methodical, avoiding wide horizons. If two things dont fit, but you believe both of them, thinking that somewhere, hidden, there must be a third thing that connects them, that's credulity. Incredulity doesn't kill curiosity; it encourages it. . . As long as you dont believe in them, the collision of two ideas - both false - can create a pleasing interval, a kind of diabolus in musica. I had no respect for some ideas people were willing to stake their lives on, but two or three ideas that I did not respect might still make a nice melody. Or have a nice beat, and if it was jazz, all the better." (Umberto Eco, Foucault's Pendulum , p.49 -50)

In music anything can be related to anything else if one tries hard enough. By substituting one system for another, be it pitch with rhythm, style with cultural associations, direct imitation of a sound within a musical text, etc, a composer has adopted or constructed a set of assumptions which articulate a certain politic/ perspective/philosophical belief about music. Most of the time this is taken for granted and so often goes unquestioned. If unquestioned this leads to the perpetuation of half truths, myths and ignorance, even falsities. Unfortunately the limitations of space prevent me from discussing the role of myth in music as it plays a very important role in represenation.

Types of representation (the role of difference)

There are many types of representation, each different to the other due to the role `difference' plays between the `original' and the `copy'. To name just a few: imitation, mimicry, simulation, onomatopoeia, parody, metaphor, metonymy, parable, allegory, naturalism, 19th century realism, appropriation, quotation. Each one of these sounds is a different approach to representation (reality/truth). What is common to all is the fact that something is used to stand for something else. What makes them individual is the role of their difference. The `copy' of the original (model) expresses a particular perspective towards the original (model). A way of knowing/seeing/hearing. A truth, but not the truth. A form of reality. (The various approaches to sound production over last fifty years regarding `naturalism' in film and radio is an example of this type of construction).

However a contradiction is implied: something is meant to be something else and yet at the same time we know it is not. This is the basis of Theodore W. Adorno's work: how is something not what it is meant to be? His analytical method attempts to show the `what is' by `what it is not' and how the `what it is not', by substitution, becomes `what it is' (i.e. dialectical analysis). Take for example his statement: "those who consider themselves apolitical are really the most political". This means that those who do not have a political interest are the ones who are the most manipulated and in doing so become a political group to be manipulated by politics. By what `they are not' they become the opposite of `what they claim to be'. It is in the understanding of how something is constructed in itself that we are able to see how it is being used to be something else. (This is Adorno's complex theory of representation - for a more thorough investigation of Adorno's work see Mihai Spariosu, (ed.) Mimesis in Contemporary Theory .)

In music we do this all the time; How does one thing represent something else in music and yet at the same time still live up to our expectations of music? Leonard Meyer's essay "Meaning in Music and Information Theory" (in Music the Arts and Ideas , p.6) addresses this problem. He distinguishes between two types of meaning in music: designative and embodied . `Designative' meaning is when a stimulus refers to something or concept outside of itself, i.e. extra-musical meaning. `Embodied' meaning is when a stimulus refers to itself. In other words, one event brings about the occurrence of another event. Meyer later goes on to write that as long as music has an uncertainty principle we as listeners will never tire of the information being presented. In other words by placing the musical sign in a constantly changing context we continually reappraise the information so that it doesnt become redundant. Consequently, we are informed about the object and about ourselves. Such is the nature of music which is completely imitative as opposed to a music which is referential to something else and yet at the same time retains its internal musical meaning to its own discourse.

Honegger's Pacific 231 is not a program of a fast train, but rather a musical model of speed. It is not a train but an analogy using its own technical parameters. Consequently it is not about a train at all but really about musical understanding of motion. The music becomes a way to understand the reality of something else, it refers but it cannot directly be the referent. It is its own referent. In other words a presence of ambiguity (uncertainty) in our listening of the music such that it is what it is referring to and yet at the same time it is not.

Imitation, Realism and Representation

In Realism in Ninteenth Century Music Carl Dahlhaus discusses the concept of `realism' saying that it is very problematic to use the term in relation to music (chapters 1 & 2). However given that, he goes on to list various approaches to the term (chapter 3). In this list Dahlhaus refutes each of them as being a theory of realism, however their value lies in helping us categorise some of the various approaches. Again I must stress that all musical representation is completely arbitrary and is a construction by the composer or listener. Therefore, according to Dahlhaus, the belief that music is a realistic art is naive and ludricous. I have abbreviated this list and added further comments:
  1. simple imitation of (non-musical) sounds : eg. the use of `onomatopoeia' in music. '
  2. music is credited with the ability to represent spatial (physical or psychological) and temporal movement . In other words: qualities. High and low, falling and rising, fast and slow, light and dark, etc.
  3. Musical representations of speech intonations . The history of music is also the history of the voice and its representations.
    "Of the various ways in which man may have become conscious of music - by blowing, banging or scraping - the use of his own voice is the most basic. We all have some music in ourselves - pitched sounds, at least, and rhythm. Common sense and common experience suggest that the earliest urges towards music, organised sound, may accompany regular movement (dance, marching, rocking a cradle) or may arise from speech. Not from all speech, however, but from certain kinds - the heightened speech of ritual and cult, the impersonal speech of heroic narrative, perhaps the anguish of pain and the release of joy (the dirge and the Alleluia)." (John Stevens, Words and Music in the Middle Ages , p.268)
  4. Musical depiction of the emotions . The following quote by Schopenhauer best explains this approach.
    "Music, unlike all other arts, does not represent ideas or phases in the will's objectification, but rather represents the will itself with nothing intervening. This is the basic reason, moreover, why music acts directly on the will, that is on a listener's emotions, passions and affections, quickly elevating or even transforming them. . .from this symphony ( Beethoven ) all human passions and affections speak at once: joy, sorrow, love, hate, fear, hope, etc., in countless nuances, yet all as if in the abstract without any particularising: it is their pure form without matter, like a world of spirits without material."(Schopenhauer, "The World as Will and Representation," in Esthetics of Music , by Carl Dahlhaus, p.42)
  5. Etymology as a mode of thought ' referring to musical symbolism associated with words: the BACH motive in German music of the 19th century where BACH equated the notes B-A-C-B flat.
  6. Music as a mirror of the structure of the world ordered according to `proportion, number and weight' .
    "But if sounds are the materia , `number' is the forma - or . . . the ratio behind the sensus . . . In plain language, the musician did not set the words of the poem to music; he set its pattern. It was this pattern, a purely numerical structure of stanzas, lines and syllables, which preceded both the melody and the poem. The pattern had to be realised in two media - the medium of words and the medium of notes - and it did not matter in the least which was realised first." (John Stevens, Words and Music in the Middle Ages , p. 499)

In any piece of music, the above six `realisms' may all be present. They are not fixed and can only be even momentary. A musical text is like any other text: multilayered, multi referential, multi historical.


"The term `mimesis' is generally used either to describe the relation between art and nature, or more recently, reality (in turn defined in a number of ways), or to describe the relation governing the works of art themselves." (Spariosu, op. cit. p.1)

Mimesis is the expression of a `reality' by the relationship between a `model' and an `imitation', or subject and object. Instead of using the term `reality', a more appropriate term is `truth' (see Dahlhaus, p.11). According to Spariosu (above) there two types of mimesis: Platonic and Pre-Platonic. Pre-Platonic mimesis is non-imitative, coming from myth and ritual. It is about the Dionysian, the ecstatic, the being (eg: Shamanism, totemism, incantations, etc). Platonic mimesis is about the imitative imperfect being idealising the perfect, the copy striving to be like the original which is no longer accessible. Pre-Platonic mimesis is sensory; Platonic mimesis is conceptual (Apollonian).

It is easy to see why people have problems with music and meaning if both of these aspects are not taken into account at the same time. In other words music is representational and also non-representational tationaltationaltationaltationali.e. conceptual and sensory): (i) When we talk of music being representational we are referring to Platonic mimesis. (ii) music as a non-representational art is an example of Pre-Platonic mimesis. There are many examples of the Pre-Platonic mimetic approach; many of them are closely related to dance (Dionysius); it being the physical manifestation of the sonorous, the physical experience.

The Pre-Platonic esthetic has become more and more predominant in the latter half of the twentieth century. As we move from a heavily coded system of signs with specific meanings, `meaning' begins to negate itself, imploding in on itself as multiple meanings become omnipresent, producing `meaninglessness'. Consequently in this age of skepticism and cynicism the experience of the object becomes more important than its meaning.

Neoclassicism, Modernism, Postmodernism

From a musical perspective the construction of `truth' can be seen in the attitudes to `Neo-Classicism', Romanticism/Modernism, and Post-Modernism. Neo -Classicism places an authority in an a priori `truth' greater than the artist. Romanticism/Modernism holds that the artist is the `truth'(i.e. the concept of genius). Post-Modernism is cynical of all truths, arguing that there can be no such thing as truth since all truths are a construction based on ideology. Hence the reason for the existence of `multiple truths'/multiple theories. It is not necessarily about eclecticism as music by its very nature is eclectic in structure: eg. the multiple styles in Mozart's Don Giovanni or Monteverdi's Orfeo . Postmodernism in music is an open acknowledgement of the Dionysian (sensory /experiential) as well as the Apollonian (conceptual/ meaning). It claims that no one theoretical discipline is able to cope with the realities of life: its contradictions, gaps, inconsistencies, multipicities. In announcing the `meaninglessness' of signs it is at the same time obsessed with `a search for meaningfulness' hence the need to give meaning to experience.

For many writers, John Cage's work has been considered to be the apotheosis of the Pre-Platonic side of Postmodern theory in music as it is about the pleasure of the moment (the sonic). It denies a teleological (causal) approach by severing or ignoring constructed relationships. Meanings which are found are purely there because the listener has `found' them: i.e. given meaning to the events. Cage's music is about the celebration of the listener. Sound is unleashed and meant to be heard as sound existing in a Dionysian play with other sounds.) Certain quarters of Postmodernism also provide us with a language not available within the Modernist approach to analysis. For example, Modernist analysis in music does not deal with psychoanalytic models of the `body' or the `corporeal', it being only concerned with the musical matter one hears and how it is constructed. Conversely, analysis usually associated with form and structure is also very useful in a Postmodern reading of a work. Stravinsky's work may originate from a sensory base, as he says of his compositional process, but it is also structural in form.

Musical Technique and Representation

It is essential not to confuse musical technique and structure with the role of representation in music. What we define as music is based on technical considerations, eg. the perception of rhythm, register, melody, sound, contrast, tempi, phrasing, etc. This is the above mentioned `embodied meaning' from Leonard Meyer: "a stimulus or process may acquire meaning because it indicates or refers to something which is like itself in kind" (Meyer, p.6). In other words, this type of musical meaning refers to itself. Representation here is only concerned with representation of the musical process. For example: a consequent phrase in a period form only has meaning because of the presence of the antecedent phrase. (Another example are the structural relationships of the repeated high register notes in the first movement of Webern's opus 27 to the repeated low notes.)

It is in the distinction between `embodied' representation and designative `representation', and how these representations are organised within a work which makes us define a music to be within a particular esthetic although even this can sometimes be a bit tenuous if taken literally.

Music as a mirror

In the light of this brief discussion concerning the presentation of reality/truth I hope it is possible to understand the complex relationship between music and representation (mimesis). Let us call the process of represenation the sonic mirror. Some sonic mirrors are literal, e.g. onomatopoeia, others are more formal (abstract) mirrors. Some are openly `fraudulant' such as sampling, others assume a `truth' such as very expensive recording techniques. It is in the role of the `difference' that we observe the function of the mirror (i.e. the relationship between the model and the copy). Eg: the recognition of the use of word painting, irony, satire, allegory, parody, cliche, sampling, etc. Each of these `mirrors' represents a particular point of view by the composer which is combined with the musical texture in order provide a particular insight into a `truth/reality'.

". . .all those mirrors. Wherever you see a mirror - it's only human - you want to look at yourself. But here you can't. You look at the position in space where the mirror will say `You are here, and you are you,' you look, craning, twisting, but nothing works, because Lavoisier's mirrors, whether concave or convex, disappoint you, mock you. You step back, find yourself for a moment but move a little and you are lost. This catoptric theatre was contrived to take away your identity and make you feel unsure not only of yourself but also the very objects standing between you and the mirrors." (Umberto Eco, Foucault's Pendulum , p. 13)

Analyses and Commentaries

The following analyses will look at two songs from Schoenberg's Pierrot Lunaire and Kurt Weill's Canon Song from the The Threepenny Opera . They will use empirical analytical methods combined with hermeneutical approaches.

1. Serenade , no. 19 from Pierrot Lunaire

With grotesquely giant-sized bow
Pierrot draws cat-squeals from his viola.
Like a stork on one leg balanced,
He plucks a doleful pizzicato.
Out pops furious Pantaloon
Raging at the night-time virtuoso
With grotesquely giant-sized bow
Pierrot draws cat-squeals from his viola
So the player drops his fiddle
Delicately, with his skilled left hand
Grabs old baldy by the collar
And dreamily plays upon his pate
With grotesquely giant-sized bow.

Schoenberg's use of the violincello instead of the viola automatically avoids the obvious reference in the text. The whole range of the cello is used to represent the sweeping grotesqueness of his bowing. For the dark low range see bar 25 also referring to the virtuosity of Pierrot in the text at 29. Again the displacement of the word association is far enough to function more as a memory and so avoid a cheap imitation. For the cat squeals of his viola, notice from bars 30 to 33 the use of the cello's top register. Again Schoenberg does not represent cat squeals, but rather portrays, in a rapsodic three/four meter Pierrot's intensity of playing, his passion.

At bar 38 notice how the music implies a more assertive two/four metric structure as Pierrot bows his viola bow on Pantaloon's head. The two/four slapstick allusions are obvious. Notice that the word pizzicato is never accompanied by the sound of a pizzicato in the cello directly - this only occurs at bar 10, 11 and 12 establishing it in our memories for when the singer does sing the word at bar 21. The waltz introduction acts as a prelude, a musical example of what the singer is about to tell us. The music hides behind the masquerade of a cello to tell us about Pierrot's viola playing. The musical accompaniment of the `Serenade' achieves an independence to the text which refers to the text but still retains its own musical autonomy.

2. Der Mondfleck , No. 18, from Pierrot Lunaire

With a fleck of white - bright patch of moonlight
On the back of his black jacket.
Pierrot strolls about in the mild evening air
On his night-time hunt for fun and good pickings.
Suddenly something strikes him as wrong,
he checks his clothes over and sure enough finds
A fleck of white - bright patch of moonlight-
On the back of his jacket.
Damn he thinks, there's a spot of white plaster!
Rubs and rubs but cant get rid of it,
So goes on his way, his pleasure poisoned,
Rubbing and rubbing till dawn comes up -
At a fleck of white, a bright patch of moonlight!

A close examination of the music reveals two canons between: (i) the viola and cello and (ii) the piccolo and clarinet. Why has Schoenberg used canons in this song? Is it purely for reasons of generating the music. Obviously yes, but then there are many other formal processes which could have done so as well. Pierrot Lunaire is about a Pierrot who looks at the moon and sees his own reflection. The work is literally scattered with various types of canons. In a sense the comes is the reflection of the dux . One part following (reflecting) the other. Furthermore a close look at these two canons reveals that they are both retrogradable around mirror points. Canon (i) the viola and cello at bar 9, on the fourth quaver; canon (ii) piccolo and clarinet, at bar 10 on the sixth semiquaver.

The text refers to a white moon fleck on his back following him. Note the canon in the viola and cello. Here is an example of word painting in that the short-demi semi quavers are obviously referring to the irritating moonfleck. However it is not such an obvious equation because this rhythmically persistent moonfleck is continually regenerated by the use of the canonic devices explained above. Again the use of a canon seems highly appropriate in that one part is following the other: the moonfleck on his back. Note also that the mirror points for both canons occur when Pierrot checks his back. He turns around and looks back; the canon turns around and looks back. A closer look at the piano part reveals that it also is canonic. Compare bar one right hand with bar two left hand.

Pierrot Lunaire and its cultural context

Pierrot Lunaire is not only about words and music but is also a collision of diverse esthetics. It is sophistication combined with decadence. The composition is considered to be an ironic cabaret work obviously referring to the german cabaret of the period. Schoenberg's singing direction, as revolutionary as it is in the Western repertoire, is also not too far away from the cabaret singing style of the late ninteenth century. In this work Schoenberg has collided two very oppositional esthetics:
  1. the developmental formal processes of the german art music tradition, and
  2. the popular musical forms of the day found in cabaret.
Hence the work produces an irreconcilable contradiction. Popular forms rely on repetition as a main means for communicating musical ideas, however Schoenberg's own musical style was based on continual variation. To repeat something exactly was anathema to Schoenberg which is why the music from this period is so transparent, fluid and mobile. Non-repetition produces a sense of amorphousness, vagueness, indefiniteness. We must also remember that this period of composition was heavily influenced by the intellectual spirit of its times. Freud's theories on hallucination are an appropriate reference point. Schoenberg's work from this period can also be perceived as musical hallucinations; musical visions. (see opus 19, opus 16 in particular). Erwartung (op. 17) can be considered to be the apotheosis of this free association esthetic as it is about a woman searching for her absent lover:

"Past and present merge into one another as fright, longing, jealousy and exaltation, sense impressions and memory associations, pass through her stream of consciousness. . .she mistakes a tree trunk for a body" (Malcolm Macdonald, Schoenberg , p. 183)

Pierrot Lunaire is another example of this hallucinatory process. Fragmented melodies, notes freely combine (associate) with other notes, momentary ruptures, grotesque shapes and images, a sense of the formlessness, although beneath all of this lies a rigorous approach to form as seen in the above discussion on Serenade . Since Pierrot Lunaire is about a moondrunk, hallucinating Pierrot it seems appropriate that its musical surface complements the poetic text with its own hallucinatory compositional style. 

3. Song of the Heavy Canon from The Threepenny Opera by Kurt Weill.

This song is a perfect example of what Brecht calls `musical gest': the music takes an attitude towards the text and presents this attitude. The song makes a pun on the word canon as it refers to Macheath and Brown singing about a soldiers' life with canons. Here is the first dramatic twist - the music uses a canonic structure in the verse. This coincides with Dahlhaus' fourth definition of musical realism: the use of etymology. In actual fact this piece is an etymologists delight in that Macheath (head of the criminal underworld) begins to sing the dux of the canon, and is followed by Brown (chief of police), who sings the comes . The canon, which is also a period structure, as a form makes a satirical comment on the relationship between Macheath and Brown: one ( antecedent/dux ) followed by the other ( consequent/comes ).

The song begins with a foxtrot whose theme is in canon. This quickly slips by way of a cycle of fifths progression (Db-G-C-B-E) using quasi augmented sixth chords (A - D flat - E flat - G, & C - F# - B flat) to the soldiers' march. The foxtrot was considered to be a sleazy popular dance. One could even extrapolate that the very name foxtrot suggests two very cunning characters (Macheath and Brown) dancing in canon (" as cunning as a fox "). Notice how the foxtrot reenters at the end of each verse functioning as a musical comment to the soldiers' life of thuggery and murder. The song is a satirical death dance juxtaposing a hedonistic leisure activity (dancing represented by the sleazy foxtrot rhythm) with professional soldiers who kill for a living (march rhythm). Such is the bite in the words "Soldaten wohnen auf den Kanonen", (what soldiers live on is heavy canon). The opposition between the two rhythms, one syncopated the other straight downbeats, achieves a synthesis in the final chorus. Both rhythms are juxtposed as the soldiers march becomes a dance of death.

Harmonically the song juxtaposes chords based on fifths (027 type chords) with chords based on tritones (in particular 026 harmonies). Although the song uses tonally based chords (French augmented sixth type chords), they are used in a non-functional way by avoiding cadences, juxtaposing consonant chord sets with the more dissonant chords. Key sections are present however: the foxtrot sections have a D tonality; the march is in E while the chorus is basically centered around F sharp. Each section of the piece starts off relatively simply harmonically and ends with an accumulation of pitch information often presenting two sets of harmonies inconsistent with each other which is the song's harmonic dialectic: foxtrot versus march. Notice also the use of parallel perfect fifths on the words "vom Capis Couch Behar" signifying the exotic life of a soldier as they travel from country to country. The bass lines follow a relentless standard I - V - I march progession. Only in moments is it in accordance with the upper voices consequently creating a harmonic tension which rarely resolves itself. Note also the tonalities referred to by the bass line progression throughout each verse of the song: D-E-F#-(G#). Just as the rhythmic component is at its most unstable in the final chorus so also is the harmony. (Fig 1 of score: tritones and fifths in juxtaposition). The juxtaposition betwen (i) the `demoniac' tritone which symmetrically divides the octave into two halves creating harmonic instability in a tonal context, and (ii) the seductive, exotic fifths combine harmonically to represent an analogy to the soldiers life: murder plus travel.


This has been an overview and introduction to the complex world of representation (meaning) in music. What we call music in our culture is a complex interaction betwen the sensory and the representational (conceptual). This lecture has focused on the representational side of music: the construction of `meaning' even though I hope we do not forget that music is also essentially sonorous and non imitative. To make a dialectical observation in the style of Adorno: music, in being not representational of anything is capable of being representational of anything. The final part of the paper has been to demonstrate the use of musical technique and the application of analogous `meanings' to this technique. Each area discussed is worthy of a chapter in itself: eg, mythology and music, Brecht's concept of Gest and Alienation, musical techniques in representation such as orchestration, narrative techniques, the role of style, sound production and realism, parody, irony and other literary forms in music.


  1. Words and Music in The Middle Ages , John Stevens
  2. The Film Sense , Sergei Eisenstein
  3. The Work of Art in the Age of Mechanical Reproduction , Walter  Benjamin, in Illuminations .
  4. Hermeneutics and Politics , Stanley Rosen
  5. Music,the Arts , and Ideas , Leonard B. Meyer
  6. Image-Music-Text , Roland Barthes
  7. Mimesis in Contemporary Theory , Mihai Spariosu
  8. Realism in Ninteenth Century Music , Carl Dahlhaus
  9. Alternative Voices , Istvan Anhalt
  10. Postmodernism , ICA Documents
  11. Film Sound , Elizabeth Weiss & John Belton(eds)
  12. Schoenberg , Charles Rosen
  13. Schoenberg , Malcolm MacDonald
  14. Brecht on Theatre , John Willett(ed)
  15. Foucault's Pendulum , Umberto Eco
  16. Aesthetic Theory , T.W. Adorno
  17. Esthetics of Music , Carl Dahlhaus

© 2003 NMA Publications and Richard Vella. Back to NMA magazine index.