Archive for January, 2013

It is widely accepted that listeners to abstract music written during some historical periods at least commonly make narrative interpretations of the music. They do this in order to make sense to themselves of the listening experience, or to communicate their experience or understanding of the music to others. Perhaps the listener is the true narrator of such music?

The significance of this narratisation of the listening experience is questioned on the basis that this tendency to narratise abstract music is just one of many examples of the way human beings narratise experience in order to comprehend it and communicate it. It therefore does not demonstrate the existence of narrativity in the music itself.

It is therefore suggested that the key questions on the existence of narrativity in abstract music concern the nature of the systems of signs established by the composers, and to the narrative personae which composers may adopt during the creative process.

In seeking to define the identity of the narrator in music, I would like to begin by questioning the validity and significance of listener-generated narrative interpretations of Western abstract art-music, on two grounds. One is that they amount to a potential misapplication of concepts based on written literature. The other is that such narrative interpretations are subjective impressions, and do not show whether narrativity is inherent in the music, or how it is operating.

1. Lyric or epic narratives in music?

To begin with the first objection, in Book III of The Republic, Plato (2004) distinguishes three types of poetry:

– tragedy and comedy, in which the poet speaks through characters and conceals them self;
– lyric poetry, where the poet speaks in their own voice;
– epic poetry, where the poet recounts the actions of characters.

Aristotle makes the same distinction, and regards the performance of music as equivalent to the recitation of poetry, (Aristotle 1974: 33) and therefore operating in the lyric mode.

I would argue that most narrated, diagetic, art forms, and above all the novel, are descended from the concept of epic poetry. A narrator tells of the actions, words, and thoughts of a range of characters.

Western art-music, I would argue, is descended from the Greek concept of lyric poetry. There is no evidence of other characters, and only one voice speaks. The question is the identity of this voice.

It might be argued that there are certain musical works where the composer specified a programme which refers to or implies the presence of a range of characters. Berlioz’ Symphonie Fantastique, Op, 14, is a work which might be seen as close to a novel. It has an detailed programme which refers to characters and events. However, Cone (1974) argues that:

… Belioz’s intent [in the programme] … was not to describe the scenes and incidents, but to depict his hero’s reactions to them. (Cone 1974: 83)

Cone’s (1974) interpretation of Berlioz’ Symphonie Fantastique might be read as the communication of a sequence of emotional states within a third person, the hero. If so, it could be said to be in the epic mode. However, we know from Berlioz’ own life that this ‘hero’ is in fact closely related to the composer. It would therefore be more accurate to say that the composer has adopted the narrative persona of the hero in order to write this particular piece. It is therefore in the lyric mode.

Relevant to this point is Abbate’s (1991 or 1989, 23), view of certain narrative analyses of Beethoven’s Symphony No. 3 in E flat major (Eroica, Op. 55). The writers she comments on suggest that the music is depicting events in the mimetic mode. She suggests that the writers should be referring to the epic mode, in which the poet describes the actions or words of characters.

However, I would argue that the music seeks to convey the poet’s/composer’s reaction to actions or emotions, rather than to describe the behaviour of characters. The music is therefore operating in the lyric mode. All roads appear to me to lead back to the view that music of this period at least operates in the lyric mode. The poet speaks in their own voice.

Therefore I would argue that music of this type operates on a basis distinct from literary narratives, since there is only one voice, that of the poet or narrator. Parallels from the study of literary narratology should therefore be applied with care.

2. Narrative and non-narrative art-music.

However, I would not wish to argue that this parallel with lyric poetry is appropriate to all music or even all Western art-music. It is argued that narrative in music is not universal but may depend on the intentions of the composer or on the aesthetic context within which the composer was writing. Hatten (1991) argues that:

… certain musical styles would seem more capable of exploiting the basic technique of narrative: the meaningful ordering of expressive events, and their meaningful description. (Hatten 1991: 96)

He considers nineteenth century music to be the prime example.

Seaton (2005) also regards narrativity in music as an historically specific phenomenon, and that the music of the Classical era could ‘… better be regarded as dramatic than as narrative.’ (Seaton 2005: 69) He considers that the change to a narrative style took place with the transition to the romantic era. While the classical style had introduced elements which can be read as a plot, characteristic of drama, romantic music introduced the idea of voice into music and therefore created the sense of a narrator. (Seaton 2005: 69)

In seeking to characterise narrative in music from the Romantic era onwards, Karl (1991) points out that while music unfolds in time, like drama, it ‘… does not present its stories in real time.’ (Karl 1991: 44) This is characteristic of narrative rather than drama. We therefore perceive a persona recounting experiences to us. The question is the identity of this persona.

We can therefore see that music of different eras appears to function in different ways, and that music from Beethoven onwards is regarded as possessing narrativity. This point is central to the following discussion of listener-generated narratives.

3. Narratives in the mind of the listener?

There have been many scholarly narrative interpretations of music from Beethoven onwards. They share two characteristics. On the one hand they are made in representational terms. They often come close to presenting music as if it were a literary narrative, in the epic mode.

On the other hand, while evidence is presented that the interpretation suggested may be close to the narrative intentions intentions of the composer, in no case are the scholars able to demonstrate the composer’s intentions directly.

Examples are the analyses of Schumann (Newcombe 1984 and 1987), Beethoven (Jander 1985 and 1995) Beethoven (Hatten 1991) Schumann (Kramer 2002) Humpnerdinck (Neuberger 1997), Beethoven (Kerman 1992) Beethoven again (McLary 2004), and Chopin (Klein 2004).

In the absence of direct evidence of composer intentions, I would therefore argue that these remain listener-generated narratives. If the composers conceived the music in terms of representational narratives, we do not know exactly what they were. Equally, the composers may have conceived the music purely as an abstract narrative in sound, without specifying a representational narrative even to themselves in the process of composition.

The most we have as evidence of the composers’ intentions are titles, markings, or comments by composers which hint at a narrative intent in the music. For instance, Tchaikovsky gave his Symphony No. 6 in B minor, Op.74 the title Pathetique. He considered describing it as a programme symphony but changed his mind. We are therefore invited to speculate on the narrative which may lie behind it. Such works set up an expectation of a narrative experience.

I would argue that it is possible to experience such works entirely as abstract sound narratives, and that this is probably how listeners initially encounter the music. This does not invalidate the idea of narrativity being present in the music. It implies only that narratives may be experienced in abstract as well as representational terms.

The human brain is known to work in a way whereby neurons are stimulated in similar ways by parallel experiences. For instance, Raffensperger (2013) shows that the same sets of neurons are stimulated by being physical hurt or being insulted by someone. The human brain readily interprets something experienced in one way, for instance an abstract musical narrative, as a different form of experience, such as a representational narrative. A full description of this process has been presented by Gallese and Wojciehowski (2011).

When a listener encounters a piece of abstract music, they are clearly unable to explain it in terms of its original medium. They therefore summarise the experience in words as a representational narrative.

I would therefore argue that listener-generated narratives, including those presented by scholars, are not evidence of the existence of narrativity in the music itself or of the composers’ intentions. They remain, however, valid in their own terms, for enabling listeners to understand and discuss a piece of music, and are invaluable as scholarly interpretations to aid listeners and interpreters.

4. Does it require a listener to create the narrative in music?

It is therefore instructive to look at the work of writers who have argued that it is the listener is the true narrator in music, and at the work of those who have dismissed the listener-narrator and therefore questioned the validity of the idea of narrativity in music.

Almén (2008) for instance has published a comprehensive theory of narrative in music, and takes the view that:

Musical narrative is the process through which the listener perceives and tracks a culturally significant transvaluation of hierarchical relationships within a temporal span. (Almén 2003: 12)

However, he (Almén 2008) goes on to argue that narrativity in music arises from:

… a hierarchy set up within a system of signs [that] is subjected to change over time; this change, filtered through an observer’s design or purpose, is interpreted as being a cultural hierarchy. (Almén 2008, 40)

In other words, it requires both a composer to set up a system of signs, and a listener to interpret the signs, to reveal narrative meaning. This implies that some music may be amenable to narrative interpretation while other music may not be. There must therefore be a question of composer intention to make the music concerned amenable or otherwise to narrative interpretation.

However, Almén considers that the narrative does not come into existence until the listener interprets it in representational terms. This appears to reflect a view that, unless a narrative can be described, and that necessarily means that it must be expressed in representational terms, then it has not been brought into existence. In this context, it is significant that Almén considers that all narratives are expressions of a limited range of narrative archetypes of the kind defined by Northrop Frye (1957) and Liszka (1989). Such narratives are inevitably representational.

Other writers, however, have dismissed the idea of narrativity in music on the grounds that it only exists in the mind of the listener, and is not inherent in the music itself.

Nattiez (1990) refers to the experience of listening to music, and ‘… an intuition of common sense: through the work, the composer speaks to us.’ He regards this as an example of the pattern of human behaviour described by Ricoeur (1984) of interpreting any sequence of events in narrative terms. His objection to this listener experience as evidence of narrative in music is that the listener has:

… a wish to complete through words what the music does not say because it is not in its semiological nature to say it … (Nattiez 1990, 245)

While I agree with Nattiez’s questioning of the idea of the listener as the true narrator in music, he is surely taking too narrow an approach. A narrative can exist as a series of emotional or psychological states communicated through abstract sound – the familiar experience at the end of a symphony that one has been taken through a process or on a journey.

However, the moment a listener develops a representational interpretation of abstract music, they replace the poet’s narrative with their own. In seeking to grasp the nature of the poet’s narrative, it slips away between their fingers, and we are left only with the listener’s subjective interpretation.

I would therefore disagree with Nattiez’ (1990) questioning of the idea that there can be narrativity inherent in the music itself. The issue is whether the systems of signs set up by composers from Beethoven onwards have narrativity inherent in them, and in what way this suggests the identity of the narrator in such music.

I would also question the opposite view, presented by Almén, that musical narrative depends on the listener to articulate it. Instead I would argue that there can be narrativity inherent in the systems of abstract sounds established by composers, and that is is not necessary to make a representational interpretation in order to demonstrate its existence.

I agree with Hatten, who queries the validity of listener-generated representational narrative interpretations of music. He argues for an avoidance of ‘… overly impressionistic metaphors or overly specific programmatic analyses …’ (Hatten 1987, 208), although arguing that the works he has analysed demonstrate a ‘… semiotic level of dramatic closure.’ (Hatten 1987, 208) which cannot adequately be described through formal or harmonic analysis.

The fact that narrative responses to abstract music relate primarily to music of the nineteenth century and beyond suggests that the music of that period invites narrative interpretation in a way which earlier music does not. The composers set up their systems of signs differently from composers of earlier eras. It remains to be defined how this process takes place, and to identify the narrator whom we appear to encounter.

5. The composer, or the narrative persona?

It is necessary to ask to what extent we encounter the composer themselves as a narrator when we listen to music from Beethoven onwards.

Hatten (1991) describes abstract music as ‘… a sequence of emotional states, rather than referential events …’ (Hatten 1991, 75). He considers therefore that the composer is the implied narrator.

In order to understand this concept of the composer as narrator, it is necessary to examine how the process of composition works. The problem impeding this is the scarcity of information on composer intentions and the compositional process.

Two writers have set out models of how the process operates. Tarasti (2004) puts forward a model which attempts to define this relationship. He (Tarasti 2004: 298-9) puts forward the concept of the Implied Composer, and explains that:

… the implied composer is someone with a certain competence, who provides his musical message with signs that the implied listener can … receive and decode correctly. (Tarasti 2004: 300)

I would read this as meaning that his Implied Composer is the narrative persona who we experience as addressing us in music from Beethoven onwards. They are not identical to the physical composer.

Sloboda (1985: 107/8) refers to musical ideas coming ‘… unbidden …’ to composers, and that ‘those that please’ are retained and used. He sets out (Sloboda 1985: 118) a model of the process. The composer conceives ideas for a piece within the context of their musical knowledge and the constraints on the form of the piece. They develop these into an interim form through the application of their repertoire of compositional devices, and then into the final form of the piece, with a constant process of interaction between their imagination, their technical knowledge, and the constraints on the type of piece required.

He considers (Sloboda 1985, 121) that there is much evidence that the creative process is largely subconscious, with the creator perceiving themselves as a bystander.

He goes on to give an example (Sloboda 1985: 125-136) of his own compositional process in writing choral music, describing it as consisting of a mix of ideas popping into his head, and the application of critical judgement and technical expertise in selecting which ideas to use.

From my own experience as a composer, I would summarise the process as follows. Working within the context of of each new piece, a composer begins to identify musical ideas, which arrive largely unbidden, but originate in their subconscious through the interaction of the constraints on the nature of the piece and its instrumentation, and the composer’s fund of technical knowledge and life experience. The composer interrogates these ideas, and investigates their potential. They choose ideas which please them, but also which have potential in the context of the piece concerned.

They begin to build these ideas into a longer structure, allowing the musical ideas to grow and develop as seems right, but subject this subconscious process to the scrutiny of musical craftsmanship. The draft piece is completed and revised.

Whether the final piece contains narrativity would therefore be a question of the position which the composer adopted in relation to it, which in turn will relate to both their nature as a individual and to the aesthetic within which they operate. Composers of the Baroque, and much of the Classical, eras did not seek to compress into a single work material which reflected a longer process of experience than could be encountered within the duration of the piece concerned. Composers from Beethoven onwards are commonly seen as doing so.

Seeking to compress experience in this way has the implication of the adoption of a narrative stance towards the material in order to summarise and recount it. The composer has to adopt a posture towards the piece and the material in it. This appears to me to be the key to the question, and it has best been described as the adoption of a narrative persona during the process of composition. This concept of the narrative persona relates closely not only to my experience but also to the models of the process developed by Tarasti and Sloboda.

6. Further study

The question, therefore, is how to study the extent of narrativity in pieces of music while resisting the temptation to interpret the music in representational terms. A potential answer to this question is to seek information directly from composers, and to seek it in a form other than a post-hoc rationalisation of their piece as a representational narrative, since that would not be fundamentally different from any listener-generated narrative.

What is required is information from composers, and accounts of the nature of pieces of music and the musical ideas which they contain, in technical rather than representational terms, yet in terms which illustrate the extent of the narrativity inherent in the music.

I am therefore planning to examine my own music, which I have long regarded as mainly consisting of forms of narrative in the abstract. I will try to develop a way of describing the narrativity, if any, in some my own music, working while the compositional process is fresh in my mind.

If I have any success in establishing a method for analysing music in this way, I then plan to invite other composers to provide similar analyses of their music as a basis for further research.


Abbate, C. (1991) Unsung Voices: Opera and Musical Narrative in the Nineteenth Century, Princeton, Princeton University Press.

Almén, B. (2003) ‘Narrative Archetypes: A Critique, Theory, and Method of Narrative Analysis’ Journal of Music Theory 47/1: 1–39.

Almén, B. (2008) A Theory of Musical Narrative, Bloomington, Indiana University Press

Aristotle (1974) Poetics, Trans: Butcher S.H. In Dramatic Theory and Criticism: Greeks to Grotowski Dukore B.F. (ed) Florence KY. Heile and Heile: 31-35
Cone, E. T. (1974) The Composer’s Voice, Berkeley, University of California Press.
Gallese, V., and Wojciehowski, H. C., (2011) How stories make us feel: towards and embodied narratology, published in California Italian Studies Journal, 2(1), Studies Multicampus Research Group, UC Office of the President, available from eScholarship, University of California – http://escholarship.org/uc/item/3jg726c2 Accessed on 17.2.12
Hatten, R.
Hatten, R.S. (1987) ‘Aspects of dramatic closure in Beethoven: A semiotic perspective on musical analysis via strategies of dramatic conflict.’ Semiotica Vol. 66, 197-210

Hatten, R.S. (1991) ‘On Narrativity in Music: Expressive Genres and Levels of Discourse in Beethoven’ Indiana Theory Review, Vol. 12: 75-96
Jander, O. (1985) ‘Beethoven’s ‘Orpheus in Hades’: The ‘Andante con moto’ of the Fourth Piano Concerto’ Nineteenth-Century Music 8/3: 195–212.

Jander, O. (1995) ‘Orpheus Revisited: A Ten-Year Retrospective on the Andante con moto of Beethoven’s Fourth Piano Concerto’ Nineteenth-Century Music 19/1: 31–49.
Karl, Gregory. (1991) ‘The Temporal Life of the Musical Persona: Implications for Narrative and Dramatic Interpretation’ Music Research Forum 6: 42–72.

Kerman, J. (1992) ‘Representing a Relationship: Notes on a Beethoven Concerto’
Representations 39: 80–101
Klein, M. (2004) ‘Chopin’s Fourth Ballade as Musical Narrative’ Music Theory Spectrum 26/1: 23-56
Kramer, L. (2002) ‘Rethinking Schumann’s Carnaval: Identity, Meaning, and the Social Order’ In Musical Meaning: Toward a Critical History. Berkeley: University of California Press, 100–132
Liszka, J. J. (1989) The Semiotic of Myth: A Critical Study of the Symbol, Bloomington: Indiana University Press
McLary, S, (2004, first published 1997) ‘The impromptu which stood on a loaf’ in Narrative theory: Critical concepts in cultural studies, Bal, M. (ed.) London, Routledge 20-35
Nattiez, J.-J. (1990) ‘Can One Speak of Narrativity in Music?’ trans. Ellis K., Journal of the Royal Musical Association 115: 240–57.
Neuberger, J. (1997) ‘Tales of Hoffman and Others: On Narratizations of Instrumental Music ‘ in: Lagerroth, U.-B., Lund, H., and Hedling, E., (eds.) Interart Poetics: Essays on the Interrelations of the Arts and Media. Amsterdam, Rodopi, 117-136
Newcomb, A. (1987) ‘Schumann and Late Eighteenth Century Narrative Strategies’ Nineteenth-Century Music, 11/2: 164-74
Newcomb, (1984) ‘Once More “Between Absolute and program Music”: Schumann’s Second Symphony’ Nineteen-Century Music 7/3 233-250
Northrop Frye, H. (1957) Anatomy of Criticism:Four Essays, New Jersey: Princeton Univ. Press
Raffensperger, l. (2013) ‘Words can never hurt me?’ New Scientist, 12: 37-39
Plato, (2004) ‘Extracts from The Republic,’in Narrative Theory: Critical Concepts in Literary and Cultural Studies, Vol 1, Part 3. Bal, M. (ed.) Abingdon, New York, Routledge: 135-138. Source: Plato, The Republic, Ed/trans: Shorey, P. (1930) Cambridge MA, William Heinemann Ltd and Harvard University Press
Ricoeur, P. (1984, 1985, 1988) Time and Narrative (3 Volumes). Chicago, University of Chicago Press
Seaton, D. (2005) ‘Narrative in music: The case of Beethoven’s “Tempest” Sonata’ in: Narratology beyond literary criticism: Mediality, Disciplinarity, Meister, J.C.(eds.) Berlin and New York, Walter de Gruyter, 65-81
Sloboda, J. A. (1985) The Musical Mind: The cognitive psychology of music. Oxford, New York; Clarendon Press
Tarasti, E. (2004) ‘Music as a narrative art’ in: Ryan, M-L. (ed.), Narrative across media- the languages of storytelling, Lincoln NE, Board of Regents of the University of Nebraska


Read Full Post »