Mark DeBellis: Music

In Berys Gaut and Dominic Lopes, The Routledge Companion to Aesthetics (2001) 


This document is a summary of DeBellis.
My personal comments are in red. 
These comments have been added to help students understand his arguments.

The Issues

Why is music of value in our lives?

  1. Expression theories value music for its ability to arouse emotion. 

  2. Referential theories emphasize its capacity to reveal aspects of the extramusical world (including, but not restricted to, emotions). [In other words, cognitivism]

  3. Formalism focuses on non-referential musical qualities, such as musical structure. We find our perception of these structures to be rewarding. [In other words, aestheticism]

There are also exponents of HYBRID views, since none of the above have to be mutually exclusive. For example, awareness of musical structure may be the basis for having an emotional response.

Expressive Character

It is common to attribute emotive characteristics to music, and to value it for the presence of these characteristics. For instance, we distinguish sad music from happy music. But what are we saying when we say this? Many theorists defend the thesis that we attribute emotions based on how it makes us feel. [This proposal is just an application of the broader idea known as the arousal theory.]

Peter Kivy famously argues that arousal theories are false: there is no need to feel sad (or pity, or anything else!) in order to recognize the sadness of a piece of music. [I don't need to feel sad to recognize YOUR sadness, do I?] Kivy regards expressive character as a combination of natural resemblances and musical conventions. The contour thesis: the overall movement of a sad song naturally resembles the movements of a sad person (slow, dragging), and we naturally see this resemblance. [While one doesn't fully appreciate expressive music if one cannot see what it expresses, there is nothing wrong with being a dry-eyed critic.]

Plato held that the resemblance is between the distinctive speech patterns of persons feeling various emotions, and musical patterns expressing those same emotions.


Kivy's response to arousal theorists assumes that music can enter into representational relationships. (Music itself, that is--music without words.) But can it? Painting and literature involve representation, the showing of definite subject matter, but can music?

Eduard Hanslick argues that music is never representational, and specifically, it cannot represent emotional states. Some emotions feels identical, and we distinguish them according to the thoughts that are involved. But music does not present clear thoughts of the kind that can distinguish emotions. So music cannot represent definite emotions. (Click here for more about Hanslick's arguments.) A response is to allow that vague representations are still representations (e.g., a grainy photograph will represent what it shows without showing it very clearly).

Scruton argues that understanding music is prior to interpreting it as a representation. And one can fully understand all that there is to understand without finding any representational content. [Hanslick put it this way: "In the pure act of listening we enjoy the music alone and do not think of importing into it any extraneous matter."] But if understanding the music is independent of understanding any thoughts conveyed by it, representation is not part of the art of music. A response is that one cannot understand music unless one understands that one segment of it will refer to other passages; reference requires interpretation of thoughts, so music does present thoughts.

Exploiting the possibility of indeterminate representations, Walton argues that understanding music involves imaginative participation in a fictional world in which things happen. There is movement. There is conflict. But these things happen without representing specific agents whose activity we follow. We can also imagine that we are experiencing emotions by imagining that the music is an emotion (as opposed to imagining that a character has an emotion).

Reference and Signification

Goodman proposes that some non-linguistic items can refer, as when a swatch of cloth illustrates the quality and color of some cloth [or the way a food sample at the grocery store exemplifies the specific brand being sold]. The swatch doesn't represent the cloth. It literally exemplifies it. 

In addition to this literal exemplification, there may be metaphorical exemplification. Sad music might metaphorically exemplify sadness. 

Kivy does not find this useful. A swatch exemplifies, but it says nothing about what it exemplifies. But art that does not say anything "interesting or useful about" its subject is not worth having. [In other words, what ABOUT the sadness we're shown?]

Understanding Music

Formalism emphasizes the actual process of understanding music. This process is seen as parallel to the activities involved in understanding language, but where language is used to refer and represent, it is not obvious that all the structures of music have this same aim. [Most art is about the world, but music seems the one art that is art for art's sake. We find musical patterns interesting for their own sake.]

We understand music as we understand spoken language: we must integrate mind and ear. This requires education. [Is this just another way of talking about musical taste?] Consequently, music is the art form that is most immediate pleasurable [calling for understanding but not interpretation], but the most difficult to respond appropriately to.

But is conscious awareness of what is going on in musical structure of benefit to listeners? Levinson argues that awareness of large-scale form is irrelevant; moment-to-moment understanding is almost always sufficient to understand music. If Levinson is correct, a great deal of what is normally taught in "music appreciation" is useless in understanding music.


                        Last updated May 6, 2004