6.2 The idea of the background, and course summary

Mozart's music, as we have already seen, achieves its sense of coherence through a series of tensions followed by resolutions: dissonances imply resolution to consonances, and dominant chords imply resolution onto tonic chords, giving a sense of finality and order to the music. Mozart's music, like nearly all tonal music, sounds ‘goal-directed’: we expect harmonic progressions, however complex, to resolve onto the home tonic, and this arrival at the tonic provides a satisfying sense of completeness.

All the extracts studied in this course have shown the importance of the drive towards a perfect cadence; this is one of the defining characteristics of Mozart's musical language. When a perfect cadence is delayed, this has the effect of incompleteness, while the cadence itself tends to signify closure or arrival. It is this basic premise that underlies the concept of ‘background’, pioneered in the late works of Schenker.

The background itself will be introduced properly in AA314_3. It is a theoretical formulation which can sometimes seem rather abstract, or seem to deny the uniqueness of individual works, since all background structures are very similar. Indeed, the theory has always had its share of fierce opposition. In essence, it is based on four main observations about large-scale structures in tonal music.

• They tend to express underlying linear shapes.

• They conform to the rules of dissonance treatment and counterpoint in the way that voices move against one another.

• They tend to begin by establishing a tonic chord.

• They drive towards closure, defined by a perfect cadence in the tonic.

I have already tried to show how Mozart's melodic and harmonic language can be seen to grow out of a limited number of basic deep-level shapes (particularly the descending line through a third or a fifth, supported by a tonic–dominant–tonic pattern in the bass), with the foregrounds and musical surfaces acting as elaborations of this deeper structural logic.

You may be tempted to think of voice-leading theory as reductionist: a process of stripping away successive layers of beautiful and distinctive music to arrive at some abstract generative model. In fact, this is only half of the story, and I hope that, when reducing musical surfaces, we have done so as a means to an end. When analysing Mozart's music you should bear in mind that the deeper-level models are artistically worthless in themselves: it is not the abstract structure itself, nor the identification of the notes of the structure in a piece of music, that is analytically interesting; rather, analysts are more interested in the process by which the large-scale drive towards the tonic is prolonged, delayed, embellished and decorated, to produce the middleground, the foreground and ultimately the musical work itself.

You will perhaps already have a view of the value, as well as the limitations, of the methods I have employed in this course. Voice-leading theory can be extremely useful if it allows us to make observations about style which simple chordal analysis cannot explain, if it can show hidden features such as linear part-movement we hear but cannot otherwise show analytically and if it reveals motivic connections between fragments of music that seem unrelated on the surface.

Conversely, voice-leading theory is not in any way intended to supersede the more traditional approaches such as analysis of chordal harmony, rhythm or form (the last two, for instance, have not been considered very much in this course). Instead, this approach is intended to run in parallel with these other methods; in particular, large-scale voice-leading processes can (and often do) interact with the patterns of musical forms as defined through thematic identity.

Once you have completed this course, you can move on to the third of the three courses on harmonic analysis, Voice-leading analysis of music 3: the background (AA314_3).