Prolonged Pedal Point
Bach Prelude in C major (WTK 1). Mozart, Sonata KV.330/ii
The aim of this short analysis is to discuss an instance of what could be considered a ‘prolonged’ pedal point.
In an organization with rather stable articulation (the whole piece is made from similar motifs outlining a chord), implied (not explicit) melody, stable tempo and no dynamic contrasts, harmonic progression is the most dynamic parameter. The progression is from the beginning rather simple. It emphasizes fifth relations, and the chords are organized either in short semi-phrases (such as mm.1-4), or in simple sequences. Except from a couple of surprises, most of the strongest implications are realized. Or, almost no implications are denied. Or, perhaps most correctly, there are few moments with clear expectations, few strong sound terms; the music seems to be weaved from small beautiful pearls, that are almost even beautiful in their own, as they are in a string. This statement implies that, even though the harmonic analysis is perhaps most informative of the piece, it is not (only) the harmonic progression that makes this piece what it is.
The first surprise is in m.12. After three 4-bar phrases, and also because of the similarity between mm.9-11 and mm.5-7, in m.12 the listener expects a chord that could round off the current 4-bar phrase (as felt). Instead, Bach chooses to start another sequence in m.12 (realized in mm.12-16), which leaves the third phrase shorter. Next surprise is in m.23, where a very strong implication of the dominant chord is denied (the dominant is delayed, appears in m.24).
The dominant pedal, the tone G, is present in eight bars (mm.24-31). Its appearance could be divided in four stages. The first stage is the G7 chord in m.24, the moment when we do not know that the bass tone will be sustained for eight bars. The second stage (includes retroactively m.24) is the prolonged dominant harmony (first by a neighboring chord in m.25, then by the suspension in m.26). The third stage is an interpolation of two bars in mm.28-29, and the fourth stage is returning to the end of the second stage. Let us examine these in turn.
The perception of the G7 chord in m.24 is influenced by the events in mm.22-23, and the understood proportions based on the grouping in the bars heard so far. As I have already mentioned, most of the chords are understood in groups of four (progressions of four bars directed towards the tonic, thereafter several sequences organized in 2x2 bars), as indicated by the blue lines in the score. When a new ‘phrase’ begins in m.20, the expected G-chord in m.23 would again complete the four bars (notice that in that case the dominant would be the goal of the progression). The surprising chord in m.23 is perhaps not immediately understood (“where does this lead to?”), but at the moment the dominant really arrives in m.24, it could retroactively be grouped with m.22 into a pre-dominant section. Uniting m.22 and m.23 retards the harmonic rhythm, and the duration of V is expected to respond to this deceleration. The chord in m.24 could thus logically be prolonged.
The second stage (mm.24-27) is in that sense a logical continuation of the previous progression and as such it could be considered to be (just) a somewhat prolonged dominant chord. The bass tone is completely consonant throughout this stage (which, according to Schenker, disqualifies it from being a pedal). However, the G is clearly ‘working’ as a dominant pedal. Why? I think it is because of at least two reasons: 1) there are melodic and harmonic processes above the static tone in the bass, and 2) the tone collection is the same in the beginning and at the end. Imagine the following:
In m.24, instead of B there is C (Gsus4 as the sounding chord).
In this case, the C would remain dissonant until the resolution in pure G7. In this variant, the dominant harmony would be recognized in all the bars, but the dominant chord itself would arrive only at the final moment. Although still full four bars, this situation would perhaps rather be seen as a long suspension, than as a pedal. This comparison suggests the importance of starting the progression with the chord that is also the target at the end of the pedal. Rounded with the chord in m.27, the progression becomes a typical example of a dominant pedal: the static bass tone opposed to the moving harmony, ending by returning of the harmony to the bass, so that they can resolve together into the tonic. In the second stage of this G-pedal, the listener recognizes the pedal model.
This recognition incites the expectation that the pedal will resolve after the re-uniting of the bass and the harmony. The third stage begins with expectation-tension, immediately turned into denial-tension in m.28. Instead of a resolution, the harmony goes once more away from the chord. This is the first time that the G is really dissonant with the chord. This time the listener might experience also surprise-tension. The dominant pedal was so far already long enough to imply the possibility that its resolution will terminate the piece. It will, indeed, but only after a detour.
The fourth stage of this pedal is the ‘resuming’ of the stage two, returning to a prior point (c.f. mm.26-27), which, once again ‘works’ as a strong sound term. Once again, the listener is expecting the tonic.
The four stages could naturally be analyzed as two-and-two, resulting in division to mm.24-27 and mm.28-31. The first of the two is rounded by the expectation of resolution. The second begins as a pedal within pedal. In this sense, this pedal is itself ‘too long’.
Another, quite different example of a ‘too long’ pedal tone, this time on the tonic, is found in the second movement of Mozart’s piano sonata KV.330. The middle section (mm.21-40) is in small ternary form. The return of a (now a1) repeats only the first part of a, and closes with a changed cadence.
The first part (mm.21-28) opens on a short tonic pedal. Only the third beat of the first bar is dissonant with it. This characteristic pedal is immediately recognized, and - as this is a short piece that is easily remembered - the whole phrase a is recalled. The prolonged pedal tone in m.39 comes as a surprise - at least the first time the piece is heard - and draws the listener’s attention to the coming events. It raises the tension only slightly: the listener understands this gesture as a closing T-pedal, and expects this part to finish. The repetition of the original eight bars is not expected. We could say that the pedal tone has guided the listener through the piece.
 Depending on the chosen context, there are several chords that could close these four bars. The similarity of mm.9-11 and mm.5-7 could incite the expectation of the C-major chord in m.12. In case the listener feels a modulation to G major (due to the power of the - now different - bass line), the tonic chord (G) could seem to have come one bar too early. The phrase could still be rounded with, for example, E-minor chord, or a retransition towards C major, in the form of G7 chord.