A reductive voice-leading analysis can reveal many “hidden” aspects of a composition, including underlying stepwise motion, large-scale unifying features, goal-directed motion, etc. The harmony, which on the surface may sometimes be obscured (in the best sense), can be clarified and distilled to common voice-leading paradigms, too.
Knowing that the underlying harmonic progression and its contrapuntal foundation are stylistically consistent (i.e. they follow stylistic “rules”), it’s tempting to think that those “rules” can be extrapolated to account for the foreground (i.e. that the process can be reversed). That is to say, if the middleground follows “rules”, and the foreground is an elaboration of the middleground, the foreground therefore must also follow “rules”. While that statement harbors a sense of logic, it’s mostly the faulty kind.
Both levels certainly share structural and stylistic constraints, but the influence they exert in the middleground cannot account for the creative choices made by the composer in the foreground. Schenker articulated this as being the difference between “strict counterpoint” and “free-composition” (Der freie Satz). Put another way: If gravity were an analog to musical style (“rules”), and rocks were an analog to individual musical voices, and throwing them in the air were an analog to composition, then the way they ultimately fall back to the ground can be accounted for by gravity. But what gravity cannot account for is which rocks I chose to throw, how many were thrown at once, what size they were, what direction they were thrown, how hard they were thrown, and so on.
I’m not suggesting that a composer’s style and technique are unknowable; through patient study it’s possible to create a taxonomy of a composer’s musical technique and syntax, and perhaps even compose something in a similar style (David Cope’s work with EMI alone has demonstrated this). But is it possible (or plausible) to compose-out from a given middleground, and reasonably predict the composer’s choices?
For example, here are two middleground reductions of the same six measures from a piece by Bach:
It’s all pretty standard stuff: a prolongation of the tonic with a 3rd-progression (b-g) in the upper-voice, and the suggestion of a sequence among the first three structural pitches. You could even go so far as to realize the inner voices based on the lower graph and come up with something that is both pleasing to the ear and stylistically consistent. But is it possible to anticipate, if not all, at least most of Bach’s choices? It’s certainly possible. Here is the same passage one step closer to the foreground—a first pass at removing diminutions and normalizing the rhythm:
This chord progression could mostly be mistaken for undergraduate part-writing work. I say “mostly” because there are some subtle and nuanced choices, including the first-inversion chords at the very end of measures 1 and 2. And while it hasn’t been too difficult to fill in the gaps between these middleground layers, are there any clues that might help in predicting Bach’s foreground choices?
Here are the original six measures—the introduction to Sei Lob und Preis mit Ehren from cantata 167, “Ihr Menschen, rühmet Gottes Liebe”:
Two recommended recordings, both featuring the Amsterdam Baroque under the direction of Ton Koopman (de Nederlanders weten hun Bach!): Cantatas Volume 8, and an instrumental version from Simply Baroque, featuring Yo-Yo Ma.
A voice-leading graph can reveal how a composition may be a remarkable elaboration of unremarkable materials, but it’s largely a one-way street; we may be able to peek behind the curtain but it’s no guarantee we’re seeing beyond the veil, so-to-speak.
This isn’t news though; music theory has never been purposed with, nor tooled for, predicting the mind of the composer. But sometimes it can seem like you’re getting so close that it fosters second-guessing, which can result in qualifications in the form of ‘…is just…’ as well as ‘…is only…’ expressions in analyses. While those expressions can be used to amplify compositional technique (e.g. “The use of inversions subtly disguises the fact that the progression is just I – IV – V”), they can also be used to mitigate or to marginalize (e.g. “Despite the effusive chromaticism, this passage is only I – IV – V”). However, the degree to which you qualify something usually has more to do with how you feel about the piece rather than the piece itself (which usually makes for weak arguments and untenable assertions). Ultimately though, this is the burden of interpretation—not analysis.
Structures closer to the background are those that tonal compositions have in common. Closer to the foreground are the individual characteristics that make each piece unique. And even though the middleground may be only one step removed from the composer’s free-composition, in some ways it can still feel like it’s light-years away. Not to worry though—Bach himself left us with some encouraging words:
“What I have achieved by industry and practice, anyone else with tolerable natural gift and ability can also achieve.”