Yesterday I wrote the introduction for a reading in the mode of Schenkerian hermeneutics, using as a model an article by Frank Samarotto. Here is the final paragraph again:
The parallel to D779n13 should be unmistakable: we just replace ^#4 with ^#5 (that is, E#) and the same process of chromatic progression and "rethinking" or "sublimation" takes place in the second strain as E5 becomes E#5 but then is pulled back to E-natural5 for the reprise. I will explore the details of a reading based on this idea in tomorrow's post.
I will use the proto-background ^3-^5 as the framework. Any of the four readings with ^5 would be workable, but, among all the proto-backgrounds, ^3-^5 is the one I prefer here, in part because (as I wrote in the comparison post) the strongly teleological readings seem out of sync with a waltz whose sections move unpredictably, and the analyses that isolate either ^3 or ^5 (or ignore them, as in the unison ^1 and octave) are much harder to hear than the one that combines ^3 and ^5 (soprano and alto, male and female dancers).
The inflection of ^5 is deeply embedded in the design, as if, in thematic terms, the rising motion is what this waltz is about, an increasing exhilaration in the dancing couple as the waltzing turns continue. Even if he wasn't a dancer, Schubert would have known from the experience and comments of his friends that, because of a basic difference in center of gravity, women in general find it much more pleasurable to turn and spin than do men -- in this portrait of dancing, then, the upper voice expresses the exhilaration, not the lower. The action is shown in the small, as well, in the later-level N
and in the immediate, as, at the beginning, E5 barely sounds before it pushes upward to F#5 and then to A5:
After all these hints (or preliminary attempts), it is hardly a surprise when the soprano pushes (completes the middleground turn?) up to A5 in the cadence.
There is thus a definite kinship -- a shared impulse -- between the E# of the background, the rising cadential gesture of the middleground, and the neighbor note of the foreground.
It is a curiosity that ^#4 or D# never appears in D779n13 as an inflection of ^4 -- instead, D# is ^2 in C# minor/major. The role of distinctive chromatic note is thus all the more clearly thrown onto ^#5 in the second strain -- but that note has a decided inner conflict: in the immediate, it is stable and F#5 is the neighbor, the dominant seventh that must resolve to it, but in the background, E# would move further upward, to that same F# as tonic to E#'s leading tone. The moment of sublimation comes in the transition, where the E#-G# pair that are the C#-major equivalents of C#-E begin to move, not up, but down in a chain of parallel tenths (see graphic below), thus imitating the determined descent of the suspension chain in the first strain. This motion would seem quickly to cancel out E#, but the E-natural belongs to a very unstable chord. Eventually the tenths arrive at C#-E as the reprise comes into focus. But in the meantime the soprano has gone out of the voice leading altogether as the ascending fourths pile up and F#6 is reached; the soprano even reaches again for a high A -- see the text at the top of the graphic. Here is that same transcendent voice, reaching up and out of the harmony, that Lewin describes.
The sublimation, then, is a complicated process here. In the background, exhilaration/ecstasy gives way in the end to the form of the dance. In the immediate, however, the soprano holds to that feeling (or its memory) and floats free of the voice leading for a moment. And in the middleground, Schubert has accomplished something that would take a whole generation of French opera composers after him to manage by force of sheer repetition: make the rising gesture in the waltz's final cadence seem quite conventional, the most natural thing in the world.
References:
Frank Samarotto. "Sublimating Sharp ^4: An Exercise in Schenkerian Energetics." Music Theory Online 10/3 (September 2004): link.
Lewin, David. "Women's Voices and the Fundamental Bass." In Studies in Music with Text, 267-281. New York: Oxford University Press, 2006.
