Monday, December 21, 2009

Schubert's F-major experiments

Quoted from Schumann's review of D 365:
. . . lovely little genii, floating above the earth at about the height of a flower--though I do not much like . . . the last three aesthetic errors [nos. 34-36] which on the whole I cannot forgive their creator.
What did Schumann mean? What was it about these waltzes that was so troublesome? I would suggest that it was exactly the same spontaneous (and thereafter practiced) creativity that expressed itself in the abrupt shift to III in D779n13 but that in other cases found itself confined within the tiny frame of (what "should" have been) a 16-bar binary dance. (Or a simple ternary design with a transposed A-strain as its contrasting middle -- see examples in yesterday's post).

After a long series of waltzes in sharp keys (everything from G major to B major) in ns16-30, a single C-major waltz intervenes before the final five waltzes in F major. No wonder many commentators regard this last group as "tacked on" -- which may very well be true and may have been an element in Schumann's displeasure with ns34-36. The effect is all the greater because the five waltzes do hang together as a set and could easily be played independently of D365 in dance-trio groups, most likely as 32-33-32-34-35-34-36-34. This design would support a dance or a performance of nearly 5 minutes. (The grouping of 32 with 33 is supported by the appearance of this pair in two manuscripts in Schubert's hand from 1821. The grouping of ns34-36, similarly, is found in another manuscript from the same year. See Litschauer.)

The "theme" of this last group is announced immediately in n32: the exploitation of chromaticism. The non-tonic opening does refer back to n31 but retrospectively, after the phrase-aligned cadential progression plays itself out, the G7 is understood as chromatic. Nevertheless, the focus of the chromatic play is, as we would expect, in the contrasting middle of the binary form (beginning of the second strain) or of the small ternary form (the "B" section). As in D779n13, Schubert turns the affects around by making the transposed variant of the theme in the contrasting middle more stable harmonically than the original.

Schubert plays out the idea of diatonic/chromatic contrast in another way in n33. Here what would have been a 16-bar waltz with repeat signs is still 32 bars total but each "repeat" is written out: bars 9-16 = bars 1-8 but the cadence is to bIII, not I; bars 25-32 are a variant of bars 17-24 where the Ab major of the earlier bars shifts directly to the F major of the later ones. The design overall is closely related, in its blocking out of chromatically related key areas, to the ternary forms I discussed in yesterday's post. This is also a waltz that could very easily have arisen in improvisation and repetition, and seems little removed from that state as it stands.

Now, on to the three "errors." In n34, it is easy to imagine a motivic motivation in improvisation for the striking augmented sixth chord that opens the second strain: the chromatic passing tone B-natural5 and its run up to D6 is compressed into a diminished third B-natural4 to Db5 in the second strain. As the eight-bar cadential function unfolds, the bass charts the inverse: Db3-C3-B-natural2.

The extended cadential function with chromaticism and prominent cadential 6/4s sounds a bit old-fashioned and dramatic, as if it belonged to a menuet or a purely instrumental piece -- not much like "little genii" floating just above ground. Given that the first strain abuts four bars of Ländler to four bars of horn calls, perhaps Schumann disliked the topical chaos.

It is more difficult to guess what Schumann objected to in n35--the design is certainly as straightforward as it could be, and the direct chromatic shift to begin the second strain and the "falling fourths" progression that opens it are hardly uncommon in Schubert's dances. Here again, I will guess that the issue was topical dissonance: the lilting violinistic figures of the Ländler style are placed in a rather low register. Where D365n2 (the Trauerwalzer) lifts its figures out of this register to end with clear Ländler figures in the next octave, and so neatly contrasts the chromatic (lower) with the diatonic (higher), here the register is maintained throughout.

In the manuscript that includes the F major waltzes, all are in F# major. These have the earliest date (March 1821; D365 was published in November that year) and are undoubtedly the "originals" -- that is, the keys Schubert would most often have played these dances in. The F# major versions without question lie better under the hands -- some places in the published versions are so awkward as to be nearly unplayable. It's unlikely Schubert himself would have made these kinds of clumsy literal transpositions -- and so perhaps what Schumann was objecting to, unbeknownst to him, was publisher's errors and not Schubert's.

The last waltz uses a texture that is very rare in Schubert's dances: melody with block-chord accompaniment that virtually erases the dance and moves the music toward song instead. These block chord textures, often with the half+quarter rhythms found here, are used occasionally for strains or whole dances that emphasize tonic pedal points (and especially in the minor key).

Note added 12-24-09, cited from Notley, 140: "Schubert entered these five dances as F sharp major "Deutsche Tänze" in his autograph, but they came out as F major waltzes in [D365]. The dances, which bear the date March 8, 1821, play with the possibilities of modal mixture. . . . In each the chromatic inflections underpin his interpretation of the genre's even phrases and divided form."

Litschauer, Walburga. Ed. Franz Schubert. Neue Ausgabe sämtlicher Werke, Series 7, part 2: Werke für Klavier zu zwei Händen, Band 6: Tänze I. Kassel: Bärenreiter, 1989.
Notley, Margaret. "Schubert's Social Music: The 'Forgotten Genres'." In Christopher H. Gibbs, The Cambridge Companion to Schubert, 138-54. Cambridge/New York: Cambridge University Press, 1997.