This post picks up the thread of recomposition, now with concrete evidence rather than my speculative rewriting of D365 or my partitionings of D. 779 into small cycles or dance-trio groups. Relatively early during his tenure in Weimar, in the midst of an enormously fruitful period of composition, but already past the years of avid concertizing, Franz Liszt composed a set of nine pieces titled Soirées de Vienne: Valses-Caprices d'après Franz Schubert (S. 427), a still-popular set of piano compositions Humphrey Searle has praised as "among the most successful things of this kind which Liszt ever did" (62). Liszt himself remained fond of the collection to the end: he played numbers from them in concerts in 1873 and 1886; number six (of interest here) was on the program of his final concert, given days before his death in July 1886 (Williams, 480, 671, 680).
These pieces, which meander about the genres of concert waltz, virtuoso vehicle, and character piece, mine two veins of nostalgia–the concert display piece and Liszt's preoccupation with memories of musical life in Vienna and Paris during the 1830s and early 1840s (a preliminary version of one of them was actually written in 1834, only six years after Schubert's death and only three years after Schumann's Papillons exploited Schubert's dances in a slightly different way).
The sixth valse-caprice is the only one of the set to appear in two versions (the second is close to the first but with more virtuosic elaboration). It is essentially a concert transcription of three waltzes, two from Valses nobles and our A Major Waltz. The logic of the choices may be a matter of good affective contrast, but key center could also have been a factor. The majority–but by now means all–of the waltzes that Liszt cites in Soirées de Vienne are in the original key. It is not difficult to imagine the ninth waltz from Valses nobles attracting Liszt's attention:
A deutscher, it is written in A minor, the minor mode being rare in Viennese waltzes, and it clearly invokes "gypsy" rhythms. Furthermore, this A Minor Deutscher opens like a dramatic scherzo–it is barely a waltz at all–and ends with an eight-bar phrase on a Dudelsack (bagpipe) figure.
To this Mephisto-like apparition, Liszt attaches two lyrical trios. The first is the Deutscher's plausible trio–No. 10 in F major–but for the second trio Liszt ignores the two remaining waltzes in A major from the Valses nobles (nos. 2 and 8) and inserts instead D779n13 (this is the only waltz Liszt takes from D. 779 for the Soirées). Thus, the design of the sixth valse-caprice is A (=nobles 9) B (=nobles 10) A C (=sentimentales 13) coda (consisting of cadenza and a reminiscence of B). According to Edward Waters, the holograph manuscript shows that the piece originally ended with the reminiscence and "a simple chordal cadence," the final flourish of triplet eighths being added later (16).
The A Major Waltz is labelled "dolce teneramente" (the two words have been separated in the published edition), and Liszt was insistent that printers observe his articulation symbols in the right hand–a tenuto with staccato dot for the half notes, but tenuto alone for the following quarters; the effect is to create a slight separation between suspension and resolution, thereby giving a slight accent to the latter and emphasizing the two-against-three "polymeter" (Waters, 16). The "dolce teneramente" is visible in the facsimile of Liszt's holograph (Waters, opposite 17).
In this context, it should not surprise us if strange things happen to the A Major Waltz. The large-scale design, if read in a traditional Schenkerian manner, should prolong ^5 (which is unmistakable in the Deutscher) with mixture in the alto as we pass from minor to major; thus, a natural connection is made to the ^5 that will motivate the rising urlinie at the end of the A Major Waltz. Liszt, however, adds a chromatically charged third strain that drags the line down in the most determined way, through a whole-tone scale no less, to C#5, but ends peaceably with a most direct ^3-^2-^1 figure. The entire "newly composed" A Major Waltz is repeated, with elaborating figuration, and a coda follows. Using Schenkerian models, the whole looks as in the graphic below.
What I have just done is to reinterpret one of the earlier readings (the Schenkerian rising line from ^5) in light of the changes of context that Liszt creates. We might do the same for any and all of the readings already offered; for example, we might compare the shapes that arise throughout this more extended composition under the reading that gives priority to melodic shape and multiple structures; and it would be interesting to rethink the metric/rhythmic readings in light of the fact that the introduction never appears as it did in Schubert's waltz (Liszt expands the two-bar introduction to four bars the first time; when the waltz is repeated, he erases the metric problem by starting the triplet figuration on the first beat of the bar, not the second).
Liszt's recomposition of the A Major Waltz, in other words, sets in motion another series of readings, which I will not take the time to follow through here. Yet it also conjures up (even announces in its title) contexts we could follow, as well, if we chose: the waltz as a cultural category, nostalgia for a musical "golden era" of the 1820s and 1830s, the waltz series as concert or salon piece, Vienna, Paris, the early Romantic genius, the sense of historical distance from that genius, changes in Schubert reception in the fourteen years between Schumann's review and the composition of Liszt's piece, etc.
References:
Searle, Humphrey. The Music of Liszt. 2d ed. New York: Dover, 1966.
Waters, Edward N. "Liszt's Soirées de Vienne." The Library of Congress Quarterly Journal of Current Acquisitions 6/2 [1949]: 16-20.
Williams, Adrian. Portrait of Liszt: By Himself and his Contemporaries. Oxford/New York: Oxford University Press, 1990.
