JMT series, part 4a-1 postscript

Work for yesterday's post about the Scherzo in Beethoven's Second Symphony involved examining the orchestral parts. I found that the upper winds "overshot" ^8 in the final cadence, complicating my reading of a simple rising line (those "extra" notes had been deleted from the piano reduction I relied on during research for the 1987 JMT article).

Having found that, I decided to re-examine some of my analyses of Haydn symphony third movements. Symphony 100 produced some interesting results. Here is the original post: link.

In the original post I noted that the inverted arch shape of the opening melody worked against a rising line, but the orchestration in fact plays on a low-then-high registral pairing throughout that supports the rising line at a higher level.

In the A-section, the flute and the first violins begin in the same octave -- circled below -- but in the re-orchestrated repeat (bars 9-16), the flute plays an octave higher -- circled notes in bars 9-10; see also bars 14-16 in the second system below.

The upper winds rejoin the first violins in the B-section -- boxed notes in bars 17 ff above. This holds till the stop on V in bar 28 -- see boxed notes below.  After that an interesting wedge figure brings out the registral differences as the flute moves chromatically down from D6, then returns to it -- circled notes and line --  while the first violins (and first oboe) rise from D5 before likewise returning to where they started.

The reprise is 8 bars rather than 16 and it combines the orchestrations of the two versions from the A-section: brass and timpani play as in bars 1-8 while the strings and winds play as in bars 9-16, except for the addition of the persistent rising figure (boxed) that motivically connects the ends of the first and second phrases and brings particular clarity to the flute's upper-register scale in the structural cadence.

JMT series, part 4a-2 (simple rising lines)

In note 28 I wrote "Other pieces that use the simplest form of the rising Urlinie include the following (qualifying comments in parentheses)."

Among the pieces named was Haydn, Symphony no. 104, III. It's not quite as simple as I claimed, however—there is a drop from ^6 down to ^7, which also happens in the menuet of Symphony no. 86 (as discussed in my essay Ascending Cadence Gestures: A Historical Survey from the 16th to the Early 19th Century, published on Texas Scholar Works: link). About Symphony 86 I wrote:

This time [in the reprise] B5 drops to C#5-D5 for the cadence. The end result is a "circle" of sorts, from D5 back to itself, but by means of an octave's worth of a scale. This device of undercutting the rise from ^6 to ^7 is discussed in my JMT article and seems to be particularly characteristic of the later 18th century. To speculate: the conventions associated with the dominant Italian style (which we know better nowadays through research on the partimenti, evidence of methods of instruction) were so strong that Haydn felt an obligation to observe them in some situations, rather than take full advantage of the rising cadence gesture. In any case, the leap downward from a subdominant to the leading tone is very expressive in and of itself.

The key is the same in Symphony no. 104, ^5 is as firmly settled as the tonic pedal note underneath it, and a string of parallel sixths lead the melodic line down to the cadence. Only the sforzando on the last beat of bar 6 suggests anything different: B5 sticks out above, then leaps down to the dominant's C#5 (see the box).


What that sforzando hints at it is the possibility of a rising line from A5, but, as happened in Symphony no. 86, directionality is undermined by curling back to the lower octave instead of rising toward C#6 and D6.

As is well known, Haydn can't seem to leave things alone in a reprise, and the effects can easily be seen even in design features like linear patterns. In the A section, the eight-bar theme is repeated (in different instrumentation). In the reprise, the theme statement makes it through six bars before changes start, the overall result being an extension of the continuation phrase from four bars to eleven, including two bars of grand pause (!), and a clearly profiled stepwise ascent from B4 through C#5 to D5 (see the second system below). The codetta adds a little flourish that gives us C#5-D6 at last.

This condensed version shows just the ^5-^8 progress over the course of the reprise.

Haydn, Part 5

Another piece from footnotes to my "Ascending Urlinie" article: Haydn, Symphony no. 104, menuet. This piece is among those I argued use "the simplest form" of the rising background line (fn28). It's not quite that simple, however—the same drop from ^6 down to ^7 that we found in the menuet of Symphony no. 86 in an earlier post. About that one I wrote:

This time [in the reprise] B5 drops to C#5-D5 for the cadence. The end result is a "circle" of sorts, from D5 back to itself, but by means of an octave's worth of a scale. This device of undercutting the rise from ^6 to ^7 is discussed in my JMT article and seems to be particularly characteristic of the later 18th century. To speculate: the conventions associated with the dominant Italian style (which we know better nowadays through research on the partimenti, evidence of methods of instruction) were so strong that Haydn felt an obligation to observe them in some situations, rather than take full advantage of the rising cadence gesture. In any case, the leap downward from a subdominant to the leading tone is very expressive in and of itself.

The key is the same in Symphony no. 104, ^5 is as firmly settled as the tonic pedal note underneath it, and a string of parallel sixths lead the melodic line down to the cadence. Only the sforzando on the last beat of bar 6 suggests anything different: B5 sticks out above, then leaps down to the dominant's C#5 (see the box).

What that sforzando hints at it is the possibility of a rising line from A5, but, as happened in Symphony no. 86, directionality is undermined by curling back to the lower octave instead of rising toward C#6 and D6.

As we have found more than once already, Haydn can't seem to leave things alone in the reprise, and the effects can easily be seen even in design features like linear patterns. In the A section, the eight-bar theme is repeated (in different instrumentation). In the reprise, the theme statement makes it through six bars before changes start, the overall result being an extension of the continuation phrase from four bars to eleven, including two bars of grand pause (!), and a clearly profiled stepwise ascent from B4 through C#5 to D5 (see the second system below). The codetta adds a little flourish that gives us C#5-D6 at last.

This condensed version shows just the ^5-^8 progress over the course of the reprise.

Haydn, Part 4

 Today, another piece mentioned in footnotes to my article "The Ascending Urlinie" (Journal of Music Theory 1987): the menuet in Haydn's Symphony 100. In the article I wrote that motivic foregrounding and layering did not necessarily generate rising background lines, and in note 28 said this:

The Menuet of Haydn’s Symphony No. 100 is a case in point. In the first period (measures 1-8, which stand for the whole), the initial motion is strongly downward, but the final cadence produces a clear ascent from ^5 to ^8 in the upper-most part.

Thinking of this in Schenkerian terms—as I was in 1987—the rising line is not workable in the theme's first presentation because it doesn't mesh well with the bass, especially in bars 5-6, where one would have to imagine a doubling of bass and soprano, never a good idea. It's much easier to build a line in this way: D5 initiates a fifth-line; to C in bar 4, recapture C in bar 6, B on the last beat of that bar, then A in bar 7, and an implied G in bar 8. The ascending scale in the cadence is boundary play. See this version here:

In the reprise, on the other hand, the chromatic passing tone D# in the bass (from m. 6) is gone, and a string of diatonic figures, all rising, take over the lower parts, directly linking the chromatic scale fragment to the diatonic scale fragment (see the arrows in the figure below). As a result, the rising line from ^5 to ^8 has a clear path and pitch design can be read as well-matched to the important aspects of expression.

Nevertheless, nowadays I think that octave shapes work just as well as lines to describe the frame of this theme. In the first phrase, the ornamented arpeggiation that brings G5 down to G4 is only answered meagerly by the rising chromatic scale in bars 3-4. The second phrase does better, as A5 to A4 is answered by the diatonic scale that brings the close back to G5.

Haydn, Part 3

The menuet of Symphony no. 96 (1791) is a counter-example. Where the rising line was the primary figure in Symphony no. 86, III, as we saw in a previous post, and is eventually connected to a rising cadence gesture, in Symphony no. 96 the promise of same is not realized. In fact, Haydn goes out of his way to undermine (more like demolish) it.

The opening figure is more arpeggio—a "rocket"—than line, but it does establish A5 by the end of the phrase (bar 4). The primary cadence of the first strain, however, drops down the octave to close on B4-A4 (circled notes). (These examples, btw, are from a piano four-hands edition; I couldn't find a two-stave reduction.)

Below is the principal cadence in the reprise: it's down, down, down in all parts but the bass. The codetta, at least, does make a small effort to compensate, but there is nothing unusual about it: an "up and out" flourish in the final seconds is very common in the later 18th and early 19th centuries, so much so as to be a cliché for opera overtures, scenes, and arias (where the orchestra provides the flourish after the singer finishes).

Haydn, part 2: postscript

In yesterday's post about the menuet from Symphony no. 86, I observed that the line from ^1 to ^5 (D5 to A5) in the first strain is pushed "one step too far" to B5 before settling on A5 in the cadence. Here is the example:

This "one-too-far" figure has its roots in 17th century improvised embellishment practices. Here is a simple example adapted from my van Eyck series on this blog: link. The upper staff is the ending of the original tune ("Wel Jan wat drommel"), the lower staff the equivalent place in the first variation. The escape tone diminutions are circled. The last of them is not quite a diminution, as van Eyck actually reorders the notes of the original, but the effect is pretty much the same.

In tonal music of the major-minor system, the most familiar—and probably most influential—figure of this type involves scale degree ^6. In example (a) below, the motion from the consonant A through a passing tone G to a consonant F# is embellished with an escape tone B. This is rather mild business, of course, as the B is consonant with the pedal base D. Even in my rather Brahmsian version, with its third and octave doublings, the effect is sweet rather than dissonant. In example (b1) the underlying voice leading pattern is shown, this time with a change of bass, however. It's this version—embellishment of V rather than I—that is commonly found throughout the century from roughly 1770 to 1870—see example (b2) for the figure with escape tone. Examples (c1) and (c2), then, show two versions with full harmonies.

The escape tone figure was one of the most important enablers of the dominant ninth chord. All it took—as Schubert and others in his generation discovered—was to replace the passing motion with a neighbor figure by resolving ^6 back into ^5 over the chord change. 

Haydn, part 2

As the four-movement symphony model crystallized in the 1770s, the individual movements took on the familiar characteristics we associate with the late 18th century: the first movement an overture, the second an aria, the third a menuet, and the fourth a contredanse (after Leonard Ratner). Of these, the last was the least stable: only in the early to mid-1770s were the contredanses really danceable or recognizable to an audience as programmatic "portrayals" of the dance (I have written about this here: link; others who have written significantly about the two dance movements include Tilden Russell, Sarah Reichart, Wye Allanbrook, and Melanie Lowe). Apart from anomalies (such as fugal movements), by the 1780s finales as dance-finales are perhaps best characterized as overtures utilizing dance topics.

The menuet remained much closer to its dance model. Cast in virtually all instances as a dance with one trio, it was a miniature representation of the actual dance. As many writers have noted, however, the dance itself changed and the music changed with it. In the early part of the century, the menuet of the French court was a couple dance that was meant as a public display of skill and grace. After the death of Louis XIV, it gradually devolved into a perfunctory opening formality for the ball, where it was followed as soon as possible by the lively, very social intercourse of the contredanse, whose musics were almost always gavottes (duple) or jigs (triple).

In Germanophone areas, the formal menuet persisted, but it was joined by a hybrid type that was modeled on the region's "turning" dances (walzen = turning). Haydn was one of the first to exploit this opportunity, and it is no surprise, then, that the violinistic figures of the ländler should find their way into the symphony's third movement, including rising melodic gestures and cadences.

Yesterday we saw one instance of this in the menuet of Symphony no. 83 (1785). Today and in the following several days, we will explore the menuets of later symphonies. In Symphony no. 86 (composed in 1786), Haydn makes the rising gesture the main event, as the line connecting all three of the first strain's four-measure phrases shows (see below). Note that the steady progress from ^1 to ^5 (D5 to A5) is pushed "one step too far" to B5 before settling on A5 in the cadence. That bit of excessive energy has consequences in the reprise.

 As in the opening, the first two phrases of the reprise march upward from D5 to A5, then go through A#5 to B5 in the third phrase. This time, however, B5 drops to C#5-D5 for the cadence. The end result is a "circle" of sorts, from D5 back to itself, but by means of an octave's worth of a scale. This device of undercutting the rise from ^6 to ^7 is discussed in my JMT article and seems to be particularly characteristic of the later 18th century. To speculate: the conventions associated with the dominant Italian style (which we know better nowadays through research on the partimenti, evidence of methods of instruction) were so strong that Haydn felt an obligation to observe them in some situations, rather than take full advantage of the rising cadence gesture. In any case, the leap downward from a subdominant to the leading tone is very expressive in and of itself.

The coda that follows involves some play on the figures we have just heard. The humorous subversion of D5 through C5 (at the fermata) leads the line (fortissimo!) back down to ^5, but then the original cadence is repeated to end, now with a final flourish that gives us ^7 and ^8 in their "correct" register, as C#6 and D6.

Information on French dance practices after the death of Louis XIV came from Richard Semmens, The Bals Publics at The Paris Opera (1716-1763) (Pendragon Press, 2004).

Haydn, part 1

This post begins another series that will provide examples of rising cadence gestures in the later 18th century and early 19th century. In footnotes to my article "The Ascending Urlinie" (Journal of Music Theory 1987), I listed five works by Haydn: the menuets of Symphony 100 and 104, the slow movement of the string quartet Op. 76n2, and movements in two piano sonatas. In this series of posts I will add menuets from three earlier symphonies: nos. 83, 86, and 96.

Let's start with the menuet in Symphony no. 83.

Rising figures appear in both the menuet and its trio. In the former, the first strain suggests the possibility of a rising line (or other figure) that would balance the continual descent in the presentation phrase (bars 1-4), but the continuation phrase doesn't work this out at all clearly.

The reprise is another matter. Although uncertainty still exists about which note in the two-note cells is primary, it is really not all that serious a factor, as one can just build an octave line from G4-G5 if you don't like mine from F#4-F#5 with resolution to G5.

In the trio, B4 in the antecedent phrase starts a very common motion that settles on A4 (as ^2) after touching the upper neighbor C5. In the consequent phrase, C is altered to C# (another common feature) in order to settle on D5 at the end. This is the sort of thing that would be understood as motion to a cover tone, with an interruption (with implied? A4) in Schenkerian analysis.

As in the menuet itself, the reprise of the trio manages the figure a different way, though with no suggestion of an ascending cadence gesture. Here Haydn anticipates many early 18th century waltzes in leaving notes of the dominant chord "hanging" over the final tonic: E5 "might" have gone to D5 [this one is especially important to the waltz], and C5 to B4.