Blurred Harmonies in Robert Schumann and Johannes Brahms

Diego Cubero

Abstract

This article argues that Robert Schumann and Johannes Brahms gave their music a blurry, Romantic quality by distorting well-known harmonic progressions. After identifying the basic progression that underpins the music, the analyses discuss the rhythmic dislocations that cause chords to overlap and merge together. The essay bridges a gap between an important aspect of the Romantic aesthetic and our theoretical understanding of Schumann’s and Brahms’s harmonic vocabulary.

View PDF
Return to Volume 35

Keywords and phrases: schema; harmony; Romantic aesthetics; rhythm; painting; distance.

Suggested Citation

Introduction

Example 1 shows the return of the opening theme in Robert Schumann’s “Warum?” op. 12, no. 3. The passage features a descending circle-of-fifths progression from a B$$\flat$$ dominant seventh chord (VI⁷) to a D$$\flat$$ major triad (I). But the E$$\flat$$ dominant seventh chord (II⁷) within this progression does not emerge as a discrete verticality. A rhythmic dislocation in the bass causes the E$$\flat$$ dominant seventh to appear above the root of the following A$$\flat$$ dominant seventh (V⁷). As the two chords overlap, their notes blend together to form an extended tertian sonority. Schumann depresses the pedal as the two chords overlap, resulting in a rich, smooth blend of the two harmonies. The descending-fifths progression becomes blurry precisely when the opening theme comes back, causing what is typically a point of harmonic and formal clarity to appear out of focus.

Passages like this are characteristic of the music of Robert Schumann and Johannes Brahms. In every case, a rhythmic dislocation causes the chords to spread completely over one another and form illusory tertian sonorities, making the underlying progression sound blurry. The underlying progression is always a stock progression, typically a descending-fifths sequence. The chords generally overlap and blur together at the beginning and end of phrases, and frequently at the point of recapitulation.

This article examines Schumann’s and Brahms’s practice of blurring chords together. It does so first from a historical, then from a structural, and finally from a semantic perspective. The musical examples are roughly organized in order from those where the harmonic blurring is more localized to those in which it is more widespread, culminating with an extensive analysis of the second movement from Brahms’s Clarinet Sonata in F minor, op. 120, no. 1. The findings suggest that by blurring the harmonies in their music Schumann and Brahms romanticized the basic harmonic schemas they inherited from past traditions and imbued them with the quality of something experienced from afar rather than in its immediacy.

1. Historical Context

This section draws a connection between Jean Paul’s definition of the Romantic as boundless beauty, the blurry backgrounds of landscape paintings, and the view that Schumann’s and Brahms’s works share a misty, Romantic quality. I note that there are rhythmic dislocations in their works that prevent harmonies from emerging clearly and propose that the frequency of these passages in their piano works may help explain why the misty quality of their music was often associated with their playing style. 

Jean Paul’s definition of the Romantic as boundless beauty appears in a well-known passage from his Vorschule der Ästhetik ([1804] 1973) that contrasts the clear contours of the Classical with the blurry outlines of the Romantic. Jean Paul observes that “a statue with its sharp and precise contours excludes everything Romantic; painting approximates it in groups of human figures and reaches it in landscapes without men, like those of Claude Lorrain” (60). Similarly, he argues that with its walls and clear-cut hedges “a Dutch garden seems the very opposite of Romanticism, but an English garden which extends into an undefined landscape can surround us playfully with a Romantic region, that is, with the background of an imagination left free to wander in the beautiful” (60). “The Romantic,” Jean Paul concludes, “is the beautiful without boundaries [Begrenzung], or beautiful infinity, just as there is sublime infinity” (61).¹ In so defining the Romantic, Jean Paul problematized the most basic distinction in eighteenth-century aesthetics.² Whereas Kant ([1790] 1951, 81) distinguished the beautiful from the sublime by arguing that the former “concerns the form of the object which consists in having [definite] boundaries [Begrenzung]” while the latter “is to be found in the formless object in so far as in it or by its occasion boundlessness is represented,” Jean Paul blurred the boundary between these two categories, illustrating the very idea he sought to define. If Romanticism is about blurring boundaries, boundless beauty is both definition and instantiation of the Romantic.

For Jean Paul few things were as Romantic as landscape paintings. As he put it, “Painting, like poetry, is much closer than sculpture to romantic infinity and in landscapes often blurs into it completely” (Jean Paul 1973, 57). The painting reproduced in Example 2 shows why Jean Paul cited the landscapes of Claude Lorrain as examples of Romanticism. In this painting, as in many of Lorrain’s pieces, nature engulfs a classical-style building, opening its closed shape into an infinite horizon where land, water, and sky merge together.³

The blurry backgrounds of Lorrain’s landscapes exerted a powerful influence on the artists of the Romantic generation, most notably Caspar David Friedrich. As Linda Siegel (1980, 14) notes, “The revolutionary landscapes and seaport scenes of Lorrain contain many artistic principles which form an important part of Friedrich’s style: melting distances where water and sky merge as one, the portrayal of a dreamlike world beyond, seascapes in which morning light dissolves into mist.” 

The “melting distances” that Siegel describes can be seen in Friedrich’s winter landscape reproduced in Example 3. As is typical of Friedrich’s paintings, in this piece the clearly delineated foreground dissolves into the misty background: the pine trees and the cross map into the spires of the blurry, Gothic cathedral, and the rock is figuratively transformed into its foundation by evoking the well-known passage in the Bible where Jesus tells Peter, “Upon this rock, I will build my church” (Matthew 16:18).⁴ In the distance everything becomes blurry; in distance everything becomes Romantic.⁵

Listeners found this same blurry, Romantic quality in the music of Schumann and Brahms, two composers who were familiar with Jean Paul’s writings and his view of Romanticism as boundless beauty.⁶ Among the first to write about the blurriness of the Schumannian style was Franz Brendel. In his first issue as the editor of the Neue Zeitschrift für Musik, Brendel paid homage to his predecessor, likening his music to landscape paintings:

Schumann’s compositions can often be compared with landscape paintings in which the foreground gains prominence in sharply delineated clear contours while the background becomes blurred and vanishes in a limitless perspective; they may be compared to fog-covered landscapes from which only now and then an object emerges glowing in the sunlight. Thus, the compositions contain certain clear primary sections and others that do not protrude but rather serve merely as backgrounds. Some passages are like points made prominent by the rays of the sun, while others vanish in blurry contours. To this inner peculiarity corresponds the exterior one that Schumann is fond of playing with the constantly depressed pedal, so that the harmonies do not emerge with particular clarity. (Brendel [1845] 1994, 322–323. Translation modified by Hoeckner 2002.)

Shortly after Schumann’s essay “Neue Bahnen” announced Brahms to the world, Hoplit published an article on the young composer in the Neue Zeitschrift at Brendel’s request. Hoplit (1855, 162–3; my translation) observed that Brahms is of a “Schumannian nature,” for he, too, “has the longing for the boundless and misty, which characterizes the Romantics in such a unique way.” Nearly a century later, Karl Geiringer (1947, 337) recognized this longing for the boundless and misty as a central characteristic of Brahms’s entire output: “The young man adored Jean Paul Richter and Novalis . . . and wrote melodies which were incomparably tender, sweet and pensive. Later on, these traits became less obvious; always, however, Brahms loved the vague twilight moods and unreal, ghostly backgrounds.”

Listeners often associated the misty quality of Schumann’s and Brahms’s music with the composers’ playing style. As quoted above, for Brendel the blurriness of Schumann’s music resonated with his tendency to play with an open pedal so as to prevent the harmonies from emerging clearly. Oswald Lorenz similarly noted Schumann’s “almost uninterrupted, yet discreet use of pedal such that no disruptive sound of a heterogeneous harmony arose” (Jansen 1883, 72). As for Brahms, Heinrich Schenker ([1912] 1992, 14) reported that people still spoke of his “blurry type of playing,” and Reimar Riefling ([1957] 1962, 6) claimed that the Viennese public could have attested to the composer’s “blurred pedaling.”

The works of Schumann and Brahms offer some evidence of this type of pedaling. In the ending of Papillons, op. 2, Schumann depresses the pedal continuously for twenty-seven measures, causing the oscillating tonic and dominant harmonies to blur together over a long-sustained bass note. Likewise, in the penultimate piece from Schumann’s Davidsbündlertänze, the sustained bass and widely spaced texture force the music to “swim in pedal . . . blurring tonic and dominant harmony together in a single mist” (Rosen 1995, 27). Here, as in Papillons, the overall effect “easily compares to Brendel’s image of landscapes with blurred harmonic backgrounds” (Hoeckner 2002, 82). Brahms’s blurry style of playing is in evidence in the opening of his B-Minor Intermezzo, op. 119, no. 1, for here every note is held as if with an open pedal so as “to blur the various harmonies in each bar” (Beller-McKenna 2004, 6).

But Schumann’s and Brahms’s tendency to blur chords together went well beyond an idiosyncratic use of the pedal.⁷ An examination of their works reveals numerous passages like that from Schumann’s “Warum?” discussed above (Example 1), where the harmonies overlap and merge together as a result of a rhythmic dislocation, causing the underlying progression to sound blurry. Given that these passages typically involve the piano and often call for the pianist to depress the pedal as one chord overlaps the next, one can assume that these passages are in part what led many writers to describe Schumann’s and Brahms’s playing as blurry and to associate the blurriness of the harmonies with the sound of the open pedal. The use of the pedal in these passages helps create a rich, hazy blend of the overlapping chords, but it is often a rhythmic dislocation, and not a continuously depressed pedal, that causes the chords to overlap in the first place.

Despite the various references to the blurriness of Schumann’s and Brahms’s music, the composers’ use of rhythmic dislocations to blur chords together has received little analytical attention. One of the few discussions of this phenomenon is that offered by Ryan McClelland (2012) in his analysis of a sequential passage near the end of the first movement of Brahms’s E$$\flat$$-Major Sonata for Clarinet and Piano, op. 120, no. 2. McClelland identifies the descending-fifths sequence that underpins the passage and then shows how a rhythmic displacement in the outer parts “blurs the harmonic changes and creates illusory tertian sonorities” (McClelland 2012, 179). The analysis is detailed and convincing, but the discussion of the “harmonic blurring” of the passage is necessarily ad hoc, for the focus of the study is on sequences rather than on the issues of blurring (179).

The following sections offer a broader and more systematic examination of Schumann’s and Brahms’s use of rhythmic dislocations to blur the harmonies in their music. After developing a theoretical framework through which to conceptualize the idea that chords can blur together as they overlap, I will analyze various passages where the harmonies emerge in varying degrees of clarity by first identifying the underlying progressions and then discussing the rhythmic dislocations that cause the chords to overlap and merge together (this is the same analytical methodology used by McClelland to discuss the sequential passage from the Clarinet Sonata). The underlying progression is always a common harmonic schema and often a descending-fifths progression. The harmonic blurring tends to occur at the beginning and end of phrases, and frequently at the recapitulation.

2. Harmonic Blurring

In Schumann’s and Brahms’s music some chords emerge distinctly while others blur together, distorting common progressions. How clear or blurry the chords appear depends largely on two factors, one rhythmic and the other harmonic. As shown in Figure 1, when it comes to rhythm the main question is to what degree the chords overlap. Some chords do not overlap at all, each setting itself sharply apart from the next. By extending the notes of a chord through part of the duration of another, anticipations and suspensions/retardations start to soften the chord boundaries while still allowing the chords to emerge as discrete verticalities.⁸ Further along, chords start to spread completely over one another, each precluding the other from emerging as a clear verticality. As shown in Figure 2, in these cases, the chords typically appear a third, fifth, or seventh apart, one over the next, so that, when they overlap, their notes merge into a stable, albeit illusory, tertian sonority.⁹ This is an important factor, for otherwise the overlap would create a sharp conflict rather than a seamless blend of the two harmonies.

We can start fleshing out the basic outline given in Figure 2 by examining the varying degrees of clarity with which the harmonies emerge in the “hazy, broken waltz” (Robert Hatten 2014, [24]) that ends Schumann’s Davidsbündlertänze. Example 4 shows a hypothetical version of the waltz above the actual music. The hypothetical version clarifies the underlying progression. The annotations in the actual music highlight the rhythmic dislocations that cause some of the chords to blur together.

As shown in Example 4a, the waltz begins with a V⁷–I anacrusis. But the V⁷ never emerges as a separate verticality. As Example 4b illustrates, the V⁷ blends with the root of the following tonic triad in the form of an eleventh chord.¹⁰ Here, as in the passage from “Warum?” discussed above, Schumann depresses the pedal as the two harmonies overlap, resulting in a rich, hazy blend of the two harmonies.

Example 4. Schumann, Davidsbündlertänze, no. 18, mm. 1–20.

After “the haze of the bichordal introduction” (Hatten 2014, [22]), the harmonies emerge in sharply delineated contours (there are no suspensions or anticipations), only to start overlapping once again during the descending-fifths sequence that ends the first phrase.¹¹ Example 4a clearly depicts the sequential repetition of mm. 6–8 down a step in mm. 8–10. But as shown in Example 4b, the sequential repetition in mm. 8–10 becomes distorted as the melody starts lagging behind the accompaniment, causing the notes of one chord to spread through part of the duration of the next. The melodic parallelism renders these overlaps more blurry than a typical set of suspensions: the A in m. 8 and the D in m. 9 appear where one would expect to hear, by comparison to m. 7, passing tones rather than chord tones, and the C in m. 10 appears where one would expect to find, by comparison to m. 8, a chord tone rather than a suspension.¹² The rest that separates the C from the B in m. 10 underscores the ambiguous nature of the C. The melodic parallelism in mm. 6–10 thus prevents the harmonies in mm. 9–10 from emerging convincingly.¹³ Overall, as the annotations at the bottom of Example 4b illustrate, the opening theme follows a trajectory from blurriness to clarity to lesser clarity.

The harmonies of the waltz blur together once again at the boundary between the digression and the recapitulation. As the brackets above the staff in Example 4a illustrate, the digression ends with a descending sequence of dominant-to-tonic progressions. But as shown in Example 4b, the V at the end of this sequence (m. 17) does not emerge as a separate verticality. The V arrives in m. 17 with a dissonant C in an inner voice. The C is set up as a 4–3 suspension. But instead of resolving the C to a B, Schumann repeats the I–II⁷–V motion heard in mm. 17–18 twice more in mm. 18–19. The V in m. 18, like that in m. 17, has a suspended C that does not resolve. When the missing B of the V chord finally materializes in m. 19, the V blends with the root of the following tonic triad to form the eleventh chord of the opening anacrusis. The way in which Schumann problematizes the resolution of the C to the B over the V chords in mm. 17–19 recalls the end of the previous phrase (m. 10), where the C appeared above the V and the resolution of the C to the B was problematized by the intervening rest and conflicting contextual clues.

For our first example of blurry harmonies in Brahms let us turn to the opening phrase of his B-Minor Intermezzo, op. 119, no. 1. Countless authors have drawn attention to the blurriness of the descending-fifths sequence in mm. 1–3, citing the long, sustained chains of downward thirds as the source of the effect.¹⁴ What has been overlooked is the role that the rhythmic displacements between the melody and the accompaniment play in how clear the harmonies appear at the beginning and at the end of the phrase.

As shown in Example 5a, mm. 1–5 feature a descending-fifths sequence with a 7–10 outer-voice pattern. But, as shown in Example 5b, the chords in this sequence do not emerge clearly. The first pair of chords of the sequence (m. 1) merge together completely: the root, third, and fifth of the tonic triad spread over the following E seventh chord, each note being held as with the pedal. And, as the diagonal lines on the staff show, the A in the melody, which transforms the opening triad into a seventh chord, arrives above the E seventh chord in the left hand, and resolves to a G, which belongs conceptually to the E seventh chord but appears instead as part of the downward arpeggiation of the A dominant seventh chord in m. 2. The rhythmic displacement between the melody and the accompaniment shown by the diagonal lines in mm. 1–2 has led to different parsings of the descending-fifths sequence: some analyses extend the E seventh chord (IV) through the beginning of m. 2 while others place the A dominant seventh chord on the downbeat of this same measure.¹⁵ In my own reading, the overlapping roman numerals below the staff show the harmonies in mm. 1–2 spreading one over the next and blending together as a result of the aforementioned rhythmic displacement.

Example 5. Brahms, Intermezzo in B Minor, op. 119, no. 1, opening phrase.

In mm. 2–3, the melody and the bass move closer into alignment. The sustained chains of descending thirds cause the root, third, and fifth of the first chord of the measure to continue to spread over the notes of the second chord, but, in the melody, the seventh of the first chord now arrives and resolves on time, allowing the 7–10 pattern, along with its underpinning sequence, to emerge more clearly. The descending-fifths progression becomes even clearer in m. 4, for here the C$$\sharp$$-minor seventh chord overlaps the F$$\sharp$$ dominant seventh chord only partially through a pair of suspensions.

The harmonic progression becomes blurry again by the end of the phrase. Example 5a illustrates the descending-fifths sequence and 6–5 outer-voice pattern that underpins mm. 7–8. But, as shown in Example 5b, in the actual music the A and the G in the melody are held for too long, spreading the first chord of each measure into the second and distorting the basic 6–5 pattern. Overall, as the annotations at the bottom of Example 5b illustrate, the opening phrase of the Intermezzo follows a trajectory from blurriness to greater clarity to lesser clarity, the same basic trajectory found in the opening phrase of Schumann’s waltz discussed above.

All of the examples discussed so far involve V–I progressions and descending fifth sequences (this includes the passage from Brahms’s E$$\flat$$-Major Clarinet Sonata analyzed by McClelland and the passages from Schumann’s Papillons and Davidsbündlertänze discussed in the previous section). Blurry V–I progressions and descending-fifths sequences are especially common on account of the ease with which each chord can merge with the next. But Schumann’s and Brahms’s practice of allowing chords to overlap and blend together went beyond these two progressions. The following analysis of the statement and recapitulation of the opening theme of Brahms’s Intermezzo in A Major, op. 118, no. 2, illustrates just how varied the overlaps can be in this repertory.

Example 6 analyzes the opening period. Example 6a provides a chordal reduction of the theme without any rhythmic displacements; 6b adds the rhythmic displacements that distort this progression; and 6c reproduces the actual music. The chords in the antecedent phrase all appear as distinct verticalities. The only exception is the A-major chord in the upbeat to m. 2. As Example 6b illustrates, the re-articulation of the opening tonic chord in the upbeat to m. 2 overlaps with the preceding IV$$6\atop4$$ chord, resulting in an illusory D-major seventh chord. Following this overlap, a set of anticipations softens the chord boundaries (see Example 6c), but each chord emerges as a clear verticality.

Example 6. Brahms, Intermezzo in A Major, op. 118, no. 2, opening period.

The chords in the consequent phrase emerge less clearly. As the brackets below Example 6a illustrate, the second half of the consequent repeats the progression heard in the second half of the antecedent, but in the key of E major.¹⁶ The harmonic progression in question consists of two common schemas: a passo indietro (IV–V$$4\atop2$$–I⁶) followed by a cadential progression (II–V in the antecedent, II⁶–V⁷–I in the consequent). In the antecedent the chords emerge clearly; in the consequent they do not. As Example 6b illustrates, in the upbeat to m. 7 the melody re-articulates the C$$\sharp$$ of the upbeat A-major chords (now the A-major chord functions as IV in the key of E major) above the following B dominant seventh chord (V$$4\atop2$$ in E). In m. 7, the D$$\sharp$$ of the V$$4\atop2$$ spreads over the following I⁶ to form an illusory III⁷, and then the E of the of the I⁶ blends with the following II⁶ to form a II$$6\atop5$$. These overlaps distort the passo indietro that was so clearly heard at the corresponding point in the antecedent phrase, while also allowing the music to reach the PAC in m. 8 in a timely manner (rather than a beat late, as is the case in Example 6a) by compressing four harmonies into three verticalities.

To be sure, the melody in the second part of the consequent phrase could be understood as consisting of a set of retardations that resolve in the last sixteenth-note of every beat. But as shown in Example 6c, the melodic and harmonic parallelism between the two phrases colors the notes that appear on the beat as chord tones rather than as retardations, and the sixteenth notes as anticipations rather than as chord tones, thus precluding the harmonies from emerging convincingly.

The harmonies overlap and blur together at various other moments in the Intermezzo.¹⁷ Most notable among these is the retransition that leads us from the F$$\sharp$$-minor middle section back to the recapitulation of the opening theme. Example 7a shows the basic harmonic progression that underpins this passage. The progression consists of an outer-voice descent in parallel sixths from a IV⁶ through a I$$6\atop4$$ (passing) to a II$$6\atop5$$ (chromaticized) that leads to a V⁷ and then to a I. A set of 7–6 suspensions embellish the parallel sixths. This progression, with its characteristic outer-voice motion in sixths, is a common Baroque cadential schema.¹⁸ The use of this schema in a work of the late nineteenth century is noteworthy, but more remarkable still is the way in which it is stylized. As Example 7b illustrates, the chromaticized II$$6\atop5$$, the V, and the I do not emerge as discrete verticalities. The II$$6\atop5$$ arrives over the root of the following V, and, in turn, the V spreads over the recapitulatory I. As these harmonies overlap, their notes blend together into a set of extended tertian sonorities, making the underlying progression sound blurry.

Example 7. Brahms, Intermezzo in A Major, op. 118, no. 2, retransition (mm. 74–77).

The preceding examples illustrate how rhythmic displacements can cause chords to overlap and blend together, making the underlying progression sound blurry. As shown, these rhythmic displacements distort common progressions (most typically V–I progressions and descending-fifths sequences), and they generally occur at the beginning and end of phrases, including the recapitulation point (see Examples 1, 4, and 7). By affecting phrase boundaries, the overlaps make what are typically moments of formal and harmonic clarity appear out of focus.¹⁹

The examples above help substantiate the connection drawn by Brendel between Schumann’s music and the blurry backgrounds of Romantic landscapes, and they give evidence to Hoplit’s claim that Brahms’s music exhibits a Schumannian tendency toward the boundless and misty. The following section expands on the connection between this music and Romantic landscape paintings by arguing that blurry harmonies help convey a sense of distance in this repertory.

3. Blurriness and Distance

The following passage from Adam Müller’s essay on Romantic landscape paintings beautifully describes the correlation between blurriness and distance typical of this medium:

That which immediately surrounds a person—his cottage, the trees in his garden—all this appears in stark contrast concrete, defined and clear beside the formless, flowing ether; now his eye lifts up, so that it can command a greater distance, and the contours of earthly things become softer, the colors gentler: air and earth seem to run together; and they trade places with lovely intimacy: in the clouds the earth appears to step into the face of heaven, and in the seas and rivers heaven into the face of the earth—and in the farthest vastness the borders trail away, the colors fade into each other, what belongs to heaven, to earth, can no longer be told. Thus, regarded from the stark cliffs of the present, appears a person’s distant, earliest childhood: heaven and earth in near relationship, but the memory of that day monochrome as if weatherworn; so too must the future decline of old age appear to him, for distance depicts equally the beginning and the end: not a collision of the elements, but a gentle marriage. (Müller 1808, 72)²⁰

After noting that the more distant something is, the blurrier it will appear, Müller temporalizes the landscape, comparing the trailing away of borders in the distance to the effects of erosion.²¹ For Müller, then, seeing an object fading into the background of a landscape painting is like seeing it fade into the past, seeing it become a ruin. Blurriness is the defining quality of that which lies spatially and temporally removed.

The music of Schumann and Brahms exhibits a similar connection between blurriness and distance as that found in paintings. Hoeckner (2002, 82–86) has briefly explored this connection in the penultimate of Schumann’s Davidsbündlertänze. In this piece, Schumann brings back the melody from the second dance of the set and marks it “As if from a distance.” Referring to Brendel’s aforementioned comparison between music and landscape paintings, Hoeckner (2002) argues that Schumann creates the effect of distance by letting the harmonies blur together through a continuously depressed pedal.

This section examines the connection between blurriness and distance in three examples where the harmonies blur together as a result of rhythmic dislocations. The examples are organized in order from that in which the harmonic blurring is more localized to that in which it is more widespread. In each case, the chords overlap and merge together as the melody slows down and dies out toward the end of a section. This creates a type of ending where the music seems to fade away in blurry contours, like an object fading in the background of a landscape painting, or a ruin fading away across time.

The first example comes from Schumann’s C-Major Fantasy, op. 17. As is well known, Schumann composed this work to help finance a monument for Beethoven, initially entitling the first movement “Ruins.” This title resonates with reference to the last song from Beethoven’s An die ferne Geliebte (“To the Distant Beloved”), which appears at the end of the movement like a ruin amidst a vanishing canvas—everything points to it and dissolves with it.²²

Example 8 shows the quoted melodies in their original form (a) and as the Romantic ruins they become (b).²³ In Schumann’s Fantasy the quoted melodies appear at first with each chord clearly delineated, much as they do in Beethoven’s song.²⁴ But, as the melody slows down in m. 295, the harmonies begin to spread over one another. In m. 297, the high E spreads a note of the six-four chord into the time span of the V⁷ chord that follows. The E moves down to a D at the end of measure 297 but the parallelism between the various echoes of the quote makes this D sound anticipatory. When the E finally resolves to the D of m. 298, the V⁷ blends with the root of the following tonic triad to form an illusory eleventh chord. And before the D in the melody can resolve to a C, the quote breaks off, leaving the V⁷ and the I blending in the background. Schumann depresses the pedal as the chords overlap and blur together, enhancing the haziness of the passage. By the end of the movement, Beethoven’s song becomes the ruin that the movement’s subtitle alludes to.²⁵ The clear harmonic outlines of the quoted melody appear now faded, as if weatherworn.

Example 8. Quotation of Beethoven’s “An die ferne Geliebte,” op. 98, no. 6, in Schumann’s Fantasy, op. 17, mvt. 1.

Brahms’s song “Abendständchen,” op. 42, no. 1, gives further evidence of the connection between blurry harmonies and distance in this repertory. In this strophic setting of Clemens Brentano’s poem, the flute’s evening serenade appears to come from a time of quasi-canonic antiphony and open-fifth cadences. Like the sounds of the flute dying in the distant horizon, the style of late-Renaissance polyphony heard so vividly at the beginning of each strophe ends up fading in indistinct harmonic contours.

Example 9 provides a chordal reduction and the actual music (reduced score) of the phrase that ends the first strophe and, later, the song as whole. As shown, the voices begin the phrase by clearly articulating each chord, first antiphonally and then homophonically. Tenors and basses evade the cadence in m. 15 by repeating the cadential melody (marked with brackets in the example). The upper voices join in antiphonally and in imitation a measure later. As it echoes in the upper voices, the melody slows down and starts lagging behind the other parts, distorting a V–I$$7\atop\natural$$–IV–I progression, an old cadential schema described by Lori Burns (1995, 88) as a “a mixolydian plagal progression.”²⁶

Example 9. Brahms, “Abendständchen,” op. 42, no. 1, mm. 8–20.

The diagonal lines in the example highlight the rhythmic dislocation of the mixolydian plagal progression. As shown, in m. 18 the melody spreads the V⁷ over the root of the I$$7\atop\natural$$ chord in the bass. As expected, the melody resolves the seventh of the V chord to the third of the tonic triad on the downbeat of m. 19, but the resolution of the C to the B (shown by the arrow in Example 9b) appears above a IV chord. The tonic and the subdominant chords in m. 19 merge together to form a IV⁷ sonority, but the B is not treated as the seventh of an actual IV⁷ chord, for it leaps down to a G instead of resolving by step to an A. The sonorities in mm. 18–19 may be better understood as a chain of overlapping harmonies: V⁷ over I$$7\atop\natural$$ in m. 18, and I$$7\atop\natural$$ over IV in m. 19. As the chords overlap, the passage heard back in mm. 14–15 fades in blurred contours. By the end of the phase, the antiphony and the old mixolydian plagal progression dissolve in the distance, like the sounds of the flute in the evening serenade.

A similar treatment of the stile antico can be found in Schumann’s strophic song “Auf einer Burg,” op. 39, no. 7. The first stanza refers to a distant past. The song begins by describing an old knight asleep at his watch, but it is not until the last phrase of the first section that we find out just how old the knight is: he has been asleep for hundreds of years, his hair and beard have grown into one, his heart has turned to stone. It is here that the Renaissance-style polyphony heard so clearly at the beginning of the section starts to fade in blurry contours, the chords merging together like the hair and beard of the centuries-old knight.

Example 10 details the process of erosion that softens the harmonic outlines of this phrase.²⁷ As shown in Example 10a, the phrase begins with an old form of the Monte schema that Gjerdingen (2007, 98) refers to as a Monte Principale,²⁸ and ends with a cadenza doppia, a cadential schema that was “old even in the eighteenth century” (169).²⁹ The progression may be old, but at this stage, with its clearly defined harmonic boundaries, it hides any signs of old age, like a carefully preserved antique.

Example 10. Schumann, “Auf einer Burg,” op. 39, no. 7, mm. 9–18.

Example 10b shows a rhythmic reduction of the phrase in the blurry, weathered state found in the song. As indicated by the diagonal lines, the chords overlap one another largely as a result of a rhythmic displacement where the bass is offset from the melody by half a measure. In the Monte Principale, the metric displacement of the outer parts and the sustained C in the inner voices obscure ever so slightly the identity of the chords. As the phrase continues, the harmonic structure becomes blurrier. In the second half of m. 14, the displacement between the outer parts causes the A-minor chord at the end of the Monte and the D-minor chord projected by the bass to merge together into an extended tertian sonority.³⁰ Two measures later, the II chord in the bass merges with the V that begins the cadenza doppia to form a VII$$^{\circ} {4\atop2}$$.

The melody slows down in m. 16, distorting the cadenza doppia still further.³¹ As its name suggests, the cadenza doppia can be understood as a pair of V–I progressions, with the first of the two tonic chords typically appearing over a dominant pedal as a six-four sonority (see Example 10a). But, as shown in Example 10b, none of these chords emerge with their usual clarity. The G$$\sharp$$ of the V chord spreads in the melody over the C of the tonic six-four chord and resolves to A just as the C of the six-four chord returns to the B of the V chord. The A in the melody resolves down to a G$$\sharp$$ as a 4–3 suspension would, but instead of appearing as part of a distinct V chord, the G$$\sharp$$ spreads over the root-position tonic triad, blending dominant and tonic harmonies into an illusory seventh chord. Whereas the bass cadences in m. 17, the final tonic chord never materializes in the melody, fading in the distance and turning this section into a fragment. Like the stone statue of the centuries-old knight, the stile antico along with its archaic harmonic schemas appear to us as ruins from the past, old and weathered.

Past melodies, antique styles, and old schemas all evoke the past, but it is the blurriness of the harmonies that imbues the preceding examples with a sense of pastness—with the quality of something being experienced from afar rather than in its immediacy. If, as Adam Müller observed, the trailing away of boundaries in distant objects is comparable to the softening of an object’s boundaries caused by erosion, then the old melodies, styles, and harmonic schemas in the examples discussed above may be compared to ruins.

4. Brahms’s Sonata for Clarinet and Piano in F Minor, op. 120, no. 1, mvt. 2

In the slow movement of Brahms’s F-Minor Sonata for Clarinet and Piano, it is no longer the stile antico of Renaissance polyphony but the basic harmonic language of the Galant style that appears in blurry contours, like a faded memory of a bygone time. To be sure, we have encountered examples of blurry Galant schemas in other pieces, but the opposition in this movement between the Galant schemas of the outer periods and the whole-tone sequences of the middle period heightens the correlation of Galant progressions with the past. Brahms’s stylization of the harmonies in this movement supports this correlation, for while the Galant schemas that underpin the outer periods fade in blurry contours, the more modern progressions of the middle period emerge crisp and distinct.

Example 11 shows the decay of the mid-late eighteenth-century harmonic structure that underpins the movement’s opening period (A section). Example 11a reconstructs the basic harmonic progression of each phrase based on the stock progressions they employ, 11b highlights the process of erosion affecting this structure, and 11c reproduces the actual music.

Example 11. Brahms, Sonata for Clarinet and Piano in F Minor, op. 120, no. 1, mvt. 2, opening period.

As the brackets above Example 11a indicate, the antecedent phrase begins by echoing the final statement of the motto heard throughout the first movement: the repeated four-note basic idea of the motto (C–F–E$$\flat$$–D$$\flat$$, C–E$$\flat$$–D$$\flat$$–C) decays to a repeated two-note idea ([C–F]–E$$\flat$$–D$$\flat$$, [C–E$$\flat$$]–D$$\flat$$–C). The resulting $$\hat5$$-to-$$\hat4$$ and $$\hat4$$-to-$$\hat3$$ melodic descent is characteristic of two common Galant schemas, the Sol-Fa-Mi and the Fonte. The Sol-Fa-Mi typically features a I–II⁷–V⁷–I progression and it served as a common opening gambit for slow movements during the second half of the eighteenth century; the Fonte features a sequential VI⁷–II–V⁷–I circle-of-fifths progression and was among the most widely used of Galant schemas (harmonically speaking, the main difference between the two schemas is the first chord: I for the Sol-Fa-Mi, VI⁷ for the Fonte). While the appearance of this melodic descent at the beginning of a slow movement is clearly in dialogue with the Galant use of the Sol-Fa-Mi, the melodic parallelism between the end of the first movement and the beginning of the second movement invites us to hear the melodic descent harmonized more like a Fonte: the F minor and B$$\flat$$-minor seventh chords in mm. 1–2 echo the F dominant seventh chord and B$$\flat$$ minor triad with the added sixth that harmonize the first part of the motto at the end of the first movement.

Following the opening Sol-Fa-Mi/Fonte, an inner voice articulates the characteristic $$\hat1$$-to-$$\hat7$$ and $$\hat2$$-to-$$\hat1$$ (in E$$\flat$$ major) melody of the Aprile schema, a variant of the Meyer schema. Whereas in a typical Aprile an opening tonic-to-dominant progression (harmonizing the $$\hat1$$to-$$\hat7$$ melody) is answered by a dominant-to-tonic progression (harmonizing the $$\hat2$$-to-$$\hat1$$ melody), in Example 11a the first chord of each pair is elided, turning the first note of each of the descending two-note pairs into a suspension.

In mm. 7–8, two more Apriles are joined together to form a Fonte. If the Fonte in mm. 1–4 is not the most common form of this schema, that in mm. 7–8 is a prototypical exemplar: the first half of the schema tonicizes a minor chord (F minor), and the second half tonicizes a major chord a step lower (E$$\flat$$ major), with the Apriles embellishing a set of local $$\hat7$$–$$\hat1$$ ascents in the melody (a common alternative to the descending steps found in mm. 1–4). After the Fonte, the cadential progression is evaded and completed, bringing the antecedent phrase to a close.

The consequent starts with the same basic Sol-Fa-Mi/Fonte schema found in the antecedent phrase (continue to refer to Example 11a). This time the Sol-Fa-Mi/Fonte schema is answered by a 6–6–10–10 double voice-exchange, a schema widely used in the second half of the eighteenth century. The 6–6–10–10 schema is embellished with suspensions, which recall those of the Apriles found in the corresponding place in the antecedent phrase. After the 6–6–10–10 schema, the phrase comes to an end with the same basic cadential progression found in the antecedent phrase.

In Brahms’s Sonata, the Galant schemas identified above appear blurry, their harmonic outlines lost, as if by a process of erosion. Example 11b reveals the rhythmic displacements that cause the chords to spread over one another and merge together. As the diagonal lines illustrate, in mm. 1–4 the bass moves one measure ahead of the melody (the F in the bass is essentially displaced back to the end of the first movement). The dislocation between the melody and the bass causes the chords of the opening Sol-Fa-Mi/Fonte to merge into a series of tertian sonorities: the VI⁷ spreads over the II⁷ in m. 1, the II⁷ over the V⁷ in m. 2, the V⁷ is chromaticized over the I in m. 3, and the I spreads over a newly added VI chord.

Following the blurry Sol-Fa-Mi/Fonte, the Aprile in mm. 5–6 appears in clearer contours, allowing the chords that were blurry in mm. 1–4 to emerge more distinctly. In m. 5, as in m. 1, an E$$\flat$$ spreads over a B$$\flat$$  in the bass, but this time the E$$\flat$$ resolves as a suspension over the B$$\flat$$, allowing the B$$\flat$$ seventh chord (now a dominant seventh rather than a minor seventh chord) to emerge more clearly. Similarly, in m. 6, the B$$\flat$$ chord overlaps the E$$\flat$$ chord, but now the E$$\flat$$ harmony emerges more clearly (albeit in second inversion), following the resolution of the F down to an E$$\flat$$. In keeping with this parallelism, in m. 7 as in m. 3, an E$$\natural$$ appears over an A$$\flat$$ in the bass, tonicizing the following F-minor triad. The F-minor triad, however, is lost amidst the following Fonte-with-Apriles.

In mm. 7–8, as in mm. 1–4, the bass and the melody are out of sync, causing the chords of the Fonte to fade over one another. As the chords in mm. 7–8 fade together, the clear Aprile schema of mm. 5–6 blurs away. Following the blurry Fonte-with-Apriles, the F-minor triad and the B$$\flat$$ seventh chord that were blurry in m. 7 emerge distinctly above the A$$\flat$$ and the B$$\flat$$ in the bass in m. 9, marking the beginning of a clear cadential progression.

The harmonies in the consequent phrase follow a similar trajectory from blurriness to clarity as those in the antecedent (continue to refer to Example 11b). The consequent begins with the same blurry Sol-Fa-Mi/Fonte schema found in the antecedent phrase. The 6–6–10–10 progression that follows also appears blurry. In m. 17, the E$$\flat$$ in the melody spreads above the F in the bass as part of what seems like a typical 7–6 suspension, but the E$$\flat$$resolves an eighth note too late, causing the D$$\flat$$ in the melody to arrive in the time span of the E$$\flat$$ in the bass, as part of the next 7–6 suspension. Similarly, in m. 18, the G in the melody resolves an eighth note too late, causing the F to materialize over the time span of the C in the bass, as part of a 4–3 suspension. As a result, the first chord of each pair overlaps the second. While in the opening progression only one of the four chords emerged distinctly, in the 6–6–10–10 schema two of the four chords appear in clear outlines, creating a trajectory from blurriness to greater clarity that culminates in m. 19 with the clear articulation of the B$$\flat$$ dominant seventh chord. Following the clear arrival of the B$$\flat$$ dominant seventh chord, the harmonies emerge more distinctly than before with no overlaps between them save for the F of the B$$\flat$$ dominant seventh chord, which comes in over the E$$\flat$$dominant seventh chord in m. 19 and 21 as part of a slightly distorted nota-cambiata figure, and for the B$$\flat$$ of the E$$\flat$$ dominant seventh chord, which resolves as a 9–8 suspension over the concluding tonic harmony.

In the actual music the chord overlaps shown statically in Example 11b occur gradually, for the bass notes come in late, halfway through the timespan they occupy. As shown in Example 11c, in mm. 1–4 each bass note appears as the culmination of a downward gesture where the pianist holds each note as if with an open pedal. Each of the downward gestures begins with the chord in the upper voices and ends mid-measure with the bass articulating the next harmony. From this perspective, the syncopated bass notes move ahead of the melody, but they are not anticipatory as Edward Klorman (2014, 129) has argued. Instead of pointing forward to a point of clarity as anticipations do, the syncopated bass notes point back to where they came. Frank Samarotto (2003) has offered a similar reading of the bass line, but whereas he understands the upper voices in mm. 1–4 as suspensions, I hear them as arriving on time, as Klorman does, given that in both a Sol-Fa-Mi schema and a Fonte schema each of the descending steps typically proceeds from an odd to an even measure, just as is the case here. The bass continues to appear late up until m. 9 where the A$$\flat$$ and B$$\flat$$ come in on the beat. The parallelism between the bass line in mm. 7 and 9 confirms that the bass notes in mm. 1–8 appear halfway through their timespans.

Similarly, in the consequent phrase the bass notes come in midway through the timespan they occupy up until the beginning of the cadential progression, where the bass is once again shifted back to appear on the beat. Instead of pointing forward to a point of clarity as anticipations do, the syncopated bass notes point back, befogging what once seemed clear. The result is one where the harmonic boundaries blur as the music unfolds. Hearing the opening theme in this way is to hear the eighteenth-century structure that underpins the music in the process of the becoming a weathered ruin.

If the opening period portrays several Galant schemas faded, like remnants of a bygone past, the contrasting middle period (mm. 23–41) reveals the music from the opening in a new light, transforming the blurry Sol-Fa-Mi/Fonte of the opening period into a whole-tone sequence that by comparison sounds novel and “luminescent.” The transformation of the music from the opening section is apparent when one compares the presentation portions of the two loose-knit sentences that make up the first period (mm. 1–4 and 13–16) with the presentation portion of the two tight-knit sentences that form the middle period (mm. 23–26 and 31–34). As the brackets in Example 12 illustrate, both periods begin by repeating the same basic descending, two-note melodic figure: E$$\flat$$ to D$$\flat$$ and D$$\flat$$ to C in the opening period, F to E$$\flat$$ and E$$\flat$$ to D$$\flat$$ in the middle period. In both cases the melody is treated sequentially, but whereas the sequence in the opening section proceeds diatonically from model to copy (VI⁷–II continues down to V⁷-I), that in the middle section proceeds chromatically in whole tones from model to copy (D$$\flat$$M–A$$\flat$$M continues down to C$$\flat$$M–G$$\flat$$M and then partially to B$$\flat\kern-1.4pt\flat$$M/AM). While the traditional sequence of the opening period appears with the blurriness of something seen from a great distance, the comparatively more modern-sounding sequential progression in the middle section emerges with the clarity of that seen up close. The mapping of the clearly delineated sequence of the middle period onto the blurry descending-fifths sequence of the opening period is comparable to the mapping of the clearly delineated cross and tress onto the blurry, Gothic cathedral in Friedrich’s painting discussed above.

Example 12. Comparison of the presentation portion of the phrases in the opening and middle periods (sections A and B respectively) of Brahms’s op. 120, no. 1, mvt 2.

Whereas the opening and middle sections juxtapose two different perspectives of the same basic thematic idea, the retransition (mm. 41ff) superposes these two perspectives by depicting the more novel harmonic aspects of the music distinctly and the more traditional ones blurrily. What is clearly heard in the retransition is the descending cycle of major thirds: the beginning of the primary theme is repeated every four measures a major third lower, outlining in the melody a whole-tone scale across mm 41–51. What is less clearly heard is the traditional descending circle-of-fifths sequence with its embedded Fontes that underpins this cycle.

Example. 13a shows the circle-of-fifths sequence that connects the tonic chord of the middle section (D$$\flat$$/C$$\sharp$$) back to the tonic chord of the opening section (A$$\flat$$). From this archetypal progression, Example 13b reconstructs a rhythmically normalized version of the actual music. The progression is subdivided into three statements of the opening motto a major third apart (which give rise to the major-thirds cycle), each statement outlining a Sol-Fa-Mi/Fonte progression. The first two statements correspond to the retransition, the last to the recapitulation.

Finally, Example 13c shows how the rhythmic displacement between the bass and the melody blur the descending-fifths sequence that binds together the three statements of the theme. In mm. 41–44, as in mm. 1–4, a VI⁷ chord dissolves above a II⁷, a II⁷ above a V⁷, and the V⁷ above the I (the addition of the lowered ninth to the V⁷ making the Fonte a hermaphrodite type). In keeping with the one-measure displacement between the melody and the bass, one would expect the tonic chord to materialize in the upper voices in m. 44, completing a statement of the opening motto in the key of E major. Instead, the melody dissolves over a long dominant chord without ever arriving on the expected E-major tonic harmony, thereby turning the already weathered Fonte into a fragment. Another fragmented Fonte follows a third lower in mm. 45–48, now with the C-major triad missing in the upper voices. The distorted descending-fifths progression then continues in m. 49 through the recapitulation of the opening period, with the A$$\flat$$-tonic chord that concludes that last Fonte lost over an F in the bass. The trajectory from the middle section, which sounds clear and new, through the retransition, which combines the clarity of the new with the blurriness of the old, culminates with recapitulation of the blurry and old opening section.

Example 13. Brahms, Sonata for Clarinet and Piano, op. 120, no. 1, mvt. 2, mm. 41–52.

The analysis of this movement as a blurry, distant depiction of a mid-late eighteenth-century harmonic structure resonates with Brahms’s view of himself as the last member of a dying musical tradition and with the idea that his late works, in particular, are tinged with elements of reminiscence or nostalgia. While other composers looked toward the dawn of a new age, in this Sonata Brahms portrayed himself with his back turned, as a Rückenfigur, gazing at the last remnants of the common practice style fading away in his own works. And yet, by drawing a clear connection between the blurriness of the outer sections and the clarity of the more modern sounding middle section, Brahms invites us to see the old in the new and the new in the old.

Conclusion

This article has proposed that some chords in Schumann’s and Brahms’s music overlap and blur together, distorting well-known harmonic progressions. As shown, how clear or blurry the progression depends largely on the degree to which its chords overlap: the harmonies may not overlap at all, they may do so only partially through anticipations or suspensions/retardations, or they may overlap completely. (In some cases, the degree to which one chord spreads over the other can be ambiguous, as contextual clues can problematize the distinction between chord tones and non-chord tones.) When the harmonies spread completely over one another and merge together in the form of illusory tertian sonorities, the underlying progression sounds blurry.

As summarized in Table 1, in the eight pieces analyzed above, rhythmic displacements distort common harmonic schemas (descending-fifths sequences being the most common among these). These overlaps generally affect the beginning and ending of phrases, and especially recapitulations, making what are traditionally points of harmonic clarity sound out of focus. The overlaps generally involve the piano, and they often call for the use of the pedal or simulate the sound of the open pedal. This is significant, for it connects with the view that the blurriness of Schumann’s and Brahms’s music is associated with the composers’ playing style and their use of the pedal.

By examining the ways in which Schumann and Brahms distort old stock progressions, the examples also draw a connection between the blurry depiction of the temporally distant in this music and the blurry depiction of the spatially distant in Romantic landscape paintings. In drawing a connection between the blurriness of this music and that of landscape paintings, the goal, however, was not to suggest that the former is imitating the latter, but rather that these two artforms are complementary expressions of the same aesthetic—one which was seen as quintessentially Romantic.

Work	Blurred Progressions	Placement of Blurred Harmonies	Use of Pedal	Associations with Distance
Schumann, “Warum?” op. 12, no. 3.	Descending-fifths	Recapitulation	Pedal is depressed as chords overlap
Schumann, Davidsbündlertänze, no. 18	V7–I, Descending-fifths (Fonte)	Beginning and end of the first phrase, recapitulation	Pedal is depressed as chords overlap	Follows a piece marked “Come da lontano”
Brahms, B-Minor Intermezzo, op. 119, no. 1	Descending-fifths	Beginning and end of the first phrase	Sustained notes simulate the sound of an open pedal	The phrase is discussed by Beller-McKenna as an example of reminiscence
Brahms, A-Major Intermezzo, op. 118, no. 2	Passo indietro, IV6–I$$6\atop4$$–II$$\sharp$$–V7–I	Near the end of the opening period, recapitulation
Schumann, Fantasy, op. 17, mvt. 1	V7–I progression	End of the piece	Pedal is depressed as chords overlap	Quote “To the Distantly Beloved,”  and subtitle “Ruins”
Brahms, “Abendständchen,” op. 42, no. 1	Mixolydian plagal progression  (a variant of the Quiescenza)	End of each stanza		Stile antico, poetry, and use of old harmonic schema
Schumann, “Auf einer Burg,” op. 39, no. 7	Monte principale, cadenza doppia	Each of each stanza		Stile antico, poetry, and use of old harmonic schema
Brahms, Sonata for Clarinet and Piano in F Minor, op. 120, no. 1, mvt. 2	Sol-Fa-Mi/Fonte,  Fonte with Apriles, 6–6–10–10	Blurred harmonies are widespread, but their use is particularly evident at the beginning of the two phrases of the opening period, and at the recapitulation	Sustained notes simulate the sound of an open pedal	Opposition between blurry Galant schemas and clear whole-tone sequences

 In arguing that Schumann and Brahms gave their music its blurry quality by allowing the harmonies to overlap and merge together, this article has bridged a gap between an important aspect of the Romantic aesthetic and our theoretical understanding of Schumann’s and Brahms’s harmonic language. If, as Jean Paul wrote, the “Romantic is the beautiful without boundaries,” to recognize the way in which the chords in the above examples blur together is to hear the harmonic language of these pieces as being Romantic. If the largely diatonic stock progressions played in intimate chamber settings is in part what imparts the examples discussed above their beauty, the way in which their chords merge gracefully into one another imbues this beauty with a sense of boundlessness, creating moments of boundless beauty.

---

Dr. Diego Cubero
Assistant Professor of Music Theory
University of North Texas
Diego.Cubero@unt.edu

References

Aldwell, Edward and Carl Schachter. 2011. Harmony and Voice-Leading. 4th ed. Boston, MA: Schirmer Books.

Beller-McKenna, Daniel. 2004. “Reminiscence in Brahms’s Intermezzi.” American Brahms Society Newsletter 22 (2): 6–9.  

Brahms, Johannes. 2003. The Brahms Notebooks: The Little Treasure Chest of the Young Kreisler. Translated by Agnes Eisenberger. New York, NY: Pendragon Press.

Brendel, Franz. (1845) 1994. “Robert Schumann with Reference to Mendelssohn-Bartholdy and the Development of Modern Music in General.” Translated by Jürgen Thym. In Schumann and his World, edited by R. Larry Todd, 317–337. Princeton, NJ: Princeton University Press.

Brower, Candace. 2008. “Paradoxes of Pitch Space.” Music Analysis 27 (1): 51–106. 

Burns, Lori. 1995. Bach’s Modal Chorales. New York, NY: Pendragon Press.   

Cadwallader, Allen. 1983. “Motivic Unity and Integration of Structural Levels in Brahms’s B-Minor Intermezzo, Op. 119, No. 1.” Theory and Practice 8 (2): 5–24. 

———. 1988. “Foreground Motivic Ambiguity: Its Clarification at Middleground Levels in Selected Late Piano Pieces of Johannes Brahms.” Music Analysis 7 (1): 59–91.

da Vinci, Leonardo. (1651) 1835. A Treatise on Painting. Translated by John Francis Rigaud. London: J. B. Nichols and Son.  

Daverio, John. 1987. “Schumann’s ‘Im Legendenton’ and Friedrich Schlegel’s ‘Arabeske.’” 19^th-Century Music 19 (2): 150–163. 

Diergarten, Felix. 2003. “Melancholie des Unvermögens: Der Brahmssche Ton und das Intermezzo op. 119, 1.” Musik & Ästhetik 26: 42–53.

Geiringer, Karl. 1947. Brahms and His World. Oxford: Oxford University Press. 

Gjerdingen, Robert O. 1988. A Classical Turn of Phrase: Music and the Psychology of Convention. University of Pennsylvania Press.  

———. 2007. Music in the Galant Style. New York, NY: Oxford University Press.   

Hanslick, Eduard. (1862–72) 2002. “Discovering Brahms.” Translated by Kevin C. Karnes. In Brahms and His World, edited by Walter Frisch and Kevin C. Karnes, 217–232. Princeton, NJ: Princeton University Press. 

Hatten, Robert S. 1994. Musical Meaning in Beethoven: Markedness, Correlation, and Interpretation. Bloomington, IN: Indiana University Press. 

———. 2014: “Performing Expressive Closure in Structurally Open Contexts: Chopin’s Prelude in A minor and the Last Two Dances of Schumann’s Davidsbündlertänze.” Music Theory Online 20 (4). 

Hoeckner, Berthold. 2002. Programming the Absolute: Nineteenth-Century German Music and the Hermeneutics of the Moment. Princeton, NJ: Princeton University Press. 

Hoplit (Richard Pohl). 1855. “Johannes Brahms II.” Neue Zeitschrift für Musik 43 (25): 261–264.  

Jansen, Gustav. 1883. Die Davidbündler: Aus Robert Schumann’s Sturm- und Drangperiode. Leipzig: Brietkopf und Härtel.

Jean Paul. 1973. Horn of Oberon: Jean Paul Richter’s School of Aesthetics. Translated by Margaret R. Hale. Detroit, MI: Wayne State University Press.   

Kant, Immanuel. (1790) 1951. Critique of Judgment. Translated by J. H. Bernard. Royal Oak, MI: Hafner Press. 

Klorman, Edward. 2014. “On the Slow Movement of Brahms’s F-minor Clarinet Sonata: Thirds-Cycles, Diatonie, and Todesangst.” Gamut 7 (1): 126–149.

Kuzniar, Alice A. 1989. “The Temporality of Landscape: Romantic Allegory and C. D. Friedrich.” Studies in Romanticism 28 (1): 69–93.

Lerdahl, Fred and Ray Jackendoff. 1983. A Generative Theory of Tonal Music. Cambridge, MA: MIT Press. 

Lester, Joel. 1986. The Rhythms of Tonal Music. Carbondale, IL: Southern Illinois University Press.

Lewin, David. 2006. Studies in Music with Text. New York, NY: Oxford University Press. 

Mahrt, William P. 1992. “Brahms and Reminiscence: A Special Use of Classical Conventions.” In Convention in Eighteenth- and Nineteenth-Century Music: Essays in Honor of Leonard G. Ratner, edited by Wye Allanbrook, Janet Levy, and William P. Mahrt, 75–112. New York, NY: Pendragon Press.  

McClelland, Ryan. 2012. “Sequence as Expressive Culmination in the Chamber Music of Brahms.” In Expressive Intersections in Brahms: Essays in Analysis and Meaning, edited by Heather Platt and Peter H. Smith, 147–185. Bloomington, IN: Indiana University Press.   

Mirka, Danuta. 2009. Metric Manipulations in Haydn and Mozart: Chamber Music for Strings, 1787–1791. New York, NY: Oxford University Press.

Morgan, Robert P. 1999. “6 Piano Pieces, Opus 118.” In The Compleat Brahms: A Guide to the Musical Works of Johannes Brahms, edited by Leon Botstein, 193–195. New York, NY: W. W. Norton & Company. 

Müller, Adam. 1808. “Etwas über Landschaftsmalerei.” Phoebus: Ein Journal für die Kunst 4–5: 71–73.

Newbould, Brian. 1977. “A New Analysis of Brahms’s Intermezzo in B Minor, op. 119, no. 1.” The Musical Review 38: 33–43.

Notley, Margaret. 2007. Lateness and Brahms: Music and Culture in the Twilight of Viennese Liberalism. New York, NY: Oxford University Press. 

Novalis. 2002. Notes for a Romantic Encyclopedia: Das Allgemeine Brouillon. Translated and edited by David Wood. Albany, NY: State University of New York Press.  

Riepel, Joseph. 1755. Grundregeln zur Tonordnung Insgemein. Frankfurt and Leipzig: Lotter, Auspurg. 

Riefling, Reimar. (1957) 1962. Piano Pedaling. Translated by Kathleen Dale. Oxford: Oxford University Press.

Rings, Steven. 2012. “The Learned Self: Artifice in Brahms’s Late Intermezzi.” In Expressive Intersections in Brahms: Essays in Analysis and Meaning, edited by Heather Platt and Peter H. Smith, 19–51. Bloomington, IN: Indiana University Press.

Rosen, Charles. 1995. The Romantic Generation. Cambridge, MA: Harvard University Press. 

Salzer, Felix. 1962. Structural Hearing: Tonal Coherence in Music. Mineola, NY: Dover. 

Salzer, Felix and Carl Schachter. 1969. Counterpoint in Composition: The Study of Voice Leading. McGraw-Hill. 

Samarotto, Frank. 2003. “Treading the Limits of Tonal Coherence: Transformation vs. Prolongation in Selected Works by Brahms.” Paper presented at the Society for Music Theory Annual Meeting, Madison, WI.

______. 2005. “Schenker’s ‘Free Forms of Interruption’, and the Strict: Toward a General Theory of Interruption.” Paper presented at the Annual Meeting of the Society for Music Theory, Cambridge, MA.

Schenker, Heinrich. (1912) 1992. Beethoven’s Ninth Symphony: A Portrayal of Its Musical Content, with Running Commentary on Performance and Literature as well. Translated by John Rothgeb. New Haven, CT: Yale University Press. 

Schumann, Robert. 1853. “Neue Bahnen.” Neue Zeitschrift für Musik 39 (18): 121–125.

Siegel, Linda. 1980. Caspar David Friedrich and the Age of German Romanticism. Brookline, MA: Branden Press.  

Smith, Peter H. 1994. “Liquidation, Augmentation, and Brahms’s Recapitulatory Overlaps.” 19^th-Century Music 17 (3): 237–261.

———. 2005. Expressive Forms in Brahms’s Instrumental Music: Structure and Meaning in his Werther Quartet. Bloomington, IN: Indiana University Press.  

Specht, Richard. 1928. Johannes Brahms: Leben und Werk eines deutschen Meisters. Dresden: Avalum-Verlag. 

Stefaniak, Alexander. 2016. Schumann’s Virtuosity: Criticism, Composition, and Performance in Nineteenth-Century Germany. Bloomington, IN: Indiana University Press.

1. Whereas Margaret Hale translates the word “Begrenzung” as “limits,” I translated it as “boundaries” in order to match J. H. Bernard’s translation of this same word in the passage from Kant quoted later in this paragraph.
2. Berthold Hoeckner (2002, 55) makes a similar point. Kant’s distinction between the beautiful and the sublime was largely based on Edmund Burke’s well-known definitions of these categories.
3. The landscapes of Claude Lorrain employ what Leonardo da Vinci described as “aerial perspective.” Da Vinci ([1651] 1835, 96) notes, “Whenever a figure is placed at a considerable distance you lose first the distinctness of the small parts, while the larger parts are left to the last, losing all distinctness of detail and outline.”
4. The transfiguration of the foreground into the background is an important, though seldom discussed, aspect of Friedrich’s style. For an additional example of this technique, see Friedrich’s 1821 painting Moonrise by the Sea. The seascape transfigures the two anchors on the rocky shore into the two boats that float in the distant waters.
5. As Novalis (2002, 51–52) wrote, “In the distance everything becomes poesy-poem. Actio in distans. Distant mountains, distant human beings, distant events, etc., all become romantic.”
6. Robert Schumann quoted Jean Paul’s definition of the Romantic as boundless beauty in three occasions, once in his Mottosammlung and twice as an epigraph for the Neue Zeitschrift für Musik  (2/25, 1835 and 14/51, 1841). Brahms must have been familiar with Jean Paul’s definition, for he read and carefully annotated the complete works of Jean Paul, which he received as a gift from Clara Schumann in the Christmas of 1854. Brahms (2003) quotes Jean Paul more than any other author but does not quote the passage in question.
7. Franz Brendel’s comparison of Schumann’s music to landscape paintings supports this idea. As quoted above, after noting that some passages in Schumann “vanish in blurry contours,” Brendel notes, “To this inner peculiarity corresponds the exterior one that Schumann is fond of playing with the constantly depressed pedal” (Brendel [1845] 1994, 323; emphasis mine). In other words, the use of raised dampers is only the external manifestation of something that is internal to Schumann’s music.
8. In the case of retardations and anticipations, meter typically allows listeners to infer the boundary between the chords, because, in tonal music, chords generally change at metrically strong points. On the relationship of harmonic rhythm and meter, see Lester (1986, 26) and Mirka (2009, 51).
9. Most commonly, Schumann and Brahms spread a seventh chord above a chord a fifth below, resulting in an illusory eleventh chord. Examples of overlaps that produce ninth chords and seventh chords are also common. Of these three possibilities, eleventh chords represent the richest and haziest of the sonorities produced. Thirteenth chords are rare.
10. The notes of the G dominant seventh chord must not be dismissed simply as embellishments of the tonic chord. The G dominant chord connects, via a chromaticized 5–6 exchange, the B-minor triad at the end of the preceding piece to the C-major tonic triad that emerges as a verticality in m. 3 of the waltz.
11. This sequence is organized as a Fonte schema. As Gjerdingen (2007, 61–72) explains, the Fonte is a common eighteenth-century schema first discussed by Joseph Riepel (1755). This schema is a four-chord, descending-fifths sequence that alternates dominant seventh chords and triads. The schema is divided in two halves. The first half consists of a dominant-to-tonic progression in the minor mode with the melody descending from $$\hat4$$ to $$\hat3$$, and often elaborated as $$\hat6$$–$$\hat5$$–$$\hat4$$–$$\hat3$$. The second half does the same a step lower and in the major mode.
12. Lerdahl and Jackendoff (1983) note that listeners are inclined to hear parallel passages in parallel ways. One of their preference rules states, “Prefer a prolongational reduction in which parallel passages receive parallel analyses” (Lerdahl and Jackendoff 1983, 226). The claim of the above analysis is that, given the parallelism in mm. 6–10, listeners may be inclined to hear the A and D in mm. 8 and 9 as if they were passing tones, and the C in m. 10 as if it was a chord tone. The conflict prevents the harmonies from emerging as clearly as they would if no such conflict existed.
13. One can appreciate the slight obfuscation of this descending-fifths progression (Fonte) by comparing it to the clearer descending-fifths sequence (Fonte) that appears in mm. 22–26, when the opening phrase is repeated.
14. Analyzes of the opening of Brahms’s B-Minor Intermezzo, op. 119, no. 1 can be found in Salzer 1962, Newbould 1977, Cadwallader 1983, Diergarten 2003, Beller-McKenna 2004, Brower 2008, and Rings 2012. Beller-McKenna (2004) and Brower (2008) refer to the blurriness produced by the sustained chains of descending thirds. Brower (2008, 89) writes, “The unbroken succession of diatonic thirds and the sustaining of every note to the end of the bar serves to blur triadic boundaries.”
15. Newbould (1977), Cadwallader (1983), and Brower (2008) extend the E seventh chord through the downbeat of m. 2. Salzer (1960), Diergarten (2003), and Beller-McKenna (2004) place the A seventh chord on the downbeat of m. 2.
16. The consequent’s harmonic progression is interrupted on a V$$6\atop5$$ in m. 6. Following the interruption on the V$$6\atop5$$, the music re-articulates the A-major chord with the C$$\sharp$$ on the melody heard at the start of the phrase. The A-major chord progresses once more to a B dominant seventh (now in 4/2 position), and the B dominant finally resolves to an E-major triad in first inversion, as part of a passo indietro. The apparent motion from the V$$6\atop5$$ to the IV in m. 6 must not be understood as a retrogression, as the V$$6\atop5$$ does not progress to the IV. The interruption of the progression on the V$$6\atop5$$ in m. 6 is an example of what Samarotto (2005), after Schenker, calls a “free form of interruption.”
17. See, for example, mm. 20, 24, and 46.
18. Aldwell and Schachter (2011, 353) show an instance of this progression in J.S. Bach’s chorale “Jesu, Meine Freude,” noting that “a passing $$6\atop4$$ between IV⁶ and II$$6\atop5$$ (less often a II⁶) is very frequent; the outer voice motion is usually in 6ths.” Bach’s chorales offer dozens of examples of this cadential progression. Measures 9–10 of Bach’s chorale no. 80, “O Haupt voll Blut und Wunden,” feature the same exact progression shown in Example 7a: same key, same harmonies, same voicing in the outer parts. The use of this harmonic schema in Baroque chorales resonates with Brahms’s use of a chorale topic in the middle section of this Intermezzo.
19. Alexander Stefaniak (2016, 91–110) has argued that Schumann blurs the structure of his works by employing a “nontransparent style of virtuosity” that subverts the melody and obfuscates phrase boundaries. Peter H. Smith (1994) has shown how Brahms overlaps the end of the development with the start of the recapitulation in several sonata-form movements, thus “blurring the recapitulatory articulation.” Smith (2005) expands the concept of formal overlap to include other sections. The findings of this article complement the idea that Schumann and Brahms blurred phrase boundaries and the recapitulatory articulation.
20. As Alice Kuzniar (1989, 75) notes, when Müller described the blurriness of the landscape, he is implicitly referring to Caspar David Friedrich’s paintings.
21. It was common for Romantic artists to temporalize space. On the Romantic fusion of space and time, see Kuzniar (1989) and Hoeckner (2002, 80).
22. As John Daverio (1987, 157) shows, the various themes of this movement prefigure the quotation from Beethoven. The movement, then, dissolves Schumann’s music into Beethoven’s, the present into the past, foreground into background.
23.  Several scholars have noted the connection between the passages marked x in Example 8, but the connection between the passages marked y has gone unnoticed. The arpeggiation of the V chord up to scale-degree 2 makes the connection between Beethoven’s original and Schumann’s quotation particularly clear.
24. Beethoven’s song begins with a Prinner, a stock Galant progression that fell sharply into disuse in the Romantic period. As Gjerdingen (2007, 45–60) explains, the prototypical form of this schema features a $$\hat6$$–$$\hat5$$–$$\hat4$$–$$\hat3$$ melodic descent above a $$\hat4$$–$$\hat3$$–$$\hat2$$–$$\hat1$$ bass descent, resulting in four distinct harmonies: IV, I⁶, V$$4\atop3$$ (or VII⁶) and I. Schumann uses the same basic set of sonorities to harmonize Beethoven’s melody but with some alterations: the IV becomes a $$\sharp$$IV⁷ (spelled enharmonically), the I⁶ becomes a I$$6\atop4$$ (cadential), and the V$$4\atop3$$ becomes a V⁷.
25. Here and elsewhere, I use the word “ruin” in its positive, Romantic sense rather than in a pejorative, colloquial way.
26. As Burns’s (1995) label suggests, the origins of this progression can be traced back to modal music. Burns (1995) notates this progression as V–V⁷/IV–IV–I. I have notated the V⁷/IV as I$$7\atop\natural$$ in order to keep the chord labels consistent throughout the article. This closing formula is closely related to the schema that Robert Gjerdingen (2007, 181–82) calls the Quiescenza.
27.  Even though one may question the value of labeling the sonorities of the sequence in m. 9ff with roman numerals, I have done so in order to show how these sonorities overlap in the music.
28. As Gjerdingen (2007, 98) notes, a Monte Principale is an old type of Monte that proceeds diatonically with all the chords in $$5\atop3$$ position, rather than alternating first-inversion and root-position triads. The main difference between a typical Monte and a 5–6 sequence is one of grouping: while in a 5–6 sequence the chords are grouped in 5–6 pairs, in a Monte they are grouped in 6–5 pairs. Just as in a 5–6 sequence the second chord of each 5–6 pair can appear in root position, so, too, in a Monte the first chord of each 6–5 pair can appear in root position, creating a Monte Principale. The Monte Principale is distinct enough from a 5–6 sequence with all its chords in root position that Aldwell and Schachter (2011, 319) label this sequence as one that ascends “by step with voice-leading 5/3 chords” rather than as a variant of the 5–6 sequence.
29. Salzer and Schachter (1969, 187–189) offer a similar reduction of this passage. While they read an ascending 5–6 sequence beginning in m. 10, I read a Monte Principale starting a measure earlier. The Monte Principale fits better the internal organization of the poetry than the 5–6 sequence. In mm. 9–14 the melodic line is divided into three two-measure units, the first starting with the G in m. 9 and ending with the A in m. 10 (“Eingewachsen Bart und Haare”), the second starting with the A in m. 11 and ending with the B in m. 12 (“Und versteinert Brust und Krause”), and the last starting with the B in m. 13 and ending with the C in m. 14 (“Sitzt er viele hundert Jahre”). The G-to-A, A-to-B, and B-to-C melodic units in mm. 9–14 help articulate the CM-to-FM, Dm-to-GM, and Em-to-Am chord-pairs of the Monte Principale shown in Example 9, while contradicting the FM-to-Dm and GM-to-Em chord-pairs of the 5–6 sequence shown by Salzer and Schachter. Hearing this passage as a Monte Principale rather than as a 5–6 sequence is an important part of hearing this phrase as evocative of the past.
30. Similarly, David Lewin (2006, 175) sees “two structural harmonies compressed into the climactic chord of m. 14.”
31. The slowdown of the melody is evident when one compares the fact that lines 1–3 of the second quatrain each occupy two measures of music, while line 4 occupies four measures. In m. 16 the pacing of the text’s declamation slows down from four syllables per measure to two syllables per measure. In m. 17, the declamation slows down further, to one syllable per measure.