Beyond the Classical Cadence: Thematic Closure in Early Romantic Music

Beyond the Classical Cadence: Thematic Closure in Early Romantic Music Abstract Building on my 2004 article “The Classical Cadence: Conceptions and Misconceptions,” this article examines the fate of the classical cadence in the hands of the first generation of Romantic composers (Schubert, Chopin, Schumann). It identifies seven characteristics of Romantic compositional style that, compared to classical practice, bear on issues of cadential morphology and function: (1) a more extensive use of chromaticism and dissonance; (2) a greater emphasis on root-position harmonies; (3) a more uniform harmonic rhythm and harmonic density; (4) a circularity of formal organization; (5) an ambiguity between sequential and cadential harmonies; (6) a lack of cadential closure for thematic units; and (7) an ambiguity between penultimate and ultimate dominants at points of potential cadence. Though I focus largely on cadence and closure, I also take the opportunity of pointing out other aspects of musical form that distinguish Romantic practice from the earlier classical style. In my 2004 article “The Classical Cadence: Conceptions and Misconceptions,” I argued that musical cadence in general was poorly understood by many musicians and that one way of regaining control over the idea is to examine cadence in the context of a relatively narrow, stylistically unified repertory.1 In particular, I advocated that the concept of cadence is best conceptualized at first in relation to music of the high classical style (Haydn, Mozart, and Beethoven). I concluded my study by suggesting that a greater involvement by scholars in theories of musical form could lead to a better understanding not only of the classical cadence but also of cadence as manifest in earlier and later repertories as well. The present article extends my approach by exploring cadential processes found in the instrumental music of first-generation Romantic composers, such as Schubert, Mendelssohn, Schumann, Chopin, and Berlioz.2 Before turning to this repertory, however, I should first summarize the essential features of the classical cadence, since not only does music of the nineteenth century continue to employ that cadence on a regular basis, but also some novel cadential techniques introduced in the Romantic period are best understood in relation to prior classical practice. Central to my definition of the classical cadence is the idea that it is supported by a specific type of harmonic progression (Ex. 1[a]), one that features an initial tonic (usually in first inversion), a pre-dominant harmony (typically II6, but sometimes IV), a penultimate dominant in root position (often embellished by the cadential six-four), and a final tonic, also in root position.3 Incomplete forms of this authentic cadential progression may sometimes eliminate either the initial tonic or pre-dominant (or both), but the dominant must always first appear in root position and remain as such. Indeed, this notion is axiomatic to my theory of cadence, for I believe that any inversion of the dominant undermines the potential of such a progression to be truly cadential.4 In the case of a half-cadential progression (Ex. 1[b]), the final harmony is an ultimate dominant, which must take the form of a consonant triad in root position. We can also identify a deceptive cadential progression (Ex. 1[c]), in which the final tonic is replaced by some other harmony, typically VI. EXAMPLE 1. View largeDownload slide Cadential progressions (Caplin 2013a, 4) EXAMPLE 1. View largeDownload slide Cadential progressions (Caplin 2013a, 4) In my own particular take on harmonic theory, I distinguish such cadential progressions from two other progression types, which I term prolongational and sequential (specific cases of these are shown in Exx. 2 and 3).5 Note that prolongational progressions often bring inverted dominants embellishing the tonic harmony.6 In addition to its harmonic content, a classical cadence typically features a conventionalized melody, most often of descending contour. Example 4 illustrates some representative melodic-motivic ideas that clearly express to the listener “this is a cadence.” EXAMPLE 2. View largeDownload slide Prolongational progressions (Caplin 2013a, 3) EXAMPLE 2. View largeDownload slide Prolongational progressions (Caplin 2013a, 3) EXAMPLE 3. View largeDownload slide Sequential progressions (Caplin 2013a, 6) EXAMPLE 3. View largeDownload slide Sequential progressions (Caplin 2013a, 6) EXAMPLE 4. View largeDownload slide Classical cadences: (a) Beethoven, Piano Sonata in C, Op. 2, No. 3, third movement, mm. 15–16; (b) Haydn, Piano Trio in D, H. 16, third movement, mm. 7–8; (c) Mozart, Piano Sonata in B♭, K. 333, second movement, mm. 20–21; (d) Haydn, Piano Sonata in B Minor, H. 32, second movement, mm. 9–10; (e) Haydn, Piano Sonata in A, H. 30, third movement, mm. 7–8; (f) Beethoven, Piano Sonata in C Minor, Op. 10, No. 1, third movement, mm. 3–4. EXAMPLE 4. View largeDownload slide Classical cadences: (a) Beethoven, Piano Sonata in C, Op. 2, No. 3, third movement, mm. 15–16; (b) Haydn, Piano Trio in D, H. 16, third movement, mm. 7–8; (c) Mozart, Piano Sonata in B♭, K. 333, second movement, mm. 20–21; (d) Haydn, Piano Sonata in B Minor, H. 32, second movement, mm. 9–10; (e) Haydn, Piano Sonata in A, H. 30, third movement, mm. 7–8; (f) Beethoven, Piano Sonata in C Minor, Op. 10, No. 1, third movement, mm. 3–4. But now we must distinguish fundamentally between the cadential content of a musical idea and its actual formal function. For in the classical style, cadences serve to provide formal closure at specific levels in the musical hierarchy, namely the middleground levels of phrase and theme. And for a cadential idea to create closure, it must follow something that functions as a formal beginning or a formal middle. Cadence is therefore a mechanism of ending, though, as will be discussed later on, not all formal endings involve cadences. Likewise, closure at very low hierarchical levels (such as that of a two-measure basic idea) and closure at high, or background, levels (such as the entire exposition section of sonata form) are not normally cadential at all, but rely on other formal processes to effect the sense of ending.7 Another distinction central to my concept of cadence is that between cadential function, a time-span leading up to the moment of cadential arrival, and post-cadential function, a time-span that follows the arrival and prolongs the final harmony of the cadence. Such post-cadential activity normally consists of a closing section (made up of codettas) following an authentic cadence or a standing on the dominant following a half cadence (HC).8 In the course of a formal process leading to cadential closure, a promised authentic cadence frequently fails at first to materialize, only to be fully realized eventually by a genuine cadence. For the classical style, I have identified three such cadential deviations (deception, evasion, and abandonment), of which the evaded cadence is most relevant for the present study.9 A cadence is evaded when the event that follows the penultimate dominant seems not to represent the final event of the cadential process, no matter what its harmonic support. In other words, this event does not group backwards with the preceding events of the cadence (those supported by an initial tonic, pre-dominant, and penultimate dominant), but rather groups forward with the subsequent events. A clear case of cadential evasion is shown in Example 5, where a perfect authentic cadence (PAC) is promised on the downbeat of m. 16. What occurs at this moment, however, does not represent the goal of the ongoing cadential process; even though supported by a root-position tonic, this event marks the onset of a new phrase, which brings the cadential idea “one more time”10 to close the theme at m. 19. EXAMPLE 5. View largeDownload slide Mozart, Piano Sonata in C, K. 309, third movement, mm. 13–19 (Caplin 2013a, 142) EXAMPLE 5. View largeDownload slide Mozart, Piano Sonata in C, K. 309, third movement, mm. 13–19 (Caplin 2013a, 142) With this all-too-brief introduction to the classical cadence, let us turn to the main topic at hand, namely, how phrase and thematic closure is realized in the early Romantic period (roughly the second quarter of the nineteenth century). First, though, we must acknowledge a central fact—Romantic composers continue to employ the classical cadence throughout their compositions. Many phrases, and most thematic units, close cadentially in the manner just described. A complete movement, moreover, is almost always marked by cadential closure of its final theme.11 Yet, despite the powerful influence that the classical cadence continues to exert, composers in the Romantic period explore some new compositional procedures that modify or distort in interesting ways the classical means of creating closure. In my research to date, I have identified seven stylistic characteristics of Romantic compositions that have the potential to affect the ways in which cadence and closure are achieved. Compared to classical practice, Romantic works may feature: (1) a more extensive use of chromaticism and dissonance; (2) a greater emphasis on root-position harmonies; (3) a more uniform harmonic rhythm and harmonic density; (4) a circularity of formal organization; (5) an ambiguity between sequential and cadential harmonies; (6) a lack of cadential closure for thematic units; and (7) a confusion between penultimate and ultimate dominants at points of potential cadence. The remainder of this article will examine these seven points in greater detail, with specific examples drawn from the Romantic repertory. Though I focus largely on cadence and closure, I also take the opportunity, where appropriate, of highlighting other aspects of musical form that distinguish Romantic practice from the earlier classical style. CHROMATICISM AND DISSONANCE That music in the nineteenth century witnesses a marked increase in the use of chromaticism and dissonance over that of the eighteenth is widely accepted. Indeed, we might think that greater chromaticism would significantly impact, perhaps even distort, cadential expression. In fact, this seems not to be the case for the Romantic generation. Most often, chromatic alterations of a given harmony can be understood in relation to a basic diatonic model, and the harmonic function defined by that model is normally retained in the chromatic version. Thus, even if a given cadential harmony is extensively chromaticized, its harmonic function within the progression remains comprehensible, and its cadential effect can be fully realized. Similarly, the addition of various dissonances does not normally disrupt the functional expression of the cadential harmonies.12 One manifestation of greater dissonance treatment, however, does have significant cadential implications. In the classical style, the final harmony of both the authentic cadence and the half cadence is normally a consonant triad, either a tonic or a dominant, in order that this final sonority has sufficient stability to express a sense of ending. In the Romantic period, this restriction begins to be eased for the ultimate dominant of the half-cadential progression. With Chopin and Schumann especially, the effect of an HC continues to be projected even when the final harmony is represented by a dominant seventh. For this reason, Janet Schmalfeldt has specifically identified a nineteenth-century half cadence at points where a classical HC would be expected (such as at the end of an antecedent phrase) but in which the ultimate dominant contains a dissonant seventh already at the moment of cadential arrival.13 A typical case is found in Chopin’s Waltz in A Minor (Ex. 6). The overall form of this theme is a compound period, whose opening compound antecedent is structured as a simple sentence.14 Following a regular presentation phrase (made up of a repeated basic idea),15 m. 21 brings the start of an expanded cadential progression (ECP), with V56 of IV substituting for the more standard I6 (shown in brackets). As the pre-dominant IV6 in m. 23 moves to the dominant in the following measure to produce the final harmony of the phrase, the pitch D is held over, thus creating a clear dissonant seventh within the dominant of this nineteenth-century HC. Not to consider this moment a type of HC seems overly restrictive, since Chopin regularly employs the dissonant seventh in this kind of formal context. This cadence nonetheless represents a clear departure from classical practice, where the use of such a dissonance for the final cadential harmony is rare. (I return to this example to discuss its consequent phrase after introducing some other characteristics of the Romantic style.) EXAMPLE 6. View largeDownload slide Chopin, Waltz in A Minor, Op. 34, No. 2, mm. 17–36 EXAMPLE 6. View largeDownload slide Chopin, Waltz in A Minor, Op. 34, No. 2, mm. 17–36 ROOT-POSITION HARMONIES One striking feature of early Romantic harmony has largely gone unremarked by theorists: harmonies frequently succeed each other in root position. In fact, a long string of root-position chords seldom arises in baroque, pre-classical, and classical works, yet such a succession is regularly found in Romantic ones.16 This emphasis on root-position harmonies has an impact on the specific harmonic progressions that tend to be used cadentially. Example 7 illustrates some typical cadential progressions found in Romantic works. Note that the pre-dominant harmony tends to favor either the subdominant or the supertonic in root position (a and b), rather than the more typical II6 of the classical style. The initial tonic of nineteenth-century cadences is more often placed in root-position (a and b) than in classical practice, and this harmonic function may be embellished by a root-position VI, which may even fully substitute for that tonic (c). Another initial tonic substitute sees the third degree in the bass (which would normally support I6) holding a III harmony, or even chromatically, a dominant of the submediant (V7 of VI), which resolves deceptively to the subdominant (d). The resulting bass line resembles a classical one, but the chromaticism betrays a Romantic touch. EXAMPLE 7. View largeDownload slide Nineteenth-century cadential progressions EXAMPLE 7. View largeDownload slide Nineteenth-century cadential progressions Example 8(a), an excerpt from Chopin’s Nocturne in G Minor, shows an extreme case of root-position harmonic succession in a passage that obviously references the genre of chorale, though clearly not in the style of Bach. The final beat of m. 44 seems to be a clear enough perfect authentic cadence (PAC), which itself is preceded by a deceptive resolution of the dominant. And we cannot entirely rule out the expression of a PAC at the end of m. 42 as well. In both places, however, the cadential progression is quite incomplete and, especially in the case the opening two-measure unit, we might want to regard the final two beats of m. 42 as continuing the prolongational progressions of the prior measure and a half. Indeed, the excessive use of root-position harmonies helps to blur the fundamental distinctions among prolongational, sequential, and cadential progressions that are so essential to classical harmony in its relation to form. Although this example of chorale texture is not unique in Chopin’s oeuvre, it still remains somewhat exceptional.17 Yet the root-position emphasis found in his chorale textures is also present at the very opening of this same Nocturne (Ex. 8[b]), which features a more standard melody and accompaniment texture. We might initially be tempted to identify various harmonic inversions within the first two measures (VII7, VI6, VII56); however, the functional bass seems to sound in a lower tessitura of the piano. If so, we can hear the opening two measures as projecting a tonic pedal (the seeming dominant pedal thus residing in a tenor voice) and then observe that all of the harmonies in the theme stand in root position, with the exception of the I6 in m. 7 (a detail to be discussed momentarily).18 EXAMPLE 8. View largeDownload slide Chopin, Nocturne in G Minor, Op. 37, No. 1, (a) mm. 41–44; (b) mm. 1–8 EXAMPLE 8. View largeDownload slide Chopin, Nocturne in G Minor, Op. 37, No. 1, (a) mm. 41–44; (b) mm. 1–8 Though I am using this example to illustrate the Romantic tendency to exploit root-position harmonies, it is worth pausing briefly to examine details that reveal some other non-classical traits. In the first place, the form of the theme seems to be a standard period—an opening antecedent with weak cadential closure followed by a consequent bringing a stronger cadence. Entirely counter to classical norms, however, the antecedent phrase already modulates to the relative major, ending there with a nineteenth-century HC.19 (Note, by the way, that the local E♭ sonority, VI6, on the downbeat of the second measure not only adds a major-mode coloration to the otherwise pathos-laden minor quality of the opening two-measure idea but also subtlety alludes to the upcoming modulation to B♭ major.) The consequent phrase begins with the tonic of the new key, but quickly modulates back to G minor. To solidify this return, Chopin brings a more marked sense of harmonic progression in mm. 5 and 6 (four distinct harmonies there compared to a single prolonged harmony in mm. 1–2), and he makes sure that the downbeat of m. 6 holds a strong root-position tonic, rather than the more luminous VI6 of the antecedent (cf. m. 2). The final cadential idea is especially interesting, since its upper-voice melodic content ( 1^– 2^– 3^– 4^– 5^– 1^) brings the main elements of a classical bass line (the scalar rise from ① to ⑤ and subsequent leap back to ①, a line that regularly supports an entire thematic unit).20 To avoid parallel octaves (except at the very end, a license found frequently enough in most tonal styles), Chopin brings a non-classical cadential bass line (⑥—③—②—⑤—①), featuring a root-position VI as the initial tonic of the progression. After a subsequent I6—an embellishing chord, which is the only inverted harmony in the theme—the bass pushes downward to support a root-position half-diminished II7, a pre-dominant rarely used in a classical cadential progression.21 Thus, the theme as whole exemplifies many ways in which what might seem to be a fairly ordinary period actually departs considerably from the norms of the classical style, while bringing a series of typically Romantic elements of harmony and form. UNIFORM HARMONIC RHYTHM AND DENSITY In a manner that harkens back to baroque practice, the Romantic style often features a relatively uniform rhythm of harmonic change, especially compared to the classical style. In many cases, this uniformity of harmonic rhythm arises from the use of various dance genres (mazurka, waltz) or from character pieces that tend to feature a single emotional affect, thus emulating, again, a baroque aesthetic.22 This rate of harmonic change can often be quite fast as well, especially at the very opening of a thematic unit. (In the classical style, on the contrary, the harmonic rhythm of a theme tends initially to be relatively slow and then systematically increases toward the cadence.) To avoid overburdening the harmonic texture with rapidly changing prolongational progressions, Romantic composers will vary the progression type by introducing both sequential and cadential progressions in initiating formal positions. In addition, the possibility of embedding one progression type within another emerges as a distinct option for creating a degree of harmonic diversity within a foreground uniformity of harmonic rhythm. The opening of Schumann’s Davidsbündlertänze No. 8 (Ex. 9) illustrates well the kind of uniformity and density of harmonic activity found in a Romantic character piece. The opening measure already brings a cadential progression in C minor, which upon repetition in the following measure proves to be a mere accompaniment to a rather fragmented melodic idea. Taken together and in the context of the entire theme, these opening two measures represent a basic idea. The accompanimental pattern in m. 3 is then varied to become a cadential progression in E♭, and the complete melodic fragment is repeated a third higher. We thus see that Schumann embeds cadential progressions within a broader ascending-third sequence.23 This use of both cadential and sequential progressions for the very opening unit of a theme—here, a kind of presentation phrase—is especially characteristic of the Romantic style.24 At the beginning of the continuation phrase, m. 5, the progression resolves deceptively, thus shifting the tonal focus back to C minor. Observe that up to this point, most of the harmonies of mm. 1–4 have appeared in root position. With the deceptive resolution of the dominant at the beginning of m. 5, the bass line now acquires an ascending stepwise melodic profile, which lasts most of the way through the modulation to the minor dominant and the confirmation of the new key by a genuine cadence (a fully classical one), which closes this sentence theme type. Note that the embedded cadential progressions at the opening of the theme only project cadential content, not any sense of cadential function—nothing is ended by these cadence-like ideas.25 EXAMPLE 9. View largeDownload slide Schumann, Davidsbündlertänze, Op. 6, No. 8, mm. 1–7 EXAMPLE 9. View largeDownload slide Schumann, Davidsbündlertänze, Op. 6, No. 8, mm. 1–7 Let us return to Example 6 and consider the compound consequent phrase in light of the uniform harmonic rhythms typical of a Romantic theme. The phrase begins at m. 25 with varied material from the opening of the compound antecedent. Measure 29 sees the start of the ECP but now with a new one-measure melodic motive in the right-hand part, which is repeated in the following measures. The insistent upper pedal on the high A makes it difficult for cadential melodic closure to occur, and so the ECP is abandoned when at m. 33 the bass leaps down to the leading tone to support a non-cadential V56.26 Following the resolution to I, a simple V7–I progression creates the final PAC. Note that it would have been perfectly possible for the ECP that began at m. 29 to have continued all the way to the end of the theme by bringing a root-position dominant at m. 33 and sustaining that harmony until the tonic at m. 36, as shown in the bracketed harmonic analysis. Such an option would have been entirely conceivable within the classical style. Chopin, however, avoids the harmonic deceleration that would so arise, probably for several reasons. First, he has established throughout most of the theme a fairly uniform rhythm of each measure bringing either a new harmony, or a change of bass, or both. By abandoning the cadential progression, he is able to maintain this consistent harmonic rhythm to the end of the theme. Second, by bringing the progression V56–I in mm. 33–34, he can provide the same basic harmonic support to those measures as for the similar material in mm. 29–30. As a result, m. 33 sounds like the beginning of a sequential repetition, which is effectively given up when the need finally to create thematic closure brings the compressed, and highly incomplete, cadential progression. That the cadential idea emerges out of what is initially heard as a sequential situation is a Romantic characteristic to which I will return in a later section. FORMAL CIRCULARITY In the classical style, themes normally begin with a relatively “characteristic” idea, which, in accord with the powerfully teleological aesthetic of the style, leads to the goal of thematic closure. Part of the broader mechanism for motivating that closure resides in the breaking down of this opening material—a process Schoenberg graphically termed liquidation—such that all that remains by the end of the theme is the relatively “conventional” cadential idea. In contrast, a newly arising Romantic orientation to musical form began to favor a more circular mode of organization, one that manifests itself at various structural levels.27 At the level of the theme, this tendency toward formal circularity can be realized when the melodic–motivic material of the opening basic idea, rather than becoming fully liquidated, returns at the end to provide the melodic content of the cadential idea itself. Whereas this technique has some precedents in the classical style, it is more frequently encountered in Romantic practice, where, in fact, it occurs in a manner that differs markedly from earlier manifestations of the technique. Example 10, the opening of the scherzo of Schumann’s Second Symphony, is a case in point. The theme starts off-tonic in C major with a two-measure basic idea that is immediately repeated to make a presentation phrase; the continuation brings one-measure fragments as expected. Rather than further liquidating the material to yield a conventionalized cadential idea, however, Schumann brings back the melodic content of the basic idea for an imperfect authentic cadence (IAC) in E♭ (♭III) (m. 8). A repeat of the continuation phrase then modulates further to the minor dominant, closing there with another variant of the opening material to create a PAC in m. 12.28 EXAMPLE 10. View largeDownload slide Schumann, Symphony No. 2, Op. 61, third movement, mm. 1–12 EXAMPLE 10. View largeDownload slide Schumann, Symphony No. 2, Op. 61, third movement, mm. 1–12 As mentioned before, some classical precedents for this re-use of the opening basic idea can also be found, such as shown in Example 11, from Haydn’s Symphony No. 100, a passage discussed extensively by Leonard B. Meyer.29 Here, the basic idea itself has a distinctly “cadential” melodic component, one that classical composers used as a closing gesture in several other works.30 And in an obvious realization of this closing potential, Haydn brings back the idea, lightly varied, to forge the PAC at m. 36.31 Although the compositional situation here may remind us of Schumann’s theme, a significant difference lies in the nature of the melodic material used for both the basic and cadential ideas. With Haydn, the melody is already cadential in implication, and so it is hardly surprising that he allows the idea eventually to fulfill its implied formal function. With Schumann, on the contrary, the melodic gesture at the opening of the theme has nothing stereotypically closing about it; in fact, it is a highly “original” idea, which, even when re-harmonized so as to function cadentially, retains its highly individualized quality. We thus see operative here another aspect of Romantic style, namely an expression of the “aesthetics of originality.” Whereas classical composers are content to allow their initiating melodic-motivic material to become liquidated in the course of a theme, Romantic composers are more attached to their “novel” melodic ideas, and are thus inclined to retain them for use at the very end of the theme in a manner that promotes a circular design.32 EXAMPLE 11. View largeDownload slide Haydn, Symphony No. 100 in G, second movement, mm. 29–36 EXAMPLE 11. View largeDownload slide Haydn, Symphony No. 100 in G, second movement, mm. 29–36 SEQUENCE VERSUS CADENCE The Romantic practice of emphasizing root-position harmonies can give rise to an important source of harmonic and cadential ambiguity. If we consider the harmonies shown in Example 12, we can observe that such a progression can be analyzed as either a circle-of-fifths sequence, which could project a medial formal function of some sort (such as a continuation), or as a cadential progression, which could bring formal closure.33 Of course, a specific musical context will normally clarify which type of progression is involved: thus the progression in mm. 7–8 of Example 8(b), discussed earlier, arises in a context that unambiguously suggests its formal function to be cadential (analogous to mm. 3–4 in the antecedent phrase). In the classical style especially, composers are usually very careful to distinguish between the progression types. A typical case arises in the first subordinate theme from Mozart’s G-Minor Symphony (Ex. 13), where the continuation phrase (mm. 48–51) begins with a series of root-position harmonies organized by descending fifths. In its formal context and in its intrinsic content, the harmonies are obviously sequential, which Mozart makes clear by following the sequence with a distinct cadential progression to close the theme. EXAMPLE 12. View largeDownload slide Sequential versus cadential EXAMPLE 12. View largeDownload slide Sequential versus cadential EXAMPLE 13. View largeDownload slide Mozart, Symphony in G Minor, K. 550, first movement, mm. 44–51 EXAMPLE 13. View largeDownload slide Mozart, Symphony in G Minor, K. 550, first movement, mm. 44–51 Among Romantic composers, this practice of separating descending-fifths sequential progressions from cadential ones is not always observed, thus giving rise to ambiguities of cadential articulation. Consider the subordinate theme from the first movement of Schubert’s “Unfinished” Symphony (Ex. 14[a]), a melody so familiar to us that it is easy to overlook how utterly non-classical it really is. We can first note that all of its harmonies stand in root position, which already suggests that we may encounter ambiguities in the function of the harmonic progressions. After two measures of introduction, the theme begins with a two-measure basic idea (supported by I going to V), followed by a contrasting idea (V to I). Matching this harmonic reversal is a motivic one, as shown by the analysis of motives “a” and “b.” (This kind of motivic palindrome is rarely, if ever, found in classical themes.) Though we might be tempted to recognize a cadence at the end of this four-measure phrase, it is probably better to understand it as concluding without cadential closure, primarily because the harmonic pattern I–V–V–I, supporting an opening four-measure phrase, more typically projects four measures of tonic prolongation rather than a two-measure prolongation followed by a two-measure cadential progression. Here especially, it is hard to hear the harmony in the third measure of the theme (m. 46) as a distinctly cadential dominant (one that differs from the prolongational dominant of the prior measure), all the more so given the commonality of melodic-motivic material in both measures. At m. 48, the phrase begins to be repeated, but motive b of the basic idea is now sequenced up a step (supported by V7 of II). The “contrasting idea” (mm. 50–51) then remains a step higher than it was in the first phrase. In order to return to tonic, Schubert brings a varied sequence of the idea back down a step in mm. 52–53. Though the rhythm has been modified, the sequential relationship between these two ideas could be seen clearly if Schubert had written an exact sequence, as shown in the second staff of Example 14(b). What this rewrite also helps to illuminate is how the rhythmic modification of the sequence makes it sound as though it is beginning with a sequence of “motive a” whose descending fourth leap is now extended into a fifth. Finally, Schubert lets the melodic line end decisively on 1^ (see again Ex. 14[a])—decisive in that the cello melody does not directly leap down to 5^, as in m. 47. In another sense, however, we do hear this melodic leap, since the downbeat of m. 53 also brings an elided repeat of the entire theme, with its opening “a” motive.34 But now the question becomes, has the theme actually ended with a cadence? Most listeners will probably answer “yes” readily enough, but we should nonetheless be sensitive to a formal ambiguity here, for what we would be construing as the “cadential” idea, mm. 52–53, arises as a sequential repetition of the contrasting idea of mm. 50–51. In the classical style, a cadential idea almost always distinguishes itself from any preceding sequential material, as we saw in the previous Mozart example. Because a sequence brings a repetition of something that is already medial in function, it should continue to project the sense of being-in-the-middle and not sound cadential. However, here it seems to do just that, namely to bring a semblance of closure to the theme. In other words, we can hear the harmonies in mm. 49–52 as not only sequential (shown in line 1 of the harmonic analysis of Ex. 14[a]), but also as cadential, with V7 of II representing a chromatic alteration of VI, a substitute for the initial cadential tonic (shown in line 2). Given the harmonic and cadential ambiguities seen thus far, it is fascinating to observe some of the implications that Schubert draws from them as the movement proceeds. In the first place, the repeat of the theme in the violins leads to a complete lack of cadential closure when the music breaks off with the grand pause at m. 62 (Ex. 14[c]). This drastic evaded cadence is unusual in that not only is there no harmonic event (a tonic) at the moment of expected cadential closure, but also a complete absence of any event (the grand pause) at all. In addition, this moment is not followed by a “one more time” repetition as often occurs with evaded cadences.35 Instead, an entirely new augmented version of motive “a” returns, set as a dramatic and ominous tutti fortissimo outburst.36 In the recapitulation, Schubert handles this passage somewhat differently (Ex. 14[d]). Here, he brings the music back from D major (the key in which the theme appears in the recapitulation) to the home key of B minor by more fully realizing the sequential possibilities of the contrasting idea as it appears in Example 14(b), staff 1, and, in so doing, significantly minimizing the contrasting idea’s cadential effect. By the time we hear the grand pause at m. 280, we are not at all certain that an evaded cadence has even taken place; rather, the music seems more to be interrupted “mid-sequence,” with little sense of any cadential function. Returning to the exposition (Ex. 14[c]), we see that the fortissimo outbursts at mm. 63–66 propose the beginning of an expanded cadential progression by means of a prolonged pre-dominant (IV–[ 46]–Gr+6– VII56 of V); however, the progression rhetorically runs out of steam with the syncopated chords in mm. 71–72 and is effectively abandoned when, at m. 73, Schubert brings back motive “b” in an extensive call-and-answer passage, one that more fully realizes another potential for sequence in the contrasting idea, namely that proposed in Example 14(b), staff 2. Following a decisive cadence at m. 93 to end the subordinate theme group (Ex. 14[e]), Schubert brings back the opening of the subordinate theme,37 but now in a manner that resolves the ambiguity of sequence and cadence. Like the original version, the third measure of the theme (m. 96) repeats, albeit in a non-sequential manner, the second half of the basic idea (motive “b”). A completely new idea then appears (mm. 97–99) that is more genuinely cadential in nature, with nothing sequential in the picture. The only disrupting aspect is the harmonic support, whereby the dominant reached at m. 95 is held as a pedal point until the tonic resolution at m. 99. If Schubert had supported mm. 96–99 with a true cadential progression, as shown in the brackets, then we could speak of a PAC at m. 99. The retention of the dominant pedal from m. 95 on, however, blurs a sense of real cadence, thus allowing this passage to function formally more as an expanded codetta of a closing section rather than as a genuine thematic unit (i.e., another subordinate theme). EXAMPLE 14(A) AND 14(B). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (a) mm. 42–54; (b) alignment of sequences EXAMPLE 14(A) AND 14(B). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (a) mm. 42–54; (b) alignment of sequences EXAMPLE 14(C) AND 14(D). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (c) mm. 60–76; (d) mm. 274–82 EXAMPLE 14(C) AND 14(D). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (c) mm. 60–76; (d) mm. 274–82 EXAMPLE 14(E). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (e) mm. 90–100 EXAMPLE 14(E). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (e) mm. 90–100 To conclude this discussion of the “Unfinished” Symphony’s subordinate theme, we can now observe that a significant source of ambiguity in the passage arises from Schubert having placed all of the harmonies in root position; as a result, the opening phrase can be heard as either prolongational or cadential, and the second phrase as either sequential or cadential. Various aspects of motivic manipulation contribute to the ambiguous effect as well. PROLONGATIONAL CLOSURE; LACK OF FUNCTIONAL ENDING Not all phrases in the classical style close with a cadence: the opening presentation phrase of the sentence theme type is a case in point.38 But almost all complete themes in that style conclude with one of the three genuine cadences—PAC, IAC, or HC. With the Romantic style, we see a loosening of this thematic requirement and begin to find themes that close not with cadential progressions, but with prolongational ones instead. Schumann’s “Valse noble” from Carnaval (Ex. 15[a]) illustrates well how a self-contained thematic unit ends with such prolongational closure, as I will term it.39 The movement, starting very much in medias res, brings two matching four-measure phrases in a manner that suggests a periodic theme-type. Within the broader form of the piece as a whole, these opening eight measures constitute the A section of a small ternary. In the classical style, this section would normally end cadentially (with a PAC in either the home key or a subordinate key).40 Here, however, both phrases end prolongationally, the first with V24 to I6, the second with V56 to I. To be sure, the norms of the period form are referenced by the first phrase ending with a weaker closure (on an inverted tonic) than the second phrase (on a root-position tonic), yet neither mode of closure is truly cadential. Moreover, the bass line supporting the complete theme is entirely non-classical, featuring a descending line from ♭⑥ down to ⑦, which resolves up to ① at the very end.41 EXAMPLE 15. View largeDownload slide Schumann, “Valse noble,” Carnaval, Op. 9. No. 4, (a) mm. 1–8; (b) mm. 33–40 EXAMPLE 15. View largeDownload slide Schumann, “Valse noble,” Carnaval, Op. 9. No. 4, (a) mm. 1–8; (b) mm. 33–40 Following a contrasting middle (B section), the A′ section first brings back the material of the complete A section without any alterations. Schumann however, obviously not satisfied in allowing the complete movement to close prolongationally, begins to sound the theme again (Ex. 15[b]) and rewrites the harmonic support for the final phrase to make a genuine cadential progression (whose constituent harmonies, not surprisingly, are all set in root position), thus concluding the movement as a whole with a real cadence.42 We see that, unlike classical practice, prolongational closure in the Romantic style may take place at the level of the simple theme (the eight-measure A section),43 but the final closure of broader formal units still requires an authentic cadence.44 A more radical mode of “ending” arises when an ongoing thematic unit fails to project any intrinsic sense of closure, yet we realize that the unit is “finished” when the onset of a subsequent unit initiates a new set of formal processes. Such situations are rare in eighteenth-century works and are hardly common in the Romantic style. Some interesting examples, however, do arise, thus revealing the breakdown of classical means of formal closure. A casual hearing of the opening theme of Chopin’s Mazurka in F♯ Minor (Ex. 16[a]) suggests that it comprises a sixteen-measure compound period, whose antecedent forms an eight-measure sentence. The return of the basic idea at m. 9 quite clearly signals the onset of a consequent. If we look more closely, though, we must ask, how does the antecedent actually end? From the perspective of classical norms, we would be expecting a closing HC in m. 8. Instead, the harmony preceding the return of the basic idea is a pre-dominant II34, which does not mark the end of any cadence type. Not only does this formal unit not end with a cadence, but we also cannot speak of any prolongational closure at this point, since this harmony appears as the last link in a broad descending-fifth sequential progression begun at m. 5.45 Little in the intrinsic musical content projects any sense of closure whatsoever for this “antecedent” phrase. We only know that it ends because the consequent so obviously begins. Of course, what helps motivate the formal oddities observed in this Mazurka is the way in which the opening basic idea is harmonized: namely with a prominent V7–I prolongational progression.46 Thus, in order to set up the formal return at m. 9, Chopin lets the antecedent conclude on the pre-dominant II34, which connects easily with the initiating dominant to tonic motion of the basic idea.47 EXAMPLE 16. View largeDownload slide Chopin, Mazurka in F♯ Minor, Op. 6, No. 1, (a) mm. 1–16; (b) recomposed version by Rothstein (1989, Ex. 2.23) EXAMPLE 16. View largeDownload slide Chopin, Mazurka in F♯ Minor, Op. 6, No. 1, (a) mm. 1–16; (b) recomposed version by Rothstein (1989, Ex. 2.23) At this point, we might consider an option proposed by William Rothstein, who wants to hear a half-cadential goal for the antecedent on the downbeat of m. 9.48 His recomposed version, shown in Example 16(b), eliminates the formal overlap of Chopin’s original. Presumably to make the dominant closing the antecedent more effective as a goal, Rothstein even rewrites this harmony as a triad. Though this interpretation has merit, we should observe that the final dominant still emerges out of an ongoing sequential process, as supported all the more by the melodic line, which continues the same motivic content of the sequence. In other words, no distinct cadential idea is specifically associated with this proposed HC. Rothstein, though, perhaps overlooks one detail that points to the pre-dominant of m. 8 as the real “ending” of the phrase, namely, the rhythmic alteration on the second beat of that measure from an expected dotted figure to even eighth notes. This subtle change to what is otherwise a sequential repetition helps differentiate m. 8 as a possible ending, even if it is one that cannot be construed as either cadential or prolongational. This rhythmic alteration not only serves the specific formal purpose just described but also helps prepare for the double eighth-note configuration that appears in the returning basic idea on the second beat of m. 9. Moreover, the use of two eighth notes in place of a dotted-eighth and sixteenth (as in the prior sequential measures) effects a slight sense of rhythmic deceleration that further helps project m. 8 as a potential ending moment. Indeed, a performance by Arthur Rubenstein highlights the formal arrival there by effecting a noticeable ritardando within the measure.49 ULTIMATE VS. PENULTIMATE DOMINANTS We tend to think that the fundamental difference between a half cadence and an authentic cadence involves its final harmony, being either a dominant or tonic respectively. But just as important—perhaps even more so—is the distinctively different roles that the dominant harmony itself plays in these cadential formations. For in the HC, the dominant is the ultimate harmony of the progression, whereas in an authentic cadence, the dominant serves as the penultimate harmony. Since both cadences include a dominant—always in root position—our sense of how that harmony is functioning, as ultimate or penultimate, very much determines which cadential type is being projected by the music. In the classical style, composers rarely allow this distinction to become blurred; in the Romantic style, and especially with Schubert, ambiguities about the role of the cadential dominant give rise to problematic cadential interpretations. Several passages from the first movement of Schubert’s Piano Sonata in A, D. 959, will illustrate such problems, even allowing us to identify a new type of cadential deviation, which I term the dissipated cadence.50 We begin with the main theme, built seemingly as a small ternary, and first ask how the A section ends (Ex. 17[a]). The final measure would appear to be m. 6, because the complete change in materials in m. 7 can mark the beginning of a new unit, namely, a standing on the dominant to comprise the whole of the B section,51 which stretches to the return of the opening materials at m. 16, thus initiating the A′ section. As already mentioned, the opening A section of a small ternary normally ends with a PAC; but here, the dominant seventh in m. 6 does not resolve as expected, thus giving the impression that an evaded cadence has occurred instead (see again the discussion of Ex. 5). Unlike normal classical practice, however, this cadential evasion is not followed by a one more time technique, that is, by some repetition of immediately prior material that would lead to a concluding authentic cadence. Instead, the A section seems to be left without any formal closure whatsoever. There is, of course, an alternative interpretation; namely, that the dominant of m. 6 is ultimate, not penultimate. We would then recognize a “nineteenth-century HC” concluding the A section (thus referencing the practice of Haydn, as mentioned in n. 40). Such a view also suggests that we might revise our interpretation of the immediately following mm. 7–15, hearing them as a post-cadential standing on the dominant, not as a contrasting middle (B section), and thus recognize the return of opening materials at m. 16 as the start of a potential consequent phrase, one matching an opening antecedent. Several problems, though, present themselves if we believe that m. 6 is an ultimate dominant. Unlike Chopin or Schumann, Schubert does not normally employ the dominant seventh in half-cadential situations.52 More importantly, the A′ section leads to a similar dominant at m. 21, and there, it clearly behaves as a penultimate harmony (albeit an inverted one) that resolves to tonic in the following measure, bringing prolongational closure to the small ternary form as a whole. (Just why Schubert does not place the dominant in root position to create a regular PAC at m. 22 will be addressed shortly below.) Before leaving the A section and its ambiguous closure, we should consider in detail some aspects of its formal organization and melodic-motivic content. In the first place, its six-measure length means that we are dealing with either a deviation of a standard eight-measure theme type or else some nonconventional type. The opening two measures could be considered a basic idea, one comprising two one-measure motives.53 What follows is contrasting (with strong continuational characteristics), and a cadential idea begins with the upbeat to m. 5, continuing on through m. 6, whose ambiguous status we have just examined. Thus, whereas we can recognize three ideas that express the general functions of beginning, middle, and end, a conventional formal type does not emerge. What also does not materialize is a highly profiled melodic component to these ideas. The uppermost sounding voice remains fixed on the tonic degree until the decorated resolution of the suspension in the final measure (6); the bass voice also brings a tonic pedal for four measures until it finally moves up to ③ for the start of the cadential progression in m. 5. The only substantial melodic motion occurs in the two inner voices, which, bringing a modicum of melodic motion, consist mostly of scalar patterns moving up and down from 3^ in an alto voice, doubled a third lower all the way by the tenor. In short, the texture features outer-voice pedals surrounding scalar motion in parallel thirds by the inner voices. The alto voice essentially functions as the true “upper voice” of the contrapuntal structure, with the actual soprano having a more conventional inner-voice function. If we follow the course of parallel thirds consistently, then we would want to consider the dissonant-seventh D ( 4^) as the active melody at the end of the section, which further helps to support the idea of an evaded cadence, inasmuch as this tone strives to return back to its starting point, 3^.54 Given the consistency of the doubled-third motion throughout the section, we might accord this texture a significant motivic function and observe its development throughout the rest of the movement. The 4^ that is left hanging at the end of the A section is transferred to the very top of the arpeggiated texture that takes over at the beginning of the B section (m. 7) and continues to sound as the principal melodic element in every measure until the liquidation in mm. 14–15. As for the bass voice, it now becomes more active, rising chromatically stepwise throughout the section from ⑤ to ②, such that at the fortissimo climax at m. 13, the outer voices sound the doubled third pair B–D, just the same two motivic pitches from the end of the A section in m. 6. The resolution of 4^ finally occurs at the beginning of the A′ section, when 3^ is picked up again, now transferred to the tenor voice and doubled a third lower by the bass. In addition, the top voice has a new countermelody that sounds above the parallel-third motion of the lower voices. Formally and harmonically, the A′ section is organized just like the A section, except that as discussed, the final dominant resolves to tonic, thus extending the unit by one measure. Just before that happens though, the F♯ in the upper line of the parallel third motion is doubled by the soprano in the second half of m. 20. That voice then continues the line with the whole-note D, while the tenor loses its pitch, and sounds the root of the dominant, E, just the note we might have expected to appear in the bass voice, which instead sustains the parallel third doubling. As a result, a wonderful reversal of texture occurs such that in m. 21 the upper and lower voices now sound the doubled third, while the inner voices take on their more appropriate function as fillers, including the decorated suspension. At this point we are ready to discuss the cadential situation arising at the end of the A′ section. Following the bass as it ascends from ① to ④ by the end of m. 20, we would normally be expecting the line to reach its goal on ⑤ to bring the cadential dominant. As we have seen, however, Schubert lets the bass fall down to ②, in parallel thirds with the soprano, and then to ① for the tonic resolution. Consequently, we cannot speak here of an authentic cadence, but recognize instead prolongational closure both for the section and for the theme as a whole. As already discussed, Schubert’s motivation for failing to bring a genuine cadence most likely lies in the motivic work that he relentlessly pursues throughout this section, namely, the use of parallel motion as the essential contrapuntal activity of the theme.55 Let us now turn to the subordinate-theme area of the exposition, where we continue to encounter cadential uncertainties similar to those discussed so far. Example 17(b) shows the first of two subordinate themes.56 It opens with a phrase that ends at m. 58 with an HC in the subordinate key of E major, just like an antecedent. Note the parallel thirds in the left-hand part, an obvious reference to motives from the main theme. The phrase begins to be repeated as a possible consequent, but this function “fails” when the music suddenly veers off to G major, partially confirming that key at m. 63 with the same half-cadence gesture; as a result, the failed consequent “becomes” another antecedent.57 We thus understand that more in the way of thematic activity must occur. The music that continues at m. 65 prolongs I6 of G, suggesting that an ECP may be in the making, one that would support a continuation⇒cadential phrase following upon a presentation, made up of the repeated antecedents.58 A continuation of the implied ECP is reconstructed in Example 17(c), showing how the theme could easily have cadenced in G major. At m. 70, however, the cadential progression is abandoned when the music shifts suddenly back to E major: the bass note B, which functioned as the third of the tonic of G, continues to sound (it is even emphasized by being placed an octave lower) to become the root of the dominant of E. Following a deceptive resolution to VI at m. 71, a more compressed cadential progression begins with the pre-dominant IV and continues in the next measure with the cadential dominant, one that is clearly perceived as penultimate, in that we expect an authentic cadence to close this subordinate theme, now in the normal subordinate key of E major. The cadence is evaded, however, and the music quickly backs up to the cadential six-four for another try, one that is evaded yet again at the end of m. 73. At this point, something quite unusual takes place, for the music now seems to get stuck on the cadential dominant as the piano’s right hand brings a new triplet figure, which gets drawn out and liquidated much like a standing on the dominant. Thus, the resolution to tonic at m. 78 does not sound like a cadential goal but rather is exclusively heard as a beginning harmony, one that initiates the next phrase, a return to material from the very opening of the theme. What seems to have taken place is that a penultimate dominant (one that promises an authentic cadence) is somehow converted into an ultimate dominant to support a standing on the dominant, a “post-cadential” function that entirely removes any potential for cadential closure. In association with the complete liquidation of the motivic content and a highly recessive dynamic, we can have the impression that the impending authentic cadence has been entirely dissipated; indeed, I will term this deviation a dissipated cadence. Something similar, but even more dramatic, takes place toward the end of the second subordinate theme, which begins at m. 82 with a large-scale model-sequence plan (not shown in the examples) somewhat in the manner of a developmental core, though essentially remaining in the subordinate key.59 At m. 95 (Ex. 17[d]), Schubert initiates an ECP, starting first in E minor but then shifting back to major two measures later. The conditions for a PAC literally appear at the downbeat of m. 101, but the sudden dropping out of the chordal texture and abrupt change back to the minor mode could also be a sign that the cadence is evaded. Both options seem available for analytical (and performative) interpretation, though for reasons of broader formal organization, I prefer the latter. In any case, the passage strives forward with the same music, though now the left-hand plays the running triplets and the right-hand the slower-paced chordal blocks. The texture changes again at m. 105, and the harmonic progressions are suggestive of cadential activity. A clear cadential progression emerges at m. 110, strongly implying PAC closure for the theme. This time, though, the cadence is evaded by a grand pause (recalling a similar situation in the “Unfinished” Symphony, Ex. 14[c], m. 62), and when the music picks up again with the texture reduced to a single line, we understand that the dominant harmony of m. 111, which we construe to be penultimate, continues to be prolonged in a manner that resembles the very end of a standing on the dominant, one that normally follows upon an ultimate dominant. The texture begins to fill out at m. 116 in preparation for a return of the opening phrase of the first subordinate theme, whose tonic at m. 117 sounds entirely initiating, not ending. In other words, we confront again a case of a dissipated cadence, whereby the subsequent standing on the dominant exhausts the potential for cadential closure to the extent that the resolution to tonic brings a new beginning, not a cadential goal, thus converting an initially perceived penultimate dominant into an ultimate dominant. EXAMPLE 17(A). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (a) mm. 1–26 EXAMPLE 17(A). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (a) mm. 1–26 EXAMPLE 17(B). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (b) mm. 55–82 EXAMPLE 17(B). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (b) mm. 55–82 EXAMPLE 17(C). View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (c) reconstruction of mm. 69–72 EXAMPLE 17(C). View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (c) reconstruction of mm. 69–72 EXAMPLE 17(D). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, (d) first movement, mm. 95–124 EXAMPLE 17(D). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, (d) first movement, mm. 95–124 Up to this point, we have assumed that the dominant achieved at m. 111 is penultimate, largely for reasons having to do with formal context: we normally expect that at this late stage of a subordinate theme, any cadential function would be aiming toward authentic cadential closure, especially when an evaded authentic cadence immediately preceded at m. 101. Some details of pitch organization in m. 111, however, prompt us to entertain the idea that this dominant could be perceived even initially as an ultimate one, projecting a strong sense of HC, after which the grand pause supports the perception that the music has “stopped” with that cadence. After all, no dissonant seventh appears there; indeed, the texture is ultimately reduced to octaves doubling the root of the harmony. Though an interpretation of HC is perhaps less likely in the broader formal context that precedes it, it does make sense in light of what follows, namely, a standing on that ultimate dominant. To help project such an interpretation, a performer would want to make the dominant of this measure sound like a genuine goal, which might entail a slight ritardando and diminuendo throughout the measure. Performances by van de Laar and Pollini achieve just this effect.60 In contrast, those by Kovacevich, Schnabel, and Perahia increase the drive and intensity of the music throughout the complete measure, so that the effect of an interruption—a real cadential evasion—is dramatically projected.61 In short, Schubert leaves it open for the performer to help decide such ambiguous cases. The return of the opening of the first subordinate theme at m. 117 continues to play on the question of whether a dominant is ultimate or penultimate, since this time, the opening phrase ends at m. 120 with a dominant seventh, not a dominant triad (cf. again, Ex. 17[b], m. 58). As a result, it is perhaps unclear whether or not an HC is created at this point, since, as earlier discussed, Schubert does not normally employ the nineteenth-century HC. We might rather interpret the dominant as penultimate and recognize an evaded cadence, one that will motivate a subsequent PAC via the one more time technique; in fact, a clear PAC does occur at m. 123, preceded by that same penultimate dominant. This is the cadence that finally, and unequivocally, closes the entire group of subordinate themes. We are now in a position to step back and consider the larger formal patterns that emerge within the subordinate theme area of this exposition. As just discussed, we can find only a single unequivocal PAC that could be considered to close a subordinate theme, namely the one we examined in m. 123.62 Two other potential PACs were poised to appear—one at m. 74 (Ex. 17[b]), to close a first subordinate theme, and one at m. 112 (Ex. 17[d]), to close a second subordinate theme. In each case, these cadences were effectively dissipated by the conversion of their penultimate dominants into ultimate ones in the course of their standings on the dominant. If Schubert had written genuine PACs, the return of the motives from the opening of the first subordinate theme at mm. 78 and 117 would have functioned as post-cadential codettas. In addition, the subordinate-theme group (consisting of two complete subordinate themes) would have conformed largely to the norms of classical form. But taking the dissipated cadences into consideration, whereby a penultimate dominant becomes an ultimate one, a more complex design begins take shape. Let us begin again with the first subordinate theme (Ex. 17[b]). As we discussed, the section from m. 65 to m. 77 is supported by a bass note B. At first, this bass functions as the third of a I6 harmony in G major, but then assumes its role as dominant of the subordinate key, E major. Broadly speaking, and appealing to the spirit of retrospective reinterpretation, we could consider this entire section functioning as a contrasting middle (B section) of a small ternary structure, a kind of global “standing on the dominant” (of E). The A section would thus embrace the opening antecedent (mm. 55–59) and the failed consequent (mm. 60–64).63 The final section of the form, however, poses difficulties. In the context of a small ternary reading, the return of the opening materials from the A section strongly signals the start of the A′ section, most likely taking the form of a genuine consequent phrase (to make up for the failed consequent of the A section). Following the return of the basic idea, however, the materials quickly liquidate in the manner of codettas, such that the overall ternary form never receives its required perfect authentic cadential closure. That the first subordinate theme remains unclosed raises the possibility of recognizing an even larger-scale ternary design by regarding this incomplete theme as a magnified A section (indicated by the larger, italicized symbol);64 we could then regard the second subordinate theme (the latter half shown in Ex. 17[d]) as bringing together the remaining B and A′ sections (beginning at mm. 82 and 117 respectively). After all, the second subordinate theme begins with extensive sequential activity in the manner of a developmental core, and thus from its start has a strong intrinsic medial functionality of the type that we might associate with a contrasting middle. This enormous B section would be deemed to end with the ultimate dominant that emerges via the dissipated cadence at m. 111; that harmony is then sustained by the subsequent standing on the dominant.65 The large-scale A′ section is signaled by the return of the A section’s opening phrase at m. 117, and the theme as a whole is closed by the PAC at m. 123. With this interpretation, the exposition would see only a single subordinate theme, one built as a huge small ternary, whose A section itself is structured as a lower-level small ternary (albeit one that does not achieve cadential closure). Given that the small ternary form is never used as the basis of a classical subordinate theme (much less a group of themes), the embedding of a small ternary within an even grander small ternary shows how far from classical norms Schubert is willing to go. Moreover, we see here a further expression of the formal circularity that marks a Romantic approach to form, in that we encounter two different returns of the opening materials of the theme (mm. 78 and 117), each signaling the start of an A′ section (at two different levels of the form). Let us come full circle and give one final consideration to the main theme of the exposition (Ex. 17[a]). In light of the two dissipated cadences that we have identified in the course of the subordinate theme area, we can now examine again the cadential situation at m. 6 and find there the potential source of these cadential deviations. For if we take the dominant in that measure to be penultimate, an option that we gave serious consideration to in our earlier discussion, we can recognize that an evaded cadence is followed by a standing on the dominant, one that eventually dissipates any possible sense of an upcoming authentic cadence and that prepares for the return of the opening section, just the kind of situation that we encounter twice in the subordinate theme area. When first discussing this passage, I was hesitant to label it a dissipated cadence for the main reason that I continued to consider it a possibility that m. 6 brings an ultimate dominant to create a nineteenth-century HC. In light of what occurs throughout the rest of the exposition, however, finding here the first of three dissipated cadences seems like an especially compelling interpretation.66 * * * * * With this last example, as well as with most of those presented in this study, we see how Romantic composers find ways of creating closure—or denying it, as the case may be—that go far beyond the techniques of cadence utilized by their classical predecessors. These novel cadential situations arise from some specific compositional traits of the Romantic style, especially in the domains of harmony, rhythm, texture, and form. Furthermore, we have seen that the various techniques of closure treated here are related to phrase-structural devices that are atypical of the classical style. There exist advantages, but also some drawbacks, in the methodology that I have adopted for this study. In general, I have surveyed the Romantic repertory to identify those places where composers of earlier generations would have employed the classical cadence to effect formal closure, but where instead, Romantic composers write something that deviates from the norms of eighteenth-century practice. To be sure, this methodology, which emphasizes in the first instance why a given situation of closure is not classical, can give the impression that there is something compositionally defective or abnormal, and the entire discussion can seem to be framed in largely negative terms.67 Unfortunately, this problem seems endemic to approaches that attempt to establish norms and then identify violations of these norms. Of course, if it can be shown that the norms are no longer really applicable (for example, using principles of tonal harmony to analyze twelve-tone music, to take an extreme case, though one that has occasionally been pursued), then such an approach is less compelling and potentially useless. As has already been stated, however, the classical cadence continues to be found widely in nineteenth-century music, especially at key moments of formal articulation, so employing this model as a measuring tool to gauge conformance seems entirely reasonable. Most importantly, it is not a question of using the classical norms to show what is wrong with Romantic music, but rather to highlight what is different about it. Given that so many theories of tonal music propose a “common practice period,” one that embraces multiple historical styles (baroque, classical, Romantic, post-Romantic), many musicians fail to attend to significant differences in harmonic, cadential, and formal practices among these styles. For even if a norm/deformation model tends toward generating negative statements, such a model at least helps to clarify differences in composition technique, leading ultimately to an enriched hearing experience of the music. I would like to think that when we focus on what may seem on the surface to be small differences in harmonic usage, for example, the inclusion of a dissonant seventh at a moment of HC, the inversion of the dominant in a case of prolongational closure, or whether a given dominant is ultimate or penultimate, then we can begin significantly to refine our hearing and to appreciate how the different styles of tonal music are expressed. Eventually, of course, we would like to formulate sound, positive reasons for why composers of a given style found it entirely acceptable to break the compositional norms developed in earlier practices. For myself, I feel that I have yet to reach that goal for issues of formal closure. As a first step, though, I am content enough to identify differences in cadential practice and am hopeful that other current and future research will provide a more comprehensive, integrated perspective on these vital issues of musical composition.68 Footnotes 1 Caplin (2004). 2 The selection of music undertaken for this study—a rather eclectic sampling of instrumental works in a variety of genres—yielded examples of specifically Romantic cadential usage primarily in compositions by Schubert, Schumann, and Chopin. Few cases arose for the classicizing Mendelssohn, though one rather exceptional situation is cited in nn. 41 and 44 below. The highly eccentric formal processes witnessed in Berlioz normally involve dimensions other than cadence; in general, it seems, French composers of the nineteenth century hold firm to classical cadential practices. 3 Note that in this example, and throughout the remaining ones, the boundaries of the cadential progression are indicated by a horizontal bracket positioned below the Roman numerals. 4 Caplin (2004, 70); Caplin (2013b, 125–27) argues this point more fully, citing empirical, theoretical, and experiential grounds for this fundamental axiom. 5 I classify sequential progressions on the basis of a consistent pattern of harmonic root motion (descending fifth, ascending step, etc.); moreover, embellishing harmonies can sometimes appear between more fundamental harmonies of the sequences, as seen in Exx. 3(c) and 3(d). 6 Now and then a tonic prolongation can appear with both dominant and tonic chords in root position. Of course, the same progression could also function as an incomplete cadential progression. The potential ambiguity in classifying such a progression can normally be resolved contextually, that is, by considering whether the progression occurs in a formal situation that is either initiating or closing. 7 Caplin (2004, 56–66). This view thus challenges the claim by Hepokoski and Darcy (2006) that both a sonata exposition and an overall sonata form are closed by specific cadences (their “essential expositional closure” and “essential structural closure,” respectively). 8 Caplin (2004, 89–96). 9 Caplin (1998, 101–7), Caplin (2013a, Chapter 5). A deceptive cadence, using a deceptive cadential progression such as that shown in Ex. 1(c), brings a final cadential event that is supported by a different harmony than the required root-position tonic. An abandoned cadence results when the penultimate dominant either fails to appear or becomes inverted prior to its resolution to the tonic. 10 Schmalfeldt (1992). 11 Note that I deliberately speak of cadential closure of a movement’s final theme, thus emphasizing that cadential closure is effectively limited to the hierarchical level of thematic organization and does not normally apply to higher levels of form (see n. 7). 12 A second source of nineteenth-century chromaticism involves the use of symmetrically organized harmonic structures, ones that are often modeled by Neo-Riemannian principles. Passages featuring these harmonies will not arise in this study, since they rarely involve mechanisms of thematic closure, where the expression of functional tonality is paramount. 13 Schmalfeldt (2011, 202–3). Considerably more work is needed to flesh out aspects of the nineteenth-century HC, investigating, in particular, which composers favor the technique and which do not (for example, cases arise now and then with Mendelssohn, whereas Schubert, as will be discussed later on, tends most often to use the classical HC ending on a dominant triad). In two recent studies, L. Poundie Burstein argues that ample precedent already exists in the classical style for regarding HCs as ending with a dominant seventh (and even being inverted) (Burstein 2014, 211–14; Burstein 2015, 97–105). Here is not the place to evaluate this claim, which is based on fundamentally different assumptions about cadence than those held by Schmalfeldt and myself. I would only note now that almost any feature of a particular musical style has precedents in an earlier style; the main question is, are such precedents central or peripheral to the prior one? I would argue that only a very limited number of classical themes bring a half-cadential articulation that genuinely includes a dissonant seventh; as such these cases should be understood as a cadential deviation, one that I term dominant arrival (see Caplin 1998, 79–81 and Caplin 2013a, 224–25). I would thus support Schmalfeldt’s assertion that the inclusion of a dissonant seventh first becomes an essential element of the HC in the early Romantic style. (I will address Burstein’s viewpoint and present a fuller examination of the harmonic content of the HC in a book-length project tentatively entitled Cadence: A Study of Closure in Tonal Music.) 14 Formal descriptions throughout this article follow the categories developed in Caplin (1998) as well as the more recent textbook version, Caplin (2013a). For the discussion of compound themes, see Caplin (1998, 63–70) and Caplin (2013a, Chapter 6). 15 Though not related to aspects of closure in this theme, the specific harmonies of this opening phrase are worth considering from the perspective of the increased chromaticism and dissonance associated with early Romantic style; with this bass line, a standard classical presentation phrase would bring the prolongational progression I– V34–I6. Here, the passing dominant is further embellished with the more dissonant VII56 (m. 19) and the I6 is replaced, most unusually, by a dissonant VI46 (m. 20), which is chromaticized as V56 of IV at the beginning of the continuation⇒cadential phrase. As a result of the irregular resolution of VII56 in m. 19, this diminished-seventh sonority actually assumes a “common-tone” voice leading, though its dominant function still makes sense in light of the VI46 being a tonic substitute harmony. 16 When they do occur in the earlier styles, such progressions are typically part of a broader sequence, e.g., a descending-fifth pattern, as discussed ahead in Ex. 13. 17 For other examples, see Chopin, Nocturne in G Minor, Op. 15, No. 3, mm. 89ff. (marked “religioso”), and Chopin, Prelude in C Minor, Op. 28, No. 20. 18 As I consider the subordinate harmonies in mm. 1–2 to occur in the context of a tonic pedal, they should not be identified as being in any particular “inversion”; however, I have retained the labels for inversions in order to acknowledge the alternative, first mentioned in the text, of hearing more literal changes of harmony within these measures. 19 This interpretation assumes that the seventh literally entering on the second beat of the measure genuinely belongs to the dominant harmony, which appears as an incomplete sonority on the downbeat. The seventh E♭ (along with the ninth G) held over from the previous chord suggests this harmonic reading. If we hear the seventh as passing, however, then the implied dominant triad supports a classical HC. 20 See Caplin (2008) for discussion of this bass-line model. Following the practice of Sanguinetti (2012) (inspired by Gjerdingen 2007), I will indicate scale degrees in the bass voice by a circled numeral and indicate scale degrees in the upper parts with a circumflex accent. 21 The cadential progression of mm. 7–8 could also be seen as a descending-fifth sequential progression, a potentially ambiguous situation that I treat in a later section on “sequence versus cadence.” 22 A significant difference between the baroque Affektenlehre and the early Romantic portrayal of “character,” however, lies in the conventionalized nature of the former compared to the more individualistic sense of personal expression in the latter. 23 Schmalfeldt (2011, Chapter 8) identifies this sequential pattern as particularly characteristic of Chopin, though here we see the sequence used by Schumann as well. The pattern is rare in the classical style, especially within a presentation phrase. 24 In the classical style, such a presentation would normally be supported by a single tonic prolongational progression. 25 See also Chopin, Prelude in C Minor, Op. 28, No. 20, mm. 1–2. 26 See Caplin (2013a, 143) on abandoned cadential progressions as opposed to abandoned cadences per se; on the latter, see Caplin (2013a, 132–33). 27 The idea of circular form in the nineteenth century is discussed by Morgan (2000) and Rodgers (2009). 28 We find in this example another instance of the I–III–V sequential pattern underlying the entire thematic unit, as discussed earlier in n. 23. 29 Meyer (1989, 26–30); see also Levy (1981, 28). Example 11 illustrates the A′ section of the main theme, an overall small ternary form (rounded binary version). 30 Meyer (1989, 26) cites Haydn’s String Quartet in B♭ Major, Op. 64, No. 3, iii, and Mozart’s Symphony No. 40 in G Minor, iii. 31 Haydn uses the same idea to close the earlier A section, with a dominant arrival at m. 7, and brings the idea back again to confirm the modulation to the dominant region in the middle of the B section, m. 24. 32 The use of a “characteristic” opening gesture as a subsequent cadential idea continues to be found in later nineteenth-century repertories; the first subordinate theme from the opening movement of Tchaikovsky’s Pathétique Symphony (cf. mm. 89–91 and 95–97) is especially exemplary of this practice. 33 The confusion of sequence with cadence arises essentially with the descending-fifth (“circle of fifths”) sequence, since this is the only sequence type that yields a potentially cadential V–I progression at its end. 34 The cello then “echoes” this motive in beats 2 and 3 of mm. 53ff., but now as an accompanimental figure, not as a genuinely melodic element. 35 Compare again Ex. 5, m. 16. Not all classical evaded cadences feature the one more time technique, though many do. Yet almost all such evasions at least bring some sounding event in place of the cadential arrival. Schubert’s evaded cadence is thus most bizarre from a classical perspective. We will see another such evaded cadence in a later discussion of the composer’s Piano Sonata in A, D. 959 (see Ex. 17[d], m. 112 below). 36 See Macdonald (1978) for a discussion of Schubert’s “volcanic temper.” 37 This return of opening subordinate-theme materials exemplifies again the Romantic impulse for formal circularity. In the classical style, by contrast, once subordinate-theme ideas are eliminated from the theme, they rarely return either in the context of the theme group itself or in the subsequent closing section. I will discuss another use of this technique below in connection with Ex. 17(b), m. 78 and Ex. 17(d), m. 117. 38 See Caplin (2004, 59). The compound basic idea (a basic idea plus a contrasting idea supported by a prolongational progression) is another four-measure phrase that does not close with a cadence; likewise, a continuation phrase may end with no cadential closure when, in looser formal contexts (such as a subordinate theme), it is followed by a unique phrase serving an exclusively cadential function. 39 Many theorists describe what I am calling prolongational closure as a weaker form of cadence proper; the terms “contrapuntal cadence” or “imperfect cadence” are regularly encountered in this context. As discussed on several occasions already, I restrict the concept of cadence to cases where both the dominant and final tonic of the cadential progression appear exclusively in root position; see again nn. 4 and 13. I fully acknowledge that a particular type of formal closure can be effected by concluding a prolongational progression, but I understand it as one that is categorically—and, more important, experientially—distinct from closure that is specifically cadential in nature. 40 Haydn now and then ends the A section of a small ternary (or rounded binary) with an HC; Mozart and Beethoven almost always conclude that section with a PAC. 41 See Caplin (2008) for the essentially ascending contour of a classical bass line. The propensity for bass lines to descend, especially at the beginning of a theme, is characteristic of the baroque style; see Lester (1999, 27–33, esp. Ex. 2.3 and Ex. 2.5) and Caplin (2014, 436). For another example of a thematic unit ending with prolongational closure, due primarily to the use of a falling bass line, see Mendelssohn, Song without Words, Op. 19, No. 1, mm. 13–15, where the penultimate dominant initially appears inverted before taking on its more conventional root position just prior to resolving to tonic. 42 The cadence at m. 40 is perfect authentic because the melody, lying below the “covering tone” F, closes on B♭ ( 1^). 43 Horton (2015, 103–7) develops the same basic point; see his insightful discussion of how the main theme in the solo exposition of Chopin’s Piano Concerto in F Minor, Op. 21, first movement, fails to close cadentially yet still projects a clear sense of thematic ending by means of prolongational progressions. 44 The Mendelssohn Song without Words cited in n. 41 is thus highly exceptional in that the entire piece concludes (mm. 42–44) with the same prolongational closure described for mm. 13–15. Later in the nineteenth century, we encounter more frequent use of prolongational closure for a complete movement (or larger section of one); my forthcoming study on cadence in various tonal styles includes a chapter on cadential practices of the mid to late nineteenth century. 45 More precisely, we could speak of the sequential progression as actually beginning in mm. 3–4 with the roots E and A and then continuing into the opening of the consequent with the roots C♯ and F♯. For the purposes of this discussion, I am focusing specifically on the sequential segment supporting the antecedent’s continuation phrase (mm. 5–8). 46 As mentioned in n. 6, a progression from root-position dominant to root-position tonic can be sometimes emerge as prolongational, not cadential. 47 See also Schmalfeldt (2011, 198). 48 Rothstein (1989, 46–48). 49 RCA Red Seal 09026 63050-2, compact disk. 50 Another work by Schubert, the first movement of his Sonata in A Minor, Op. 42, D. 845, features a number of ambiguous cases of ultimate and penultimate dominants; Schmalfeldt (2011, Chapter 5) discusses some of these cadential ambiguities, especially as they affect performance decisions. 51 See Caplin (1998, 75–77) and Caplin (2013a, 210–11) for the conventional use of a standing on the dominant in the contrasting middle of the small ternary form. 52 Schmalfeldt (2011) does not identify any nineteenth-century half cadences in connection with Schubert, only raising the topic first with her chapter on Chopin; see again, n. 13. 53 See Caplin (1998, 37) and Caplin (2013a, 39) for cases in which an opening basic idea consists of a repeated one-measure motive. 54 I am regarding all of the inner-voice Es in the chord of m. 6 as doublings of the bass voice. 55 There are perhaps additional reasons for Schubert’s not having placed an E in the bass: if the bass line of m. 20 were to continue to rise to E3, the texture would be extremely thin, and thus not appropriately balanced with the texture of m. 6. Moreover, it is not possible to add the lower E2 while still retaining D5 in the upper voice and an octave-doubled decorated resolution in the inner parts, since two hands cannot play such a configuration of notes. Finally, the upper D ( 4^) in m. 21 would awkwardly clash against the lower E (⑤), a voice-leading situation that, if not strictly prohibited, is normally shunned in such an exposed manner. 56 The discussion of Exx. 17 (b) and (c) largely reproduces that given in Caplin (1987, 246–49). 57 On the notion of failed consequent, see Caplin (1998, 89) and Caplin (2013a, 246–47). Somewhat problematic in characterizing this second phrase as an antecedent is its modulating structure; normal antecedents in a classical period do not modulate, as discussed earlier in connection with the modulating antecedent of Ex. 8(b). 58 In this broader sentential context, the half cadences ending the antecedent phrases are thus deemed to have “limited scope” (see Caplin, 2004, 86–89). 59 See de Médicis (2015, 205) for a discussion of the infusion of core-like materials within subordinate themes of Schubert’s expositions. The use of immediate model–sequence activity at the start of the theme also has clear classical precedent, since a second subordinate theme occasionally begins directly with continuation function, thus bypassing an initiating unit; see Caplin (1998, 111–13) and Caplin (2013a, 391). 60 Brilliant Classics 99678/2, compact disk (van de Laar); Deutsche Grammophon 4713502, vol. 6, compact disk (Pollini). 61 EMI Records, compact disk (Kovacevich); Pearl CDS 9271, compact disk (Schnabel); Sony Classical S2K 87706, compact disk (Perahia). 62 I am discounting the possible PAC at m. 101, since no coherent view of the form ensues from interpreting this as a moment of complete tonal and formal closure; instead, I recognize an evaded cadence there for the reasons cited earlier. 63 Once again (as mentioned in connection with the main theme), we would be confronting the exceptional case of an A section closing with an HC rather than a PAC, but I have already discussed a classical precedent for this situation in Haydn (see n. 40). 64 The expression “large-scale small ternary” might seem odd; why not speak of a “large ternary”? The problem, of course, is that the latter is itself a distinct classical formal type, one that diverges significantly from the small ternary; see Caplin (1998, 211–16) and Caplin (2013a, 566–69, 574–86) for my definition of the large ternary form and how it differs from the small ternary. 65 Or, if we believe that m. 111 brings a regular HC, shown as option 2 in the analysis, then we would not speak of a dissipated cadence. 66 A look at the coda of the movement further supports this reading. Here we find that Schubert twice brings back, in an extended form, the opening thematic unit (A or A′) and closes both with a classical perfect authentic cadence (mm. 340 and 349). As well, the dominant seventh of each cadence includes the same decorated resolution of the 4–3 suspension that we find in mm. 6 and 21, thus suggesting that these earlier moments, too, can be considered penultimate dominants. 67 See Vande Moortele (2013, 5–8) for a highly insightful discussion of the dilemma of approaching nineteenth-century form from a “positive” or “negative” perspective. 68 In addition to the studies by de Médicis, Horten, Morgan, Rodgers, Schmalfeldt, and Vande Moortele cited above, important recent contributions by Aziz (2015), Black (2015), Davis (2014), Hyland (2009), Koslovsky (2014), Martin (2010), Martin and Vande Moortele (2014), Smith (2014, 2016), and Suurpää (2011) are using in creative, “positive” ways some of the methodologies of the recently revived Formenlehre for eighteenth-century music to help enrich our understanding of early nineteenth-century form and cadence. Works cited Aziz Andrew I. 2015 . “The Evolution of Chopin’s Sonata Forms: Excavating the Second Theme Group.” Music Theory Online 21 ( 4 ). Accessed August 10, 2016. http://www.mtosmt.org/issues/mto.15.21.4/mto.15.21.4.aziz.html. Black Brian. 2015 . “Schubert’s ‘Deflected-Cadence’ Transitions and the Classical Style.” In Formal Functions in Perspective . Ed. Moortele Steven Vande , Pedneault-Deslauriers Julie , Martin Nathan John . 165 – 97 . Rochester, NY : University of Rochester Press . Burstein L. Poundie. 2014 . “The Half Cadence and Other Such Slippery Events.” Music Theory Spectrum 36 ( 2 ): 203 – 27 . Google Scholar CrossRef Search ADS ———. 2015 . “The Half Cadence and Related Analytic Fictions.” In What Is a Cadence? Theoretical and Analytical Perspectives on Cadences in the Classical Repertoire . Ed. Markus Neuwirth , Bergé Pieter . 85 – 116 . Leuven : Leuven University Press . Caplin William E. 1987 . “The ‘Expanded Cadential Progression’: A Category for the Analysis of Classical Form.” Journal of Musicological Research 7 ( 2–3 ): 215 – 57 . Google Scholar CrossRef Search ADS ———. 1998 . Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven . New York : Oxford University Press . ———. 2004 . “The Classical Cadence: Conceptions and Misconceptions.” Journal of the American Musicological Society 57 ( 1 ): 51 – 117 . CrossRef Search ADS ———. 2008 . “Schoenberg’s ‘Second Melody,’ or, ‘Meyer-ed’ in the Bass.” In Communication in Eighteenth-Century Music . Ed. Danuta Mirka , Agawu Kofi . 160 – 88 . Cambridge : Cambridge University Press . ———. 2013a . Analyzing Classical Form: An Approach for the Classroom . New York : Oxford University Press . ———. 2013b . “Teaching Classical Form: Strict Categories versus Flexible Analyses.” Tijdschrift voor Muziektheorie 18 ( 3 ): 119 – 35 . ———. 2014 . “Topics and Formal Functions: The Case of the Lament.” In The Oxford Handbook of Musical Topics . Ed. Danuta Mirka . 415 – 52 . New York : Oxford University Press . Davis Andrew. 2014 . “Chopin and the Romantic Sonata: The First Movement of Op. 58.” Music Theory Spectrum 36 ( 2 ): 270 – 94 . Google Scholar CrossRef Search ADS de Médicis François. 2015 . “Heavenly Length’ in Schubert’s Instrumental Music.” In Formal Functions in Perspective . Ed. Moortele Steven Vande , Pedneault-Deslauriers Julie , Martin Nathan John . 198 – 222 . Rochester, NY : University of Rochester Press . Gjerdingen Robert O. 2007 . Music in the Galant Style . New York : Oxford University Press . Hepokoski James , Darcy Warren . 2006 . Elements of Sonata Theory: Norms, Types, and Deformations in the Late-Eighteenth-Century Sonata . New York : Oxford University Press . Google Scholar CrossRef Search ADS Horton Julian. 2015 . “Formal Type and Formal Function in the Postclassical Piano Concerto.” In Formal Functions in Perspective . Ed. Moortele Steven Vande , Pedneault-Deslauriers Julie , Martin Nathan John . 77 – 122 . Rochester, NY : University of Rochester Press . Hyland Anne M. 2009 . “Rhetorical Closure in the First Movement of Schubert’s Quartet in C Major, D. 46: A Dialogue with Deformation.” Music Analysis 28 ( 1 ): 111 – 42 . Google Scholar CrossRef Search ADS Koslovsky John. 2014 . “Timeless Reflections: Form, Cadence and Tonal Structure in the Scherzo and Finale of Schubert’s String Quintet.” Music Analysis 33 ( 2 ): 168 – 93 . Google Scholar CrossRef Search ADS Lester Joel. 1999 . Bach’s Works for Solo Violin: Style, Structure, Performance . New York : Oxford University Press . Google Scholar CrossRef Search ADS Levy Janet M. 1981 . “Gesture, Form, and Syntax in Haydn’s Music.” In Haydn Studies: Proceedings of the International Haydn Conference: Washington, D. C., 1975 . Ed. Larsen Jens Peter , Serwer Howard , Webster James . 355 – 62 . New York : W. W. Norton . Macdonald Hugh. 1978 . “Schubert’s Volcanic Temper.” Musical Times 119 : 949 – 52 . Google Scholar CrossRef Search ADS Martin Nathan John. 2010 . “Schumann’s Fragment.” Indiana Theory Review 28 : 85 – 109 . Martin Nathan John , Vande Moortele Steven . 2014 . “Formal Functions and Retrospective Reinterpretation in the First Movement of Schubert’s String Quintet.” Music Analysis 33 ( 2 ): 130 – 55 . Google Scholar CrossRef Search ADS Meyer Leonard B. 1989 . Style and Music: Theory, History, and Ideology. Philadelphia : University of Pennsylvania Press . Morgan Robert P. 2000 . “Circular Form in the ‘Tristan’ Prelude.” Journal of the American Musicological Society 53 ( 1 ): 69 – 103 . Google Scholar CrossRef Search ADS Rodgers Stephen. 2009 . Form, Program, and Metaphor in the Music of Berlioz . Cambridge : Cambridge University Press . Google Scholar CrossRef Search ADS Rothstein William. 1989 . Phrase Rhythm in Tonal Music . New York : Schirmer . Sanguinetti Giorgio. 2012 . The Art of Partimento . New York : Oxford University Press . Schmalfeldt Janet. 1992 . “Cadential Processes: The Evaded Cadence and the ‘One More Time’ Technique.” Journal of Musicological Research 12 ( 1–2 ): 1 – 52 . Google Scholar CrossRef Search ADS ———. 2011 . In the Process of Becoming: Analytic and Philosophical Perspectives on Form in Early Nineteenth-Century Music . New York : Oxford University Press . Smith Peter H. 2014 . “Schumann’s Continuous Expositions and the Classical Tradition.” Journal of Music Theory 58 ( 1 ): 25 – 56 . Google Scholar CrossRef Search ADS ———. 2016 . “Cadential Content and Cadential Function in the First-Movement Expositions of Schumann’s Violin Sonatas.” Music Theory and Analysis 3 ( 1 ): 27 – 52 . CrossRef Search ADS Suurpää Lauri. 2011 . “Non-Congruent Temporal Functions in Chopin’s Nocturne Op. 62, No. 2.” Tijdschrift voor Muziektheorie 16 ( 1 ): 64 – 71 . Vande Moortele Steven. 2013 . “In Search of Romantic Form.” Music Analysis 32 ( 3 ): 404 – 31 . © The Author(s) 2018. Published by Oxford University Press on behalf of The Society for Music Theory. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Music Theory Spectrum Oxford University Press

Beyond the Classical Cadence: Thematic Closure in Early Romantic Music

Music Theory Spectrum , Volume Advance Article (1) – Apr 7, 2018

Loading next page...
 
/lp/ou_press/beyond-the-classical-cadence-thematic-closure-in-early-romantic-music-VnnR6YyxGR
Publisher
Society for Music Theory
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The Society for Music Theory. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.
ISSN
0195-6167
eISSN
1533-8339
D.O.I.
10.1093/mts/mty002
Publisher site
See Article on Publisher Site

Abstract

Abstract Building on my 2004 article “The Classical Cadence: Conceptions and Misconceptions,” this article examines the fate of the classical cadence in the hands of the first generation of Romantic composers (Schubert, Chopin, Schumann). It identifies seven characteristics of Romantic compositional style that, compared to classical practice, bear on issues of cadential morphology and function: (1) a more extensive use of chromaticism and dissonance; (2) a greater emphasis on root-position harmonies; (3) a more uniform harmonic rhythm and harmonic density; (4) a circularity of formal organization; (5) an ambiguity between sequential and cadential harmonies; (6) a lack of cadential closure for thematic units; and (7) an ambiguity between penultimate and ultimate dominants at points of potential cadence. Though I focus largely on cadence and closure, I also take the opportunity of pointing out other aspects of musical form that distinguish Romantic practice from the earlier classical style. In my 2004 article “The Classical Cadence: Conceptions and Misconceptions,” I argued that musical cadence in general was poorly understood by many musicians and that one way of regaining control over the idea is to examine cadence in the context of a relatively narrow, stylistically unified repertory.1 In particular, I advocated that the concept of cadence is best conceptualized at first in relation to music of the high classical style (Haydn, Mozart, and Beethoven). I concluded my study by suggesting that a greater involvement by scholars in theories of musical form could lead to a better understanding not only of the classical cadence but also of cadence as manifest in earlier and later repertories as well. The present article extends my approach by exploring cadential processes found in the instrumental music of first-generation Romantic composers, such as Schubert, Mendelssohn, Schumann, Chopin, and Berlioz.2 Before turning to this repertory, however, I should first summarize the essential features of the classical cadence, since not only does music of the nineteenth century continue to employ that cadence on a regular basis, but also some novel cadential techniques introduced in the Romantic period are best understood in relation to prior classical practice. Central to my definition of the classical cadence is the idea that it is supported by a specific type of harmonic progression (Ex. 1[a]), one that features an initial tonic (usually in first inversion), a pre-dominant harmony (typically II6, but sometimes IV), a penultimate dominant in root position (often embellished by the cadential six-four), and a final tonic, also in root position.3 Incomplete forms of this authentic cadential progression may sometimes eliminate either the initial tonic or pre-dominant (or both), but the dominant must always first appear in root position and remain as such. Indeed, this notion is axiomatic to my theory of cadence, for I believe that any inversion of the dominant undermines the potential of such a progression to be truly cadential.4 In the case of a half-cadential progression (Ex. 1[b]), the final harmony is an ultimate dominant, which must take the form of a consonant triad in root position. We can also identify a deceptive cadential progression (Ex. 1[c]), in which the final tonic is replaced by some other harmony, typically VI. EXAMPLE 1. View largeDownload slide Cadential progressions (Caplin 2013a, 4) EXAMPLE 1. View largeDownload slide Cadential progressions (Caplin 2013a, 4) In my own particular take on harmonic theory, I distinguish such cadential progressions from two other progression types, which I term prolongational and sequential (specific cases of these are shown in Exx. 2 and 3).5 Note that prolongational progressions often bring inverted dominants embellishing the tonic harmony.6 In addition to its harmonic content, a classical cadence typically features a conventionalized melody, most often of descending contour. Example 4 illustrates some representative melodic-motivic ideas that clearly express to the listener “this is a cadence.” EXAMPLE 2. View largeDownload slide Prolongational progressions (Caplin 2013a, 3) EXAMPLE 2. View largeDownload slide Prolongational progressions (Caplin 2013a, 3) EXAMPLE 3. View largeDownload slide Sequential progressions (Caplin 2013a, 6) EXAMPLE 3. View largeDownload slide Sequential progressions (Caplin 2013a, 6) EXAMPLE 4. View largeDownload slide Classical cadences: (a) Beethoven, Piano Sonata in C, Op. 2, No. 3, third movement, mm. 15–16; (b) Haydn, Piano Trio in D, H. 16, third movement, mm. 7–8; (c) Mozart, Piano Sonata in B♭, K. 333, second movement, mm. 20–21; (d) Haydn, Piano Sonata in B Minor, H. 32, second movement, mm. 9–10; (e) Haydn, Piano Sonata in A, H. 30, third movement, mm. 7–8; (f) Beethoven, Piano Sonata in C Minor, Op. 10, No. 1, third movement, mm. 3–4. EXAMPLE 4. View largeDownload slide Classical cadences: (a) Beethoven, Piano Sonata in C, Op. 2, No. 3, third movement, mm. 15–16; (b) Haydn, Piano Trio in D, H. 16, third movement, mm. 7–8; (c) Mozart, Piano Sonata in B♭, K. 333, second movement, mm. 20–21; (d) Haydn, Piano Sonata in B Minor, H. 32, second movement, mm. 9–10; (e) Haydn, Piano Sonata in A, H. 30, third movement, mm. 7–8; (f) Beethoven, Piano Sonata in C Minor, Op. 10, No. 1, third movement, mm. 3–4. But now we must distinguish fundamentally between the cadential content of a musical idea and its actual formal function. For in the classical style, cadences serve to provide formal closure at specific levels in the musical hierarchy, namely the middleground levels of phrase and theme. And for a cadential idea to create closure, it must follow something that functions as a formal beginning or a formal middle. Cadence is therefore a mechanism of ending, though, as will be discussed later on, not all formal endings involve cadences. Likewise, closure at very low hierarchical levels (such as that of a two-measure basic idea) and closure at high, or background, levels (such as the entire exposition section of sonata form) are not normally cadential at all, but rely on other formal processes to effect the sense of ending.7 Another distinction central to my concept of cadence is that between cadential function, a time-span leading up to the moment of cadential arrival, and post-cadential function, a time-span that follows the arrival and prolongs the final harmony of the cadence. Such post-cadential activity normally consists of a closing section (made up of codettas) following an authentic cadence or a standing on the dominant following a half cadence (HC).8 In the course of a formal process leading to cadential closure, a promised authentic cadence frequently fails at first to materialize, only to be fully realized eventually by a genuine cadence. For the classical style, I have identified three such cadential deviations (deception, evasion, and abandonment), of which the evaded cadence is most relevant for the present study.9 A cadence is evaded when the event that follows the penultimate dominant seems not to represent the final event of the cadential process, no matter what its harmonic support. In other words, this event does not group backwards with the preceding events of the cadence (those supported by an initial tonic, pre-dominant, and penultimate dominant), but rather groups forward with the subsequent events. A clear case of cadential evasion is shown in Example 5, where a perfect authentic cadence (PAC) is promised on the downbeat of m. 16. What occurs at this moment, however, does not represent the goal of the ongoing cadential process; even though supported by a root-position tonic, this event marks the onset of a new phrase, which brings the cadential idea “one more time”10 to close the theme at m. 19. EXAMPLE 5. View largeDownload slide Mozart, Piano Sonata in C, K. 309, third movement, mm. 13–19 (Caplin 2013a, 142) EXAMPLE 5. View largeDownload slide Mozart, Piano Sonata in C, K. 309, third movement, mm. 13–19 (Caplin 2013a, 142) With this all-too-brief introduction to the classical cadence, let us turn to the main topic at hand, namely, how phrase and thematic closure is realized in the early Romantic period (roughly the second quarter of the nineteenth century). First, though, we must acknowledge a central fact—Romantic composers continue to employ the classical cadence throughout their compositions. Many phrases, and most thematic units, close cadentially in the manner just described. A complete movement, moreover, is almost always marked by cadential closure of its final theme.11 Yet, despite the powerful influence that the classical cadence continues to exert, composers in the Romantic period explore some new compositional procedures that modify or distort in interesting ways the classical means of creating closure. In my research to date, I have identified seven stylistic characteristics of Romantic compositions that have the potential to affect the ways in which cadence and closure are achieved. Compared to classical practice, Romantic works may feature: (1) a more extensive use of chromaticism and dissonance; (2) a greater emphasis on root-position harmonies; (3) a more uniform harmonic rhythm and harmonic density; (4) a circularity of formal organization; (5) an ambiguity between sequential and cadential harmonies; (6) a lack of cadential closure for thematic units; and (7) a confusion between penultimate and ultimate dominants at points of potential cadence. The remainder of this article will examine these seven points in greater detail, with specific examples drawn from the Romantic repertory. Though I focus largely on cadence and closure, I also take the opportunity, where appropriate, of highlighting other aspects of musical form that distinguish Romantic practice from the earlier classical style. CHROMATICISM AND DISSONANCE That music in the nineteenth century witnesses a marked increase in the use of chromaticism and dissonance over that of the eighteenth is widely accepted. Indeed, we might think that greater chromaticism would significantly impact, perhaps even distort, cadential expression. In fact, this seems not to be the case for the Romantic generation. Most often, chromatic alterations of a given harmony can be understood in relation to a basic diatonic model, and the harmonic function defined by that model is normally retained in the chromatic version. Thus, even if a given cadential harmony is extensively chromaticized, its harmonic function within the progression remains comprehensible, and its cadential effect can be fully realized. Similarly, the addition of various dissonances does not normally disrupt the functional expression of the cadential harmonies.12 One manifestation of greater dissonance treatment, however, does have significant cadential implications. In the classical style, the final harmony of both the authentic cadence and the half cadence is normally a consonant triad, either a tonic or a dominant, in order that this final sonority has sufficient stability to express a sense of ending. In the Romantic period, this restriction begins to be eased for the ultimate dominant of the half-cadential progression. With Chopin and Schumann especially, the effect of an HC continues to be projected even when the final harmony is represented by a dominant seventh. For this reason, Janet Schmalfeldt has specifically identified a nineteenth-century half cadence at points where a classical HC would be expected (such as at the end of an antecedent phrase) but in which the ultimate dominant contains a dissonant seventh already at the moment of cadential arrival.13 A typical case is found in Chopin’s Waltz in A Minor (Ex. 6). The overall form of this theme is a compound period, whose opening compound antecedent is structured as a simple sentence.14 Following a regular presentation phrase (made up of a repeated basic idea),15 m. 21 brings the start of an expanded cadential progression (ECP), with V56 of IV substituting for the more standard I6 (shown in brackets). As the pre-dominant IV6 in m. 23 moves to the dominant in the following measure to produce the final harmony of the phrase, the pitch D is held over, thus creating a clear dissonant seventh within the dominant of this nineteenth-century HC. Not to consider this moment a type of HC seems overly restrictive, since Chopin regularly employs the dissonant seventh in this kind of formal context. This cadence nonetheless represents a clear departure from classical practice, where the use of such a dissonance for the final cadential harmony is rare. (I return to this example to discuss its consequent phrase after introducing some other characteristics of the Romantic style.) EXAMPLE 6. View largeDownload slide Chopin, Waltz in A Minor, Op. 34, No. 2, mm. 17–36 EXAMPLE 6. View largeDownload slide Chopin, Waltz in A Minor, Op. 34, No. 2, mm. 17–36 ROOT-POSITION HARMONIES One striking feature of early Romantic harmony has largely gone unremarked by theorists: harmonies frequently succeed each other in root position. In fact, a long string of root-position chords seldom arises in baroque, pre-classical, and classical works, yet such a succession is regularly found in Romantic ones.16 This emphasis on root-position harmonies has an impact on the specific harmonic progressions that tend to be used cadentially. Example 7 illustrates some typical cadential progressions found in Romantic works. Note that the pre-dominant harmony tends to favor either the subdominant or the supertonic in root position (a and b), rather than the more typical II6 of the classical style. The initial tonic of nineteenth-century cadences is more often placed in root-position (a and b) than in classical practice, and this harmonic function may be embellished by a root-position VI, which may even fully substitute for that tonic (c). Another initial tonic substitute sees the third degree in the bass (which would normally support I6) holding a III harmony, or even chromatically, a dominant of the submediant (V7 of VI), which resolves deceptively to the subdominant (d). The resulting bass line resembles a classical one, but the chromaticism betrays a Romantic touch. EXAMPLE 7. View largeDownload slide Nineteenth-century cadential progressions EXAMPLE 7. View largeDownload slide Nineteenth-century cadential progressions Example 8(a), an excerpt from Chopin’s Nocturne in G Minor, shows an extreme case of root-position harmonic succession in a passage that obviously references the genre of chorale, though clearly not in the style of Bach. The final beat of m. 44 seems to be a clear enough perfect authentic cadence (PAC), which itself is preceded by a deceptive resolution of the dominant. And we cannot entirely rule out the expression of a PAC at the end of m. 42 as well. In both places, however, the cadential progression is quite incomplete and, especially in the case the opening two-measure unit, we might want to regard the final two beats of m. 42 as continuing the prolongational progressions of the prior measure and a half. Indeed, the excessive use of root-position harmonies helps to blur the fundamental distinctions among prolongational, sequential, and cadential progressions that are so essential to classical harmony in its relation to form. Although this example of chorale texture is not unique in Chopin’s oeuvre, it still remains somewhat exceptional.17 Yet the root-position emphasis found in his chorale textures is also present at the very opening of this same Nocturne (Ex. 8[b]), which features a more standard melody and accompaniment texture. We might initially be tempted to identify various harmonic inversions within the first two measures (VII7, VI6, VII56); however, the functional bass seems to sound in a lower tessitura of the piano. If so, we can hear the opening two measures as projecting a tonic pedal (the seeming dominant pedal thus residing in a tenor voice) and then observe that all of the harmonies in the theme stand in root position, with the exception of the I6 in m. 7 (a detail to be discussed momentarily).18 EXAMPLE 8. View largeDownload slide Chopin, Nocturne in G Minor, Op. 37, No. 1, (a) mm. 41–44; (b) mm. 1–8 EXAMPLE 8. View largeDownload slide Chopin, Nocturne in G Minor, Op. 37, No. 1, (a) mm. 41–44; (b) mm. 1–8 Though I am using this example to illustrate the Romantic tendency to exploit root-position harmonies, it is worth pausing briefly to examine details that reveal some other non-classical traits. In the first place, the form of the theme seems to be a standard period—an opening antecedent with weak cadential closure followed by a consequent bringing a stronger cadence. Entirely counter to classical norms, however, the antecedent phrase already modulates to the relative major, ending there with a nineteenth-century HC.19 (Note, by the way, that the local E♭ sonority, VI6, on the downbeat of the second measure not only adds a major-mode coloration to the otherwise pathos-laden minor quality of the opening two-measure idea but also subtlety alludes to the upcoming modulation to B♭ major.) The consequent phrase begins with the tonic of the new key, but quickly modulates back to G minor. To solidify this return, Chopin brings a more marked sense of harmonic progression in mm. 5 and 6 (four distinct harmonies there compared to a single prolonged harmony in mm. 1–2), and he makes sure that the downbeat of m. 6 holds a strong root-position tonic, rather than the more luminous VI6 of the antecedent (cf. m. 2). The final cadential idea is especially interesting, since its upper-voice melodic content ( 1^– 2^– 3^– 4^– 5^– 1^) brings the main elements of a classical bass line (the scalar rise from ① to ⑤ and subsequent leap back to ①, a line that regularly supports an entire thematic unit).20 To avoid parallel octaves (except at the very end, a license found frequently enough in most tonal styles), Chopin brings a non-classical cadential bass line (⑥—③—②—⑤—①), featuring a root-position VI as the initial tonic of the progression. After a subsequent I6—an embellishing chord, which is the only inverted harmony in the theme—the bass pushes downward to support a root-position half-diminished II7, a pre-dominant rarely used in a classical cadential progression.21 Thus, the theme as whole exemplifies many ways in which what might seem to be a fairly ordinary period actually departs considerably from the norms of the classical style, while bringing a series of typically Romantic elements of harmony and form. UNIFORM HARMONIC RHYTHM AND DENSITY In a manner that harkens back to baroque practice, the Romantic style often features a relatively uniform rhythm of harmonic change, especially compared to the classical style. In many cases, this uniformity of harmonic rhythm arises from the use of various dance genres (mazurka, waltz) or from character pieces that tend to feature a single emotional affect, thus emulating, again, a baroque aesthetic.22 This rate of harmonic change can often be quite fast as well, especially at the very opening of a thematic unit. (In the classical style, on the contrary, the harmonic rhythm of a theme tends initially to be relatively slow and then systematically increases toward the cadence.) To avoid overburdening the harmonic texture with rapidly changing prolongational progressions, Romantic composers will vary the progression type by introducing both sequential and cadential progressions in initiating formal positions. In addition, the possibility of embedding one progression type within another emerges as a distinct option for creating a degree of harmonic diversity within a foreground uniformity of harmonic rhythm. The opening of Schumann’s Davidsbündlertänze No. 8 (Ex. 9) illustrates well the kind of uniformity and density of harmonic activity found in a Romantic character piece. The opening measure already brings a cadential progression in C minor, which upon repetition in the following measure proves to be a mere accompaniment to a rather fragmented melodic idea. Taken together and in the context of the entire theme, these opening two measures represent a basic idea. The accompanimental pattern in m. 3 is then varied to become a cadential progression in E♭, and the complete melodic fragment is repeated a third higher. We thus see that Schumann embeds cadential progressions within a broader ascending-third sequence.23 This use of both cadential and sequential progressions for the very opening unit of a theme—here, a kind of presentation phrase—is especially characteristic of the Romantic style.24 At the beginning of the continuation phrase, m. 5, the progression resolves deceptively, thus shifting the tonal focus back to C minor. Observe that up to this point, most of the harmonies of mm. 1–4 have appeared in root position. With the deceptive resolution of the dominant at the beginning of m. 5, the bass line now acquires an ascending stepwise melodic profile, which lasts most of the way through the modulation to the minor dominant and the confirmation of the new key by a genuine cadence (a fully classical one), which closes this sentence theme type. Note that the embedded cadential progressions at the opening of the theme only project cadential content, not any sense of cadential function—nothing is ended by these cadence-like ideas.25 EXAMPLE 9. View largeDownload slide Schumann, Davidsbündlertänze, Op. 6, No. 8, mm. 1–7 EXAMPLE 9. View largeDownload slide Schumann, Davidsbündlertänze, Op. 6, No. 8, mm. 1–7 Let us return to Example 6 and consider the compound consequent phrase in light of the uniform harmonic rhythms typical of a Romantic theme. The phrase begins at m. 25 with varied material from the opening of the compound antecedent. Measure 29 sees the start of the ECP but now with a new one-measure melodic motive in the right-hand part, which is repeated in the following measures. The insistent upper pedal on the high A makes it difficult for cadential melodic closure to occur, and so the ECP is abandoned when at m. 33 the bass leaps down to the leading tone to support a non-cadential V56.26 Following the resolution to I, a simple V7–I progression creates the final PAC. Note that it would have been perfectly possible for the ECP that began at m. 29 to have continued all the way to the end of the theme by bringing a root-position dominant at m. 33 and sustaining that harmony until the tonic at m. 36, as shown in the bracketed harmonic analysis. Such an option would have been entirely conceivable within the classical style. Chopin, however, avoids the harmonic deceleration that would so arise, probably for several reasons. First, he has established throughout most of the theme a fairly uniform rhythm of each measure bringing either a new harmony, or a change of bass, or both. By abandoning the cadential progression, he is able to maintain this consistent harmonic rhythm to the end of the theme. Second, by bringing the progression V56–I in mm. 33–34, he can provide the same basic harmonic support to those measures as for the similar material in mm. 29–30. As a result, m. 33 sounds like the beginning of a sequential repetition, which is effectively given up when the need finally to create thematic closure brings the compressed, and highly incomplete, cadential progression. That the cadential idea emerges out of what is initially heard as a sequential situation is a Romantic characteristic to which I will return in a later section. FORMAL CIRCULARITY In the classical style, themes normally begin with a relatively “characteristic” idea, which, in accord with the powerfully teleological aesthetic of the style, leads to the goal of thematic closure. Part of the broader mechanism for motivating that closure resides in the breaking down of this opening material—a process Schoenberg graphically termed liquidation—such that all that remains by the end of the theme is the relatively “conventional” cadential idea. In contrast, a newly arising Romantic orientation to musical form began to favor a more circular mode of organization, one that manifests itself at various structural levels.27 At the level of the theme, this tendency toward formal circularity can be realized when the melodic–motivic material of the opening basic idea, rather than becoming fully liquidated, returns at the end to provide the melodic content of the cadential idea itself. Whereas this technique has some precedents in the classical style, it is more frequently encountered in Romantic practice, where, in fact, it occurs in a manner that differs markedly from earlier manifestations of the technique. Example 10, the opening of the scherzo of Schumann’s Second Symphony, is a case in point. The theme starts off-tonic in C major with a two-measure basic idea that is immediately repeated to make a presentation phrase; the continuation brings one-measure fragments as expected. Rather than further liquidating the material to yield a conventionalized cadential idea, however, Schumann brings back the melodic content of the basic idea for an imperfect authentic cadence (IAC) in E♭ (♭III) (m. 8). A repeat of the continuation phrase then modulates further to the minor dominant, closing there with another variant of the opening material to create a PAC in m. 12.28 EXAMPLE 10. View largeDownload slide Schumann, Symphony No. 2, Op. 61, third movement, mm. 1–12 EXAMPLE 10. View largeDownload slide Schumann, Symphony No. 2, Op. 61, third movement, mm. 1–12 As mentioned before, some classical precedents for this re-use of the opening basic idea can also be found, such as shown in Example 11, from Haydn’s Symphony No. 100, a passage discussed extensively by Leonard B. Meyer.29 Here, the basic idea itself has a distinctly “cadential” melodic component, one that classical composers used as a closing gesture in several other works.30 And in an obvious realization of this closing potential, Haydn brings back the idea, lightly varied, to forge the PAC at m. 36.31 Although the compositional situation here may remind us of Schumann’s theme, a significant difference lies in the nature of the melodic material used for both the basic and cadential ideas. With Haydn, the melody is already cadential in implication, and so it is hardly surprising that he allows the idea eventually to fulfill its implied formal function. With Schumann, on the contrary, the melodic gesture at the opening of the theme has nothing stereotypically closing about it; in fact, it is a highly “original” idea, which, even when re-harmonized so as to function cadentially, retains its highly individualized quality. We thus see operative here another aspect of Romantic style, namely an expression of the “aesthetics of originality.” Whereas classical composers are content to allow their initiating melodic-motivic material to become liquidated in the course of a theme, Romantic composers are more attached to their “novel” melodic ideas, and are thus inclined to retain them for use at the very end of the theme in a manner that promotes a circular design.32 EXAMPLE 11. View largeDownload slide Haydn, Symphony No. 100 in G, second movement, mm. 29–36 EXAMPLE 11. View largeDownload slide Haydn, Symphony No. 100 in G, second movement, mm. 29–36 SEQUENCE VERSUS CADENCE The Romantic practice of emphasizing root-position harmonies can give rise to an important source of harmonic and cadential ambiguity. If we consider the harmonies shown in Example 12, we can observe that such a progression can be analyzed as either a circle-of-fifths sequence, which could project a medial formal function of some sort (such as a continuation), or as a cadential progression, which could bring formal closure.33 Of course, a specific musical context will normally clarify which type of progression is involved: thus the progression in mm. 7–8 of Example 8(b), discussed earlier, arises in a context that unambiguously suggests its formal function to be cadential (analogous to mm. 3–4 in the antecedent phrase). In the classical style especially, composers are usually very careful to distinguish between the progression types. A typical case arises in the first subordinate theme from Mozart’s G-Minor Symphony (Ex. 13), where the continuation phrase (mm. 48–51) begins with a series of root-position harmonies organized by descending fifths. In its formal context and in its intrinsic content, the harmonies are obviously sequential, which Mozart makes clear by following the sequence with a distinct cadential progression to close the theme. EXAMPLE 12. View largeDownload slide Sequential versus cadential EXAMPLE 12. View largeDownload slide Sequential versus cadential EXAMPLE 13. View largeDownload slide Mozart, Symphony in G Minor, K. 550, first movement, mm. 44–51 EXAMPLE 13. View largeDownload slide Mozart, Symphony in G Minor, K. 550, first movement, mm. 44–51 Among Romantic composers, this practice of separating descending-fifths sequential progressions from cadential ones is not always observed, thus giving rise to ambiguities of cadential articulation. Consider the subordinate theme from the first movement of Schubert’s “Unfinished” Symphony (Ex. 14[a]), a melody so familiar to us that it is easy to overlook how utterly non-classical it really is. We can first note that all of its harmonies stand in root position, which already suggests that we may encounter ambiguities in the function of the harmonic progressions. After two measures of introduction, the theme begins with a two-measure basic idea (supported by I going to V), followed by a contrasting idea (V to I). Matching this harmonic reversal is a motivic one, as shown by the analysis of motives “a” and “b.” (This kind of motivic palindrome is rarely, if ever, found in classical themes.) Though we might be tempted to recognize a cadence at the end of this four-measure phrase, it is probably better to understand it as concluding without cadential closure, primarily because the harmonic pattern I–V–V–I, supporting an opening four-measure phrase, more typically projects four measures of tonic prolongation rather than a two-measure prolongation followed by a two-measure cadential progression. Here especially, it is hard to hear the harmony in the third measure of the theme (m. 46) as a distinctly cadential dominant (one that differs from the prolongational dominant of the prior measure), all the more so given the commonality of melodic-motivic material in both measures. At m. 48, the phrase begins to be repeated, but motive b of the basic idea is now sequenced up a step (supported by V7 of II). The “contrasting idea” (mm. 50–51) then remains a step higher than it was in the first phrase. In order to return to tonic, Schubert brings a varied sequence of the idea back down a step in mm. 52–53. Though the rhythm has been modified, the sequential relationship between these two ideas could be seen clearly if Schubert had written an exact sequence, as shown in the second staff of Example 14(b). What this rewrite also helps to illuminate is how the rhythmic modification of the sequence makes it sound as though it is beginning with a sequence of “motive a” whose descending fourth leap is now extended into a fifth. Finally, Schubert lets the melodic line end decisively on 1^ (see again Ex. 14[a])—decisive in that the cello melody does not directly leap down to 5^, as in m. 47. In another sense, however, we do hear this melodic leap, since the downbeat of m. 53 also brings an elided repeat of the entire theme, with its opening “a” motive.34 But now the question becomes, has the theme actually ended with a cadence? Most listeners will probably answer “yes” readily enough, but we should nonetheless be sensitive to a formal ambiguity here, for what we would be construing as the “cadential” idea, mm. 52–53, arises as a sequential repetition of the contrasting idea of mm. 50–51. In the classical style, a cadential idea almost always distinguishes itself from any preceding sequential material, as we saw in the previous Mozart example. Because a sequence brings a repetition of something that is already medial in function, it should continue to project the sense of being-in-the-middle and not sound cadential. However, here it seems to do just that, namely to bring a semblance of closure to the theme. In other words, we can hear the harmonies in mm. 49–52 as not only sequential (shown in line 1 of the harmonic analysis of Ex. 14[a]), but also as cadential, with V7 of II representing a chromatic alteration of VI, a substitute for the initial cadential tonic (shown in line 2). Given the harmonic and cadential ambiguities seen thus far, it is fascinating to observe some of the implications that Schubert draws from them as the movement proceeds. In the first place, the repeat of the theme in the violins leads to a complete lack of cadential closure when the music breaks off with the grand pause at m. 62 (Ex. 14[c]). This drastic evaded cadence is unusual in that not only is there no harmonic event (a tonic) at the moment of expected cadential closure, but also a complete absence of any event (the grand pause) at all. In addition, this moment is not followed by a “one more time” repetition as often occurs with evaded cadences.35 Instead, an entirely new augmented version of motive “a” returns, set as a dramatic and ominous tutti fortissimo outburst.36 In the recapitulation, Schubert handles this passage somewhat differently (Ex. 14[d]). Here, he brings the music back from D major (the key in which the theme appears in the recapitulation) to the home key of B minor by more fully realizing the sequential possibilities of the contrasting idea as it appears in Example 14(b), staff 1, and, in so doing, significantly minimizing the contrasting idea’s cadential effect. By the time we hear the grand pause at m. 280, we are not at all certain that an evaded cadence has even taken place; rather, the music seems more to be interrupted “mid-sequence,” with little sense of any cadential function. Returning to the exposition (Ex. 14[c]), we see that the fortissimo outbursts at mm. 63–66 propose the beginning of an expanded cadential progression by means of a prolonged pre-dominant (IV–[ 46]–Gr+6– VII56 of V); however, the progression rhetorically runs out of steam with the syncopated chords in mm. 71–72 and is effectively abandoned when, at m. 73, Schubert brings back motive “b” in an extensive call-and-answer passage, one that more fully realizes another potential for sequence in the contrasting idea, namely that proposed in Example 14(b), staff 2. Following a decisive cadence at m. 93 to end the subordinate theme group (Ex. 14[e]), Schubert brings back the opening of the subordinate theme,37 but now in a manner that resolves the ambiguity of sequence and cadence. Like the original version, the third measure of the theme (m. 96) repeats, albeit in a non-sequential manner, the second half of the basic idea (motive “b”). A completely new idea then appears (mm. 97–99) that is more genuinely cadential in nature, with nothing sequential in the picture. The only disrupting aspect is the harmonic support, whereby the dominant reached at m. 95 is held as a pedal point until the tonic resolution at m. 99. If Schubert had supported mm. 96–99 with a true cadential progression, as shown in the brackets, then we could speak of a PAC at m. 99. The retention of the dominant pedal from m. 95 on, however, blurs a sense of real cadence, thus allowing this passage to function formally more as an expanded codetta of a closing section rather than as a genuine thematic unit (i.e., another subordinate theme). EXAMPLE 14(A) AND 14(B). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (a) mm. 42–54; (b) alignment of sequences EXAMPLE 14(A) AND 14(B). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (a) mm. 42–54; (b) alignment of sequences EXAMPLE 14(C) AND 14(D). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (c) mm. 60–76; (d) mm. 274–82 EXAMPLE 14(C) AND 14(D). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (c) mm. 60–76; (d) mm. 274–82 EXAMPLE 14(E). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (e) mm. 90–100 EXAMPLE 14(E). View largeDownload slide Schubert, Symphony No. 8 (“Unfinished”), first movement, (e) mm. 90–100 To conclude this discussion of the “Unfinished” Symphony’s subordinate theme, we can now observe that a significant source of ambiguity in the passage arises from Schubert having placed all of the harmonies in root position; as a result, the opening phrase can be heard as either prolongational or cadential, and the second phrase as either sequential or cadential. Various aspects of motivic manipulation contribute to the ambiguous effect as well. PROLONGATIONAL CLOSURE; LACK OF FUNCTIONAL ENDING Not all phrases in the classical style close with a cadence: the opening presentation phrase of the sentence theme type is a case in point.38 But almost all complete themes in that style conclude with one of the three genuine cadences—PAC, IAC, or HC. With the Romantic style, we see a loosening of this thematic requirement and begin to find themes that close not with cadential progressions, but with prolongational ones instead. Schumann’s “Valse noble” from Carnaval (Ex. 15[a]) illustrates well how a self-contained thematic unit ends with such prolongational closure, as I will term it.39 The movement, starting very much in medias res, brings two matching four-measure phrases in a manner that suggests a periodic theme-type. Within the broader form of the piece as a whole, these opening eight measures constitute the A section of a small ternary. In the classical style, this section would normally end cadentially (with a PAC in either the home key or a subordinate key).40 Here, however, both phrases end prolongationally, the first with V24 to I6, the second with V56 to I. To be sure, the norms of the period form are referenced by the first phrase ending with a weaker closure (on an inverted tonic) than the second phrase (on a root-position tonic), yet neither mode of closure is truly cadential. Moreover, the bass line supporting the complete theme is entirely non-classical, featuring a descending line from ♭⑥ down to ⑦, which resolves up to ① at the very end.41 EXAMPLE 15. View largeDownload slide Schumann, “Valse noble,” Carnaval, Op. 9. No. 4, (a) mm. 1–8; (b) mm. 33–40 EXAMPLE 15. View largeDownload slide Schumann, “Valse noble,” Carnaval, Op. 9. No. 4, (a) mm. 1–8; (b) mm. 33–40 Following a contrasting middle (B section), the A′ section first brings back the material of the complete A section without any alterations. Schumann however, obviously not satisfied in allowing the complete movement to close prolongationally, begins to sound the theme again (Ex. 15[b]) and rewrites the harmonic support for the final phrase to make a genuine cadential progression (whose constituent harmonies, not surprisingly, are all set in root position), thus concluding the movement as a whole with a real cadence.42 We see that, unlike classical practice, prolongational closure in the Romantic style may take place at the level of the simple theme (the eight-measure A section),43 but the final closure of broader formal units still requires an authentic cadence.44 A more radical mode of “ending” arises when an ongoing thematic unit fails to project any intrinsic sense of closure, yet we realize that the unit is “finished” when the onset of a subsequent unit initiates a new set of formal processes. Such situations are rare in eighteenth-century works and are hardly common in the Romantic style. Some interesting examples, however, do arise, thus revealing the breakdown of classical means of formal closure. A casual hearing of the opening theme of Chopin’s Mazurka in F♯ Minor (Ex. 16[a]) suggests that it comprises a sixteen-measure compound period, whose antecedent forms an eight-measure sentence. The return of the basic idea at m. 9 quite clearly signals the onset of a consequent. If we look more closely, though, we must ask, how does the antecedent actually end? From the perspective of classical norms, we would be expecting a closing HC in m. 8. Instead, the harmony preceding the return of the basic idea is a pre-dominant II34, which does not mark the end of any cadence type. Not only does this formal unit not end with a cadence, but we also cannot speak of any prolongational closure at this point, since this harmony appears as the last link in a broad descending-fifth sequential progression begun at m. 5.45 Little in the intrinsic musical content projects any sense of closure whatsoever for this “antecedent” phrase. We only know that it ends because the consequent so obviously begins. Of course, what helps motivate the formal oddities observed in this Mazurka is the way in which the opening basic idea is harmonized: namely with a prominent V7–I prolongational progression.46 Thus, in order to set up the formal return at m. 9, Chopin lets the antecedent conclude on the pre-dominant II34, which connects easily with the initiating dominant to tonic motion of the basic idea.47 EXAMPLE 16. View largeDownload slide Chopin, Mazurka in F♯ Minor, Op. 6, No. 1, (a) mm. 1–16; (b) recomposed version by Rothstein (1989, Ex. 2.23) EXAMPLE 16. View largeDownload slide Chopin, Mazurka in F♯ Minor, Op. 6, No. 1, (a) mm. 1–16; (b) recomposed version by Rothstein (1989, Ex. 2.23) At this point, we might consider an option proposed by William Rothstein, who wants to hear a half-cadential goal for the antecedent on the downbeat of m. 9.48 His recomposed version, shown in Example 16(b), eliminates the formal overlap of Chopin’s original. Presumably to make the dominant closing the antecedent more effective as a goal, Rothstein even rewrites this harmony as a triad. Though this interpretation has merit, we should observe that the final dominant still emerges out of an ongoing sequential process, as supported all the more by the melodic line, which continues the same motivic content of the sequence. In other words, no distinct cadential idea is specifically associated with this proposed HC. Rothstein, though, perhaps overlooks one detail that points to the pre-dominant of m. 8 as the real “ending” of the phrase, namely, the rhythmic alteration on the second beat of that measure from an expected dotted figure to even eighth notes. This subtle change to what is otherwise a sequential repetition helps differentiate m. 8 as a possible ending, even if it is one that cannot be construed as either cadential or prolongational. This rhythmic alteration not only serves the specific formal purpose just described but also helps prepare for the double eighth-note configuration that appears in the returning basic idea on the second beat of m. 9. Moreover, the use of two eighth notes in place of a dotted-eighth and sixteenth (as in the prior sequential measures) effects a slight sense of rhythmic deceleration that further helps project m. 8 as a potential ending moment. Indeed, a performance by Arthur Rubenstein highlights the formal arrival there by effecting a noticeable ritardando within the measure.49 ULTIMATE VS. PENULTIMATE DOMINANTS We tend to think that the fundamental difference between a half cadence and an authentic cadence involves its final harmony, being either a dominant or tonic respectively. But just as important—perhaps even more so—is the distinctively different roles that the dominant harmony itself plays in these cadential formations. For in the HC, the dominant is the ultimate harmony of the progression, whereas in an authentic cadence, the dominant serves as the penultimate harmony. Since both cadences include a dominant—always in root position—our sense of how that harmony is functioning, as ultimate or penultimate, very much determines which cadential type is being projected by the music. In the classical style, composers rarely allow this distinction to become blurred; in the Romantic style, and especially with Schubert, ambiguities about the role of the cadential dominant give rise to problematic cadential interpretations. Several passages from the first movement of Schubert’s Piano Sonata in A, D. 959, will illustrate such problems, even allowing us to identify a new type of cadential deviation, which I term the dissipated cadence.50 We begin with the main theme, built seemingly as a small ternary, and first ask how the A section ends (Ex. 17[a]). The final measure would appear to be m. 6, because the complete change in materials in m. 7 can mark the beginning of a new unit, namely, a standing on the dominant to comprise the whole of the B section,51 which stretches to the return of the opening materials at m. 16, thus initiating the A′ section. As already mentioned, the opening A section of a small ternary normally ends with a PAC; but here, the dominant seventh in m. 6 does not resolve as expected, thus giving the impression that an evaded cadence has occurred instead (see again the discussion of Ex. 5). Unlike normal classical practice, however, this cadential evasion is not followed by a one more time technique, that is, by some repetition of immediately prior material that would lead to a concluding authentic cadence. Instead, the A section seems to be left without any formal closure whatsoever. There is, of course, an alternative interpretation; namely, that the dominant of m. 6 is ultimate, not penultimate. We would then recognize a “nineteenth-century HC” concluding the A section (thus referencing the practice of Haydn, as mentioned in n. 40). Such a view also suggests that we might revise our interpretation of the immediately following mm. 7–15, hearing them as a post-cadential standing on the dominant, not as a contrasting middle (B section), and thus recognize the return of opening materials at m. 16 as the start of a potential consequent phrase, one matching an opening antecedent. Several problems, though, present themselves if we believe that m. 6 is an ultimate dominant. Unlike Chopin or Schumann, Schubert does not normally employ the dominant seventh in half-cadential situations.52 More importantly, the A′ section leads to a similar dominant at m. 21, and there, it clearly behaves as a penultimate harmony (albeit an inverted one) that resolves to tonic in the following measure, bringing prolongational closure to the small ternary form as a whole. (Just why Schubert does not place the dominant in root position to create a regular PAC at m. 22 will be addressed shortly below.) Before leaving the A section and its ambiguous closure, we should consider in detail some aspects of its formal organization and melodic-motivic content. In the first place, its six-measure length means that we are dealing with either a deviation of a standard eight-measure theme type or else some nonconventional type. The opening two measures could be considered a basic idea, one comprising two one-measure motives.53 What follows is contrasting (with strong continuational characteristics), and a cadential idea begins with the upbeat to m. 5, continuing on through m. 6, whose ambiguous status we have just examined. Thus, whereas we can recognize three ideas that express the general functions of beginning, middle, and end, a conventional formal type does not emerge. What also does not materialize is a highly profiled melodic component to these ideas. The uppermost sounding voice remains fixed on the tonic degree until the decorated resolution of the suspension in the final measure (6); the bass voice also brings a tonic pedal for four measures until it finally moves up to ③ for the start of the cadential progression in m. 5. The only substantial melodic motion occurs in the two inner voices, which, bringing a modicum of melodic motion, consist mostly of scalar patterns moving up and down from 3^ in an alto voice, doubled a third lower all the way by the tenor. In short, the texture features outer-voice pedals surrounding scalar motion in parallel thirds by the inner voices. The alto voice essentially functions as the true “upper voice” of the contrapuntal structure, with the actual soprano having a more conventional inner-voice function. If we follow the course of parallel thirds consistently, then we would want to consider the dissonant-seventh D ( 4^) as the active melody at the end of the section, which further helps to support the idea of an evaded cadence, inasmuch as this tone strives to return back to its starting point, 3^.54 Given the consistency of the doubled-third motion throughout the section, we might accord this texture a significant motivic function and observe its development throughout the rest of the movement. The 4^ that is left hanging at the end of the A section is transferred to the very top of the arpeggiated texture that takes over at the beginning of the B section (m. 7) and continues to sound as the principal melodic element in every measure until the liquidation in mm. 14–15. As for the bass voice, it now becomes more active, rising chromatically stepwise throughout the section from ⑤ to ②, such that at the fortissimo climax at m. 13, the outer voices sound the doubled third pair B–D, just the same two motivic pitches from the end of the A section in m. 6. The resolution of 4^ finally occurs at the beginning of the A′ section, when 3^ is picked up again, now transferred to the tenor voice and doubled a third lower by the bass. In addition, the top voice has a new countermelody that sounds above the parallel-third motion of the lower voices. Formally and harmonically, the A′ section is organized just like the A section, except that as discussed, the final dominant resolves to tonic, thus extending the unit by one measure. Just before that happens though, the F♯ in the upper line of the parallel third motion is doubled by the soprano in the second half of m. 20. That voice then continues the line with the whole-note D, while the tenor loses its pitch, and sounds the root of the dominant, E, just the note we might have expected to appear in the bass voice, which instead sustains the parallel third doubling. As a result, a wonderful reversal of texture occurs such that in m. 21 the upper and lower voices now sound the doubled third, while the inner voices take on their more appropriate function as fillers, including the decorated suspension. At this point we are ready to discuss the cadential situation arising at the end of the A′ section. Following the bass as it ascends from ① to ④ by the end of m. 20, we would normally be expecting the line to reach its goal on ⑤ to bring the cadential dominant. As we have seen, however, Schubert lets the bass fall down to ②, in parallel thirds with the soprano, and then to ① for the tonic resolution. Consequently, we cannot speak here of an authentic cadence, but recognize instead prolongational closure both for the section and for the theme as a whole. As already discussed, Schubert’s motivation for failing to bring a genuine cadence most likely lies in the motivic work that he relentlessly pursues throughout this section, namely, the use of parallel motion as the essential contrapuntal activity of the theme.55 Let us now turn to the subordinate-theme area of the exposition, where we continue to encounter cadential uncertainties similar to those discussed so far. Example 17(b) shows the first of two subordinate themes.56 It opens with a phrase that ends at m. 58 with an HC in the subordinate key of E major, just like an antecedent. Note the parallel thirds in the left-hand part, an obvious reference to motives from the main theme. The phrase begins to be repeated as a possible consequent, but this function “fails” when the music suddenly veers off to G major, partially confirming that key at m. 63 with the same half-cadence gesture; as a result, the failed consequent “becomes” another antecedent.57 We thus understand that more in the way of thematic activity must occur. The music that continues at m. 65 prolongs I6 of G, suggesting that an ECP may be in the making, one that would support a continuation⇒cadential phrase following upon a presentation, made up of the repeated antecedents.58 A continuation of the implied ECP is reconstructed in Example 17(c), showing how the theme could easily have cadenced in G major. At m. 70, however, the cadential progression is abandoned when the music shifts suddenly back to E major: the bass note B, which functioned as the third of the tonic of G, continues to sound (it is even emphasized by being placed an octave lower) to become the root of the dominant of E. Following a deceptive resolution to VI at m. 71, a more compressed cadential progression begins with the pre-dominant IV and continues in the next measure with the cadential dominant, one that is clearly perceived as penultimate, in that we expect an authentic cadence to close this subordinate theme, now in the normal subordinate key of E major. The cadence is evaded, however, and the music quickly backs up to the cadential six-four for another try, one that is evaded yet again at the end of m. 73. At this point, something quite unusual takes place, for the music now seems to get stuck on the cadential dominant as the piano’s right hand brings a new triplet figure, which gets drawn out and liquidated much like a standing on the dominant. Thus, the resolution to tonic at m. 78 does not sound like a cadential goal but rather is exclusively heard as a beginning harmony, one that initiates the next phrase, a return to material from the very opening of the theme. What seems to have taken place is that a penultimate dominant (one that promises an authentic cadence) is somehow converted into an ultimate dominant to support a standing on the dominant, a “post-cadential” function that entirely removes any potential for cadential closure. In association with the complete liquidation of the motivic content and a highly recessive dynamic, we can have the impression that the impending authentic cadence has been entirely dissipated; indeed, I will term this deviation a dissipated cadence. Something similar, but even more dramatic, takes place toward the end of the second subordinate theme, which begins at m. 82 with a large-scale model-sequence plan (not shown in the examples) somewhat in the manner of a developmental core, though essentially remaining in the subordinate key.59 At m. 95 (Ex. 17[d]), Schubert initiates an ECP, starting first in E minor but then shifting back to major two measures later. The conditions for a PAC literally appear at the downbeat of m. 101, but the sudden dropping out of the chordal texture and abrupt change back to the minor mode could also be a sign that the cadence is evaded. Both options seem available for analytical (and performative) interpretation, though for reasons of broader formal organization, I prefer the latter. In any case, the passage strives forward with the same music, though now the left-hand plays the running triplets and the right-hand the slower-paced chordal blocks. The texture changes again at m. 105, and the harmonic progressions are suggestive of cadential activity. A clear cadential progression emerges at m. 110, strongly implying PAC closure for the theme. This time, though, the cadence is evaded by a grand pause (recalling a similar situation in the “Unfinished” Symphony, Ex. 14[c], m. 62), and when the music picks up again with the texture reduced to a single line, we understand that the dominant harmony of m. 111, which we construe to be penultimate, continues to be prolonged in a manner that resembles the very end of a standing on the dominant, one that normally follows upon an ultimate dominant. The texture begins to fill out at m. 116 in preparation for a return of the opening phrase of the first subordinate theme, whose tonic at m. 117 sounds entirely initiating, not ending. In other words, we confront again a case of a dissipated cadence, whereby the subsequent standing on the dominant exhausts the potential for cadential closure to the extent that the resolution to tonic brings a new beginning, not a cadential goal, thus converting an initially perceived penultimate dominant into an ultimate dominant. EXAMPLE 17(A). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (a) mm. 1–26 EXAMPLE 17(A). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (a) mm. 1–26 EXAMPLE 17(B). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (b) mm. 55–82 EXAMPLE 17(B). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (b) mm. 55–82 EXAMPLE 17(C). View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (c) reconstruction of mm. 69–72 EXAMPLE 17(C). View largeDownload slide Schubert, Sonata in A, D. 959, first movement, (c) reconstruction of mm. 69–72 EXAMPLE 17(D). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, (d) first movement, mm. 95–124 EXAMPLE 17(D). View largeDownload slide View largeDownload slide Schubert, Sonata in A, D. 959, (d) first movement, mm. 95–124 Up to this point, we have assumed that the dominant achieved at m. 111 is penultimate, largely for reasons having to do with formal context: we normally expect that at this late stage of a subordinate theme, any cadential function would be aiming toward authentic cadential closure, especially when an evaded authentic cadence immediately preceded at m. 101. Some details of pitch organization in m. 111, however, prompt us to entertain the idea that this dominant could be perceived even initially as an ultimate one, projecting a strong sense of HC, after which the grand pause supports the perception that the music has “stopped” with that cadence. After all, no dissonant seventh appears there; indeed, the texture is ultimately reduced to octaves doubling the root of the harmony. Though an interpretation of HC is perhaps less likely in the broader formal context that precedes it, it does make sense in light of what follows, namely, a standing on that ultimate dominant. To help project such an interpretation, a performer would want to make the dominant of this measure sound like a genuine goal, which might entail a slight ritardando and diminuendo throughout the measure. Performances by van de Laar and Pollini achieve just this effect.60 In contrast, those by Kovacevich, Schnabel, and Perahia increase the drive and intensity of the music throughout the complete measure, so that the effect of an interruption—a real cadential evasion—is dramatically projected.61 In short, Schubert leaves it open for the performer to help decide such ambiguous cases. The return of the opening of the first subordinate theme at m. 117 continues to play on the question of whether a dominant is ultimate or penultimate, since this time, the opening phrase ends at m. 120 with a dominant seventh, not a dominant triad (cf. again, Ex. 17[b], m. 58). As a result, it is perhaps unclear whether or not an HC is created at this point, since, as earlier discussed, Schubert does not normally employ the nineteenth-century HC. We might rather interpret the dominant as penultimate and recognize an evaded cadence, one that will motivate a subsequent PAC via the one more time technique; in fact, a clear PAC does occur at m. 123, preceded by that same penultimate dominant. This is the cadence that finally, and unequivocally, closes the entire group of subordinate themes. We are now in a position to step back and consider the larger formal patterns that emerge within the subordinate theme area of this exposition. As just discussed, we can find only a single unequivocal PAC that could be considered to close a subordinate theme, namely the one we examined in m. 123.62 Two other potential PACs were poised to appear—one at m. 74 (Ex. 17[b]), to close a first subordinate theme, and one at m. 112 (Ex. 17[d]), to close a second subordinate theme. In each case, these cadences were effectively dissipated by the conversion of their penultimate dominants into ultimate ones in the course of their standings on the dominant. If Schubert had written genuine PACs, the return of the motives from the opening of the first subordinate theme at mm. 78 and 117 would have functioned as post-cadential codettas. In addition, the subordinate-theme group (consisting of two complete subordinate themes) would have conformed largely to the norms of classical form. But taking the dissipated cadences into consideration, whereby a penultimate dominant becomes an ultimate one, a more complex design begins take shape. Let us begin again with the first subordinate theme (Ex. 17[b]). As we discussed, the section from m. 65 to m. 77 is supported by a bass note B. At first, this bass functions as the third of a I6 harmony in G major, but then assumes its role as dominant of the subordinate key, E major. Broadly speaking, and appealing to the spirit of retrospective reinterpretation, we could consider this entire section functioning as a contrasting middle (B section) of a small ternary structure, a kind of global “standing on the dominant” (of E). The A section would thus embrace the opening antecedent (mm. 55–59) and the failed consequent (mm. 60–64).63 The final section of the form, however, poses difficulties. In the context of a small ternary reading, the return of the opening materials from the A section strongly signals the start of the A′ section, most likely taking the form of a genuine consequent phrase (to make up for the failed consequent of the A section). Following the return of the basic idea, however, the materials quickly liquidate in the manner of codettas, such that the overall ternary form never receives its required perfect authentic cadential closure. That the first subordinate theme remains unclosed raises the possibility of recognizing an even larger-scale ternary design by regarding this incomplete theme as a magnified A section (indicated by the larger, italicized symbol);64 we could then regard the second subordinate theme (the latter half shown in Ex. 17[d]) as bringing together the remaining B and A′ sections (beginning at mm. 82 and 117 respectively). After all, the second subordinate theme begins with extensive sequential activity in the manner of a developmental core, and thus from its start has a strong intrinsic medial functionality of the type that we might associate with a contrasting middle. This enormous B section would be deemed to end with the ultimate dominant that emerges via the dissipated cadence at m. 111; that harmony is then sustained by the subsequent standing on the dominant.65 The large-scale A′ section is signaled by the return of the A section’s opening phrase at m. 117, and the theme as a whole is closed by the PAC at m. 123. With this interpretation, the exposition would see only a single subordinate theme, one built as a huge small ternary, whose A section itself is structured as a lower-level small ternary (albeit one that does not achieve cadential closure). Given that the small ternary form is never used as the basis of a classical subordinate theme (much less a group of themes), the embedding of a small ternary within an even grander small ternary shows how far from classical norms Schubert is willing to go. Moreover, we see here a further expression of the formal circularity that marks a Romantic approach to form, in that we encounter two different returns of the opening materials of the theme (mm. 78 and 117), each signaling the start of an A′ section (at two different levels of the form). Let us come full circle and give one final consideration to the main theme of the exposition (Ex. 17[a]). In light of the two dissipated cadences that we have identified in the course of the subordinate theme area, we can now examine again the cadential situation at m. 6 and find there the potential source of these cadential deviations. For if we take the dominant in that measure to be penultimate, an option that we gave serious consideration to in our earlier discussion, we can recognize that an evaded cadence is followed by a standing on the dominant, one that eventually dissipates any possible sense of an upcoming authentic cadence and that prepares for the return of the opening section, just the kind of situation that we encounter twice in the subordinate theme area. When first discussing this passage, I was hesitant to label it a dissipated cadence for the main reason that I continued to consider it a possibility that m. 6 brings an ultimate dominant to create a nineteenth-century HC. In light of what occurs throughout the rest of the exposition, however, finding here the first of three dissipated cadences seems like an especially compelling interpretation.66 * * * * * With this last example, as well as with most of those presented in this study, we see how Romantic composers find ways of creating closure—or denying it, as the case may be—that go far beyond the techniques of cadence utilized by their classical predecessors. These novel cadential situations arise from some specific compositional traits of the Romantic style, especially in the domains of harmony, rhythm, texture, and form. Furthermore, we have seen that the various techniques of closure treated here are related to phrase-structural devices that are atypical of the classical style. There exist advantages, but also some drawbacks, in the methodology that I have adopted for this study. In general, I have surveyed the Romantic repertory to identify those places where composers of earlier generations would have employed the classical cadence to effect formal closure, but where instead, Romantic composers write something that deviates from the norms of eighteenth-century practice. To be sure, this methodology, which emphasizes in the first instance why a given situation of closure is not classical, can give the impression that there is something compositionally defective or abnormal, and the entire discussion can seem to be framed in largely negative terms.67 Unfortunately, this problem seems endemic to approaches that attempt to establish norms and then identify violations of these norms. Of course, if it can be shown that the norms are no longer really applicable (for example, using principles of tonal harmony to analyze twelve-tone music, to take an extreme case, though one that has occasionally been pursued), then such an approach is less compelling and potentially useless. As has already been stated, however, the classical cadence continues to be found widely in nineteenth-century music, especially at key moments of formal articulation, so employing this model as a measuring tool to gauge conformance seems entirely reasonable. Most importantly, it is not a question of using the classical norms to show what is wrong with Romantic music, but rather to highlight what is different about it. Given that so many theories of tonal music propose a “common practice period,” one that embraces multiple historical styles (baroque, classical, Romantic, post-Romantic), many musicians fail to attend to significant differences in harmonic, cadential, and formal practices among these styles. For even if a norm/deformation model tends toward generating negative statements, such a model at least helps to clarify differences in composition technique, leading ultimately to an enriched hearing experience of the music. I would like to think that when we focus on what may seem on the surface to be small differences in harmonic usage, for example, the inclusion of a dissonant seventh at a moment of HC, the inversion of the dominant in a case of prolongational closure, or whether a given dominant is ultimate or penultimate, then we can begin significantly to refine our hearing and to appreciate how the different styles of tonal music are expressed. Eventually, of course, we would like to formulate sound, positive reasons for why composers of a given style found it entirely acceptable to break the compositional norms developed in earlier practices. For myself, I feel that I have yet to reach that goal for issues of formal closure. As a first step, though, I am content enough to identify differences in cadential practice and am hopeful that other current and future research will provide a more comprehensive, integrated perspective on these vital issues of musical composition.68 Footnotes 1 Caplin (2004). 2 The selection of music undertaken for this study—a rather eclectic sampling of instrumental works in a variety of genres—yielded examples of specifically Romantic cadential usage primarily in compositions by Schubert, Schumann, and Chopin. Few cases arose for the classicizing Mendelssohn, though one rather exceptional situation is cited in nn. 41 and 44 below. The highly eccentric formal processes witnessed in Berlioz normally involve dimensions other than cadence; in general, it seems, French composers of the nineteenth century hold firm to classical cadential practices. 3 Note that in this example, and throughout the remaining ones, the boundaries of the cadential progression are indicated by a horizontal bracket positioned below the Roman numerals. 4 Caplin (2004, 70); Caplin (2013b, 125–27) argues this point more fully, citing empirical, theoretical, and experiential grounds for this fundamental axiom. 5 I classify sequential progressions on the basis of a consistent pattern of harmonic root motion (descending fifth, ascending step, etc.); moreover, embellishing harmonies can sometimes appear between more fundamental harmonies of the sequences, as seen in Exx. 3(c) and 3(d). 6 Now and then a tonic prolongation can appear with both dominant and tonic chords in root position. Of course, the same progression could also function as an incomplete cadential progression. The potential ambiguity in classifying such a progression can normally be resolved contextually, that is, by considering whether the progression occurs in a formal situation that is either initiating or closing. 7 Caplin (2004, 56–66). This view thus challenges the claim by Hepokoski and Darcy (2006) that both a sonata exposition and an overall sonata form are closed by specific cadences (their “essential expositional closure” and “essential structural closure,” respectively). 8 Caplin (2004, 89–96). 9 Caplin (1998, 101–7), Caplin (2013a, Chapter 5). A deceptive cadence, using a deceptive cadential progression such as that shown in Ex. 1(c), brings a final cadential event that is supported by a different harmony than the required root-position tonic. An abandoned cadence results when the penultimate dominant either fails to appear or becomes inverted prior to its resolution to the tonic. 10 Schmalfeldt (1992). 11 Note that I deliberately speak of cadential closure of a movement’s final theme, thus emphasizing that cadential closure is effectively limited to the hierarchical level of thematic organization and does not normally apply to higher levels of form (see n. 7). 12 A second source of nineteenth-century chromaticism involves the use of symmetrically organized harmonic structures, ones that are often modeled by Neo-Riemannian principles. Passages featuring these harmonies will not arise in this study, since they rarely involve mechanisms of thematic closure, where the expression of functional tonality is paramount. 13 Schmalfeldt (2011, 202–3). Considerably more work is needed to flesh out aspects of the nineteenth-century HC, investigating, in particular, which composers favor the technique and which do not (for example, cases arise now and then with Mendelssohn, whereas Schubert, as will be discussed later on, tends most often to use the classical HC ending on a dominant triad). In two recent studies, L. Poundie Burstein argues that ample precedent already exists in the classical style for regarding HCs as ending with a dominant seventh (and even being inverted) (Burstein 2014, 211–14; Burstein 2015, 97–105). Here is not the place to evaluate this claim, which is based on fundamentally different assumptions about cadence than those held by Schmalfeldt and myself. I would only note now that almost any feature of a particular musical style has precedents in an earlier style; the main question is, are such precedents central or peripheral to the prior one? I would argue that only a very limited number of classical themes bring a half-cadential articulation that genuinely includes a dissonant seventh; as such these cases should be understood as a cadential deviation, one that I term dominant arrival (see Caplin 1998, 79–81 and Caplin 2013a, 224–25). I would thus support Schmalfeldt’s assertion that the inclusion of a dissonant seventh first becomes an essential element of the HC in the early Romantic style. (I will address Burstein’s viewpoint and present a fuller examination of the harmonic content of the HC in a book-length project tentatively entitled Cadence: A Study of Closure in Tonal Music.) 14 Formal descriptions throughout this article follow the categories developed in Caplin (1998) as well as the more recent textbook version, Caplin (2013a). For the discussion of compound themes, see Caplin (1998, 63–70) and Caplin (2013a, Chapter 6). 15 Though not related to aspects of closure in this theme, the specific harmonies of this opening phrase are worth considering from the perspective of the increased chromaticism and dissonance associated with early Romantic style; with this bass line, a standard classical presentation phrase would bring the prolongational progression I– V34–I6. Here, the passing dominant is further embellished with the more dissonant VII56 (m. 19) and the I6 is replaced, most unusually, by a dissonant VI46 (m. 20), which is chromaticized as V56 of IV at the beginning of the continuation⇒cadential phrase. As a result of the irregular resolution of VII56 in m. 19, this diminished-seventh sonority actually assumes a “common-tone” voice leading, though its dominant function still makes sense in light of the VI46 being a tonic substitute harmony. 16 When they do occur in the earlier styles, such progressions are typically part of a broader sequence, e.g., a descending-fifth pattern, as discussed ahead in Ex. 13. 17 For other examples, see Chopin, Nocturne in G Minor, Op. 15, No. 3, mm. 89ff. (marked “religioso”), and Chopin, Prelude in C Minor, Op. 28, No. 20. 18 As I consider the subordinate harmonies in mm. 1–2 to occur in the context of a tonic pedal, they should not be identified as being in any particular “inversion”; however, I have retained the labels for inversions in order to acknowledge the alternative, first mentioned in the text, of hearing more literal changes of harmony within these measures. 19 This interpretation assumes that the seventh literally entering on the second beat of the measure genuinely belongs to the dominant harmony, which appears as an incomplete sonority on the downbeat. The seventh E♭ (along with the ninth G) held over from the previous chord suggests this harmonic reading. If we hear the seventh as passing, however, then the implied dominant triad supports a classical HC. 20 See Caplin (2008) for discussion of this bass-line model. Following the practice of Sanguinetti (2012) (inspired by Gjerdingen 2007), I will indicate scale degrees in the bass voice by a circled numeral and indicate scale degrees in the upper parts with a circumflex accent. 21 The cadential progression of mm. 7–8 could also be seen as a descending-fifth sequential progression, a potentially ambiguous situation that I treat in a later section on “sequence versus cadence.” 22 A significant difference between the baroque Affektenlehre and the early Romantic portrayal of “character,” however, lies in the conventionalized nature of the former compared to the more individualistic sense of personal expression in the latter. 23 Schmalfeldt (2011, Chapter 8) identifies this sequential pattern as particularly characteristic of Chopin, though here we see the sequence used by Schumann as well. The pattern is rare in the classical style, especially within a presentation phrase. 24 In the classical style, such a presentation would normally be supported by a single tonic prolongational progression. 25 See also Chopin, Prelude in C Minor, Op. 28, No. 20, mm. 1–2. 26 See Caplin (2013a, 143) on abandoned cadential progressions as opposed to abandoned cadences per se; on the latter, see Caplin (2013a, 132–33). 27 The idea of circular form in the nineteenth century is discussed by Morgan (2000) and Rodgers (2009). 28 We find in this example another instance of the I–III–V sequential pattern underlying the entire thematic unit, as discussed earlier in n. 23. 29 Meyer (1989, 26–30); see also Levy (1981, 28). Example 11 illustrates the A′ section of the main theme, an overall small ternary form (rounded binary version). 30 Meyer (1989, 26) cites Haydn’s String Quartet in B♭ Major, Op. 64, No. 3, iii, and Mozart’s Symphony No. 40 in G Minor, iii. 31 Haydn uses the same idea to close the earlier A section, with a dominant arrival at m. 7, and brings the idea back again to confirm the modulation to the dominant region in the middle of the B section, m. 24. 32 The use of a “characteristic” opening gesture as a subsequent cadential idea continues to be found in later nineteenth-century repertories; the first subordinate theme from the opening movement of Tchaikovsky’s Pathétique Symphony (cf. mm. 89–91 and 95–97) is especially exemplary of this practice. 33 The confusion of sequence with cadence arises essentially with the descending-fifth (“circle of fifths”) sequence, since this is the only sequence type that yields a potentially cadential V–I progression at its end. 34 The cello then “echoes” this motive in beats 2 and 3 of mm. 53ff., but now as an accompanimental figure, not as a genuinely melodic element. 35 Compare again Ex. 5, m. 16. Not all classical evaded cadences feature the one more time technique, though many do. Yet almost all such evasions at least bring some sounding event in place of the cadential arrival. Schubert’s evaded cadence is thus most bizarre from a classical perspective. We will see another such evaded cadence in a later discussion of the composer’s Piano Sonata in A, D. 959 (see Ex. 17[d], m. 112 below). 36 See Macdonald (1978) for a discussion of Schubert’s “volcanic temper.” 37 This return of opening subordinate-theme materials exemplifies again the Romantic impulse for formal circularity. In the classical style, by contrast, once subordinate-theme ideas are eliminated from the theme, they rarely return either in the context of the theme group itself or in the subsequent closing section. I will discuss another use of this technique below in connection with Ex. 17(b), m. 78 and Ex. 17(d), m. 117. 38 See Caplin (2004, 59). The compound basic idea (a basic idea plus a contrasting idea supported by a prolongational progression) is another four-measure phrase that does not close with a cadence; likewise, a continuation phrase may end with no cadential closure when, in looser formal contexts (such as a subordinate theme), it is followed by a unique phrase serving an exclusively cadential function. 39 Many theorists describe what I am calling prolongational closure as a weaker form of cadence proper; the terms “contrapuntal cadence” or “imperfect cadence” are regularly encountered in this context. As discussed on several occasions already, I restrict the concept of cadence to cases where both the dominant and final tonic of the cadential progression appear exclusively in root position; see again nn. 4 and 13. I fully acknowledge that a particular type of formal closure can be effected by concluding a prolongational progression, but I understand it as one that is categorically—and, more important, experientially—distinct from closure that is specifically cadential in nature. 40 Haydn now and then ends the A section of a small ternary (or rounded binary) with an HC; Mozart and Beethoven almost always conclude that section with a PAC. 41 See Caplin (2008) for the essentially ascending contour of a classical bass line. The propensity for bass lines to descend, especially at the beginning of a theme, is characteristic of the baroque style; see Lester (1999, 27–33, esp. Ex. 2.3 and Ex. 2.5) and Caplin (2014, 436). For another example of a thematic unit ending with prolongational closure, due primarily to the use of a falling bass line, see Mendelssohn, Song without Words, Op. 19, No. 1, mm. 13–15, where the penultimate dominant initially appears inverted before taking on its more conventional root position just prior to resolving to tonic. 42 The cadence at m. 40 is perfect authentic because the melody, lying below the “covering tone” F, closes on B♭ ( 1^). 43 Horton (2015, 103–7) develops the same basic point; see his insightful discussion of how the main theme in the solo exposition of Chopin’s Piano Concerto in F Minor, Op. 21, first movement, fails to close cadentially yet still projects a clear sense of thematic ending by means of prolongational progressions. 44 The Mendelssohn Song without Words cited in n. 41 is thus highly exceptional in that the entire piece concludes (mm. 42–44) with the same prolongational closure described for mm. 13–15. Later in the nineteenth century, we encounter more frequent use of prolongational closure for a complete movement (or larger section of one); my forthcoming study on cadence in various tonal styles includes a chapter on cadential practices of the mid to late nineteenth century. 45 More precisely, we could speak of the sequential progression as actually beginning in mm. 3–4 with the roots E and A and then continuing into the opening of the consequent with the roots C♯ and F♯. For the purposes of this discussion, I am focusing specifically on the sequential segment supporting the antecedent’s continuation phrase (mm. 5–8). 46 As mentioned in n. 6, a progression from root-position dominant to root-position tonic can be sometimes emerge as prolongational, not cadential. 47 See also Schmalfeldt (2011, 198). 48 Rothstein (1989, 46–48). 49 RCA Red Seal 09026 63050-2, compact disk. 50 Another work by Schubert, the first movement of his Sonata in A Minor, Op. 42, D. 845, features a number of ambiguous cases of ultimate and penultimate dominants; Schmalfeldt (2011, Chapter 5) discusses some of these cadential ambiguities, especially as they affect performance decisions. 51 See Caplin (1998, 75–77) and Caplin (2013a, 210–11) for the conventional use of a standing on the dominant in the contrasting middle of the small ternary form. 52 Schmalfeldt (2011) does not identify any nineteenth-century half cadences in connection with Schubert, only raising the topic first with her chapter on Chopin; see again, n. 13. 53 See Caplin (1998, 37) and Caplin (2013a, 39) for cases in which an opening basic idea consists of a repeated one-measure motive. 54 I am regarding all of the inner-voice Es in the chord of m. 6 as doublings of the bass voice. 55 There are perhaps additional reasons for Schubert’s not having placed an E in the bass: if the bass line of m. 20 were to continue to rise to E3, the texture would be extremely thin, and thus not appropriately balanced with the texture of m. 6. Moreover, it is not possible to add the lower E2 while still retaining D5 in the upper voice and an octave-doubled decorated resolution in the inner parts, since two hands cannot play such a configuration of notes. Finally, the upper D ( 4^) in m. 21 would awkwardly clash against the lower E (⑤), a voice-leading situation that, if not strictly prohibited, is normally shunned in such an exposed manner. 56 The discussion of Exx. 17 (b) and (c) largely reproduces that given in Caplin (1987, 246–49). 57 On the notion of failed consequent, see Caplin (1998, 89) and Caplin (2013a, 246–47). Somewhat problematic in characterizing this second phrase as an antecedent is its modulating structure; normal antecedents in a classical period do not modulate, as discussed earlier in connection with the modulating antecedent of Ex. 8(b). 58 In this broader sentential context, the half cadences ending the antecedent phrases are thus deemed to have “limited scope” (see Caplin, 2004, 86–89). 59 See de Médicis (2015, 205) for a discussion of the infusion of core-like materials within subordinate themes of Schubert’s expositions. The use of immediate model–sequence activity at the start of the theme also has clear classical precedent, since a second subordinate theme occasionally begins directly with continuation function, thus bypassing an initiating unit; see Caplin (1998, 111–13) and Caplin (2013a, 391). 60 Brilliant Classics 99678/2, compact disk (van de Laar); Deutsche Grammophon 4713502, vol. 6, compact disk (Pollini). 61 EMI Records, compact disk (Kovacevich); Pearl CDS 9271, compact disk (Schnabel); Sony Classical S2K 87706, compact disk (Perahia). 62 I am discounting the possible PAC at m. 101, since no coherent view of the form ensues from interpreting this as a moment of complete tonal and formal closure; instead, I recognize an evaded cadence there for the reasons cited earlier. 63 Once again (as mentioned in connection with the main theme), we would be confronting the exceptional case of an A section closing with an HC rather than a PAC, but I have already discussed a classical precedent for this situation in Haydn (see n. 40). 64 The expression “large-scale small ternary” might seem odd; why not speak of a “large ternary”? The problem, of course, is that the latter is itself a distinct classical formal type, one that diverges significantly from the small ternary; see Caplin (1998, 211–16) and Caplin (2013a, 566–69, 574–86) for my definition of the large ternary form and how it differs from the small ternary. 65 Or, if we believe that m. 111 brings a regular HC, shown as option 2 in the analysis, then we would not speak of a dissipated cadence. 66 A look at the coda of the movement further supports this reading. Here we find that Schubert twice brings back, in an extended form, the opening thematic unit (A or A′) and closes both with a classical perfect authentic cadence (mm. 340 and 349). As well, the dominant seventh of each cadence includes the same decorated resolution of the 4–3 suspension that we find in mm. 6 and 21, thus suggesting that these earlier moments, too, can be considered penultimate dominants. 67 See Vande Moortele (2013, 5–8) for a highly insightful discussion of the dilemma of approaching nineteenth-century form from a “positive” or “negative” perspective. 68 In addition to the studies by de Médicis, Horten, Morgan, Rodgers, Schmalfeldt, and Vande Moortele cited above, important recent contributions by Aziz (2015), Black (2015), Davis (2014), Hyland (2009), Koslovsky (2014), Martin (2010), Martin and Vande Moortele (2014), Smith (2014, 2016), and Suurpää (2011) are using in creative, “positive” ways some of the methodologies of the recently revived Formenlehre for eighteenth-century music to help enrich our understanding of early nineteenth-century form and cadence. Works cited Aziz Andrew I. 2015 . “The Evolution of Chopin’s Sonata Forms: Excavating the Second Theme Group.” Music Theory Online 21 ( 4 ). Accessed August 10, 2016. http://www.mtosmt.org/issues/mto.15.21.4/mto.15.21.4.aziz.html. Black Brian. 2015 . “Schubert’s ‘Deflected-Cadence’ Transitions and the Classical Style.” In Formal Functions in Perspective . Ed. Moortele Steven Vande , Pedneault-Deslauriers Julie , Martin Nathan John . 165 – 97 . Rochester, NY : University of Rochester Press . Burstein L. Poundie. 2014 . “The Half Cadence and Other Such Slippery Events.” Music Theory Spectrum 36 ( 2 ): 203 – 27 . Google Scholar CrossRef Search ADS ———. 2015 . “The Half Cadence and Related Analytic Fictions.” In What Is a Cadence? Theoretical and Analytical Perspectives on Cadences in the Classical Repertoire . Ed. Markus Neuwirth , Bergé Pieter . 85 – 116 . Leuven : Leuven University Press . Caplin William E. 1987 . “The ‘Expanded Cadential Progression’: A Category for the Analysis of Classical Form.” Journal of Musicological Research 7 ( 2–3 ): 215 – 57 . Google Scholar CrossRef Search ADS ———. 1998 . Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven . New York : Oxford University Press . ———. 2004 . “The Classical Cadence: Conceptions and Misconceptions.” Journal of the American Musicological Society 57 ( 1 ): 51 – 117 . CrossRef Search ADS ———. 2008 . “Schoenberg’s ‘Second Melody,’ or, ‘Meyer-ed’ in the Bass.” In Communication in Eighteenth-Century Music . Ed. Danuta Mirka , Agawu Kofi . 160 – 88 . Cambridge : Cambridge University Press . ———. 2013a . Analyzing Classical Form: An Approach for the Classroom . New York : Oxford University Press . ———. 2013b . “Teaching Classical Form: Strict Categories versus Flexible Analyses.” Tijdschrift voor Muziektheorie 18 ( 3 ): 119 – 35 . ———. 2014 . “Topics and Formal Functions: The Case of the Lament.” In The Oxford Handbook of Musical Topics . Ed. Danuta Mirka . 415 – 52 . New York : Oxford University Press . Davis Andrew. 2014 . “Chopin and the Romantic Sonata: The First Movement of Op. 58.” Music Theory Spectrum 36 ( 2 ): 270 – 94 . Google Scholar CrossRef Search ADS de Médicis François. 2015 . “Heavenly Length’ in Schubert’s Instrumental Music.” In Formal Functions in Perspective . Ed. Moortele Steven Vande , Pedneault-Deslauriers Julie , Martin Nathan John . 198 – 222 . Rochester, NY : University of Rochester Press . Gjerdingen Robert O. 2007 . Music in the Galant Style . New York : Oxford University Press . Hepokoski James , Darcy Warren . 2006 . Elements of Sonata Theory: Norms, Types, and Deformations in the Late-Eighteenth-Century Sonata . New York : Oxford University Press . Google Scholar CrossRef Search ADS Horton Julian. 2015 . “Formal Type and Formal Function in the Postclassical Piano Concerto.” In Formal Functions in Perspective . Ed. Moortele Steven Vande , Pedneault-Deslauriers Julie , Martin Nathan John . 77 – 122 . Rochester, NY : University of Rochester Press . Hyland Anne M. 2009 . “Rhetorical Closure in the First Movement of Schubert’s Quartet in C Major, D. 46: A Dialogue with Deformation.” Music Analysis 28 ( 1 ): 111 – 42 . Google Scholar CrossRef Search ADS Koslovsky John. 2014 . “Timeless Reflections: Form, Cadence and Tonal Structure in the Scherzo and Finale of Schubert’s String Quintet.” Music Analysis 33 ( 2 ): 168 – 93 . Google Scholar CrossRef Search ADS Lester Joel. 1999 . Bach’s Works for Solo Violin: Style, Structure, Performance . New York : Oxford University Press . Google Scholar CrossRef Search ADS Levy Janet M. 1981 . “Gesture, Form, and Syntax in Haydn’s Music.” In Haydn Studies: Proceedings of the International Haydn Conference: Washington, D. C., 1975 . Ed. Larsen Jens Peter , Serwer Howard , Webster James . 355 – 62 . New York : W. W. Norton . Macdonald Hugh. 1978 . “Schubert’s Volcanic Temper.” Musical Times 119 : 949 – 52 . Google Scholar CrossRef Search ADS Martin Nathan John. 2010 . “Schumann’s Fragment.” Indiana Theory Review 28 : 85 – 109 . Martin Nathan John , Vande Moortele Steven . 2014 . “Formal Functions and Retrospective Reinterpretation in the First Movement of Schubert’s String Quintet.” Music Analysis 33 ( 2 ): 130 – 55 . Google Scholar CrossRef Search ADS Meyer Leonard B. 1989 . Style and Music: Theory, History, and Ideology. Philadelphia : University of Pennsylvania Press . Morgan Robert P. 2000 . “Circular Form in the ‘Tristan’ Prelude.” Journal of the American Musicological Society 53 ( 1 ): 69 – 103 . Google Scholar CrossRef Search ADS Rodgers Stephen. 2009 . Form, Program, and Metaphor in the Music of Berlioz . Cambridge : Cambridge University Press . Google Scholar CrossRef Search ADS Rothstein William. 1989 . Phrase Rhythm in Tonal Music . New York : Schirmer . Sanguinetti Giorgio. 2012 . The Art of Partimento . New York : Oxford University Press . Schmalfeldt Janet. 1992 . “Cadential Processes: The Evaded Cadence and the ‘One More Time’ Technique.” Journal of Musicological Research 12 ( 1–2 ): 1 – 52 . Google Scholar CrossRef Search ADS ———. 2011 . In the Process of Becoming: Analytic and Philosophical Perspectives on Form in Early Nineteenth-Century Music . New York : Oxford University Press . Smith Peter H. 2014 . “Schumann’s Continuous Expositions and the Classical Tradition.” Journal of Music Theory 58 ( 1 ): 25 – 56 . Google Scholar CrossRef Search ADS ———. 2016 . “Cadential Content and Cadential Function in the First-Movement Expositions of Schumann’s Violin Sonatas.” Music Theory and Analysis 3 ( 1 ): 27 – 52 . CrossRef Search ADS Suurpää Lauri. 2011 . “Non-Congruent Temporal Functions in Chopin’s Nocturne Op. 62, No. 2.” Tijdschrift voor Muziektheorie 16 ( 1 ): 64 – 71 . Vande Moortele Steven. 2013 . “In Search of Romantic Form.” Music Analysis 32 ( 3 ): 404 – 31 . © The Author(s) 2018. Published by Oxford University Press on behalf of The Society for Music Theory. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Music Theory SpectrumOxford University Press

Published: Apr 7, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off