On Performing Chopin’s Barcarolle

[1] Frédéric Chopin’s masterful Barcarolle, op. 60, a late work from 1846, poses particular challenges to the performer.⁽¹⁾ Despite its many musical virtues, it presents a sprawling, idiosyncratic formal plan, a certain uniformity of texture and melody for roughly the first two-thirds of the piece, and a succession of short but dense, powerfully expressive subsections toward the end, all culminating in similarly strong cadential arrivals. In my own practice and performance of the piece, I felt certain shortcomings and the need for ideas to address them. Since I had not done so before, for the purposes of this study I set about analyzing the music to see what ideas might emerge. From the viewpoint of performance strategy, such an exercise could potentially address nagging trouble spots—the “problem-solving” mode identified by Jonathan Dunsby (1989, 8) and John Rink (2002, 39)—but it could also suggest fresh approaches to passages that had not seemed particularly problematic beforehand. Moreover, from my particular stance as a theorist of mid-nineteenth century chromatic harmony, a secondary interest in this exercise was to see if my conception of “common-tone tonality” might be relevant in some way.⁽²⁾ In fact, all three types of outcomes materialized.

[2] Any inquiry of this type must necessarily engage with the issues raised by the relevant literature of the last few decades on the relationship of analysis to performance. The classic pitfalls are well-documented: the prioritization of the analyst’s way of thinking over the performer’s, with a certain disdain for the latter; the related dependence on analytic notions of musical structure foreign to typical performance lingo; and the tendency toward highly prescriptive conclusions dictating one preferred way of performing a passage, all epitomized in publications by Wallace Berry (1989) and Eugene Narmour (1988). Tempering this attitude was the prior claim by Janet Schmalfeldt that “there is no single, one-and-only performance decision that can be dictated by an analytic observation” (1985, 28).

[3] One can trace two broad and related lines of response to the analysis-informs-performance attitude of superiority. The first of these preserves one of its basic assertions, the existence of what John Rink has called “the vast terminological gulf between analysts and performers” (1990, 323), but draws different conclusions: rather than being inferior in musical understanding to academic analysts, performers have their own equally valid and highly developed means of conceiving and analyzing music, since after all they have been performing it convincingly for generations, and they, not the analysts, are the experts in this domain. From this viewpoint, performance has perhaps the greater potential to inform analysis than analysis does performance. Beyond this, the focus of inquiry has shifted toward discovering how performers think, act, and interact musically, in their own terms, while the abstract evidence of the score is supplemented by the concrete evidence of live and recorded performances, resulting in a fruitful and wide-ranging interdisciplinary outpouring of research.

[4] The second line of response takes a different and perhaps less radical tack, focusing less on the terminological gulf and the great divide it perpetuates, and instead viewing analysis and performance as having the potential to inform each other in limited but genuine ways, as complementary expressions of thinking and acting musically that can be embodied in collaboration or in the same individual. Joel Lester (1995, 198), Jonathan Dunsby (1989, 14), and Nicholas Cook have argued along these lines, with Cook noting a “new emphasis on the mutuality of the analyst/performer relationship” (1999, 245). With this shift, the more traditional piece-based performance-and-analysis study—less concerned with prescriptive recommendations for demonstrating structure-oriented analytical findings through performance, less bound by what Cook has called the theorist’s “paradigm of representation” (1999, 256), and mindful of the many other factors influencing the act of performance—remains viable. In a more recent special issue of this journal, Cook (2012) cites an article by Daphne Leong and David Korevaar (2005) as an example of the new model, in which the authors, as pianists and analysts, succeed in employing an integrated and balanced approach, although he has since tempered his praise on account of their analytic approach (2014, 42). The spirit of the earlier articles cited above is reflected in their approach: carefully reasoned analytic observations drawing on familiar theoretical concepts; detailed performance decisions involving articulation, emphasis, and timing; and the study of recorded performances for evidence of similar influence. What distinguishes the article from its predecessors is the presence of these authors’ complete musical selves. Most authors of the performance and analysis studies of the ‘80s and ‘90s were also accomplished performers. But striving to achieve the proper scholarly perspective of the time involved closing the door on the very aspect of themselves that presumably motivated the inquiry. Schmalfeldt’s (1985) solution to this dilemma, the artificial separation of her musical self into Analyst and Performer for the bold purpose of incorporating the performer’s own voice into music-theoretical discourse, stands as a sign of the times. Her experiment drew criticism both for a perceived bias toward the Analyst, and for the implication that the separation of the two personas was beneficial to scholarship.⁽³⁾ A generation later, Leong and Korevaar achieved the symbolic opposite, melding the insights of two individuals into a single scholarly narrative in which analysis and performance mutually inform each other.

[5] I would like to position the present study modestly, to provide a further example of how analysis, in response to performance, can contribute positively to performance strategy. Limiting performance recommendations to my own playing should help to avoid the pitfalls of overly prescriptive language.⁽⁴⁾ Expressing these recommendations on a conceptual rather than a technical level, where possible, should better allow for the multiple realizations implicit in Schmalfeldt’s dictum, quoted above. Treating performance, analysis, and my experience of the music as interdependent should guard against the tendency toward raising artificial distinctions. And beyond these aims, I hope that this study might offer some useful contribution to a general understanding of Chopin’s Barcarolle.

[6] This article addresses aspects of the piece from several analytic perspectives: form, harmonic structure, rhythm and meter, a landmark opening motive, and the fate of a referential pitch-class and pitch-class pair. Other perspectives interact with the analytic ones: Chopin’s dynamic and tempo indications and their variants, and an extramusical association stimulating the musical imagination. Rink’s (1988) comprehensive analysis of the Barcarolle will serve as a useful departure for the discussion, although my present concerns are different from his.

[7] Example 1 shows Rink’s representation of the Barcarolle’s formal layout (1988, 197). A short introduction is followed by a tripartite section in tonic F♯ major, built around what Rink calls theme A. Its tonal structure exemplifies what Edward Cone (1968, 83) describes as Chopin’s idiosyncratic adaptation of the sonata principle in this piece.⁽⁵⁾ Thirty bars in length, a self-contained form, it comprises more than a quarter of the Barcarolle.⁽⁶⁾ Next, a transition carried through by a single-line texture leads to another section of roughly equal length, whose tonal center is A major, a chromatic mediant key of great significance in the work. This section contains two different themes, Rink’s B(1) and B(2), grouped together more by their common tonic than by any fundamental similarity; in fact, theme B(2) has notable similarities to theme A, as I will show below. This second principal section is followed by an interlude that moves from an insistent pedal on the local dominant, E, through a staggered descending chromatic motion to settle on the global dominant, C♯ major, as a pedal point, not quite a full-fledged key. At nearly three-quarters of the way through the piece, this brief, serene interlude contains the only music that actually centers on C♯ major, other than a brief tonicization in the first section. It culminates in a dramatic return to material from the first two principal sections, now all in tonic F♯ major, and strikingly attenuated: both the first and second main sections of the piece are represented by their last parts only (A' and B(2)'), each no more than ten measures in length. Despite their brevity, these reprises are much more grandiose than their original appearances and carry considerable structural and rhetorical weight. Rink identifies them as the locus of the principal structural descent and cadence, and he notes the presence of what Edward T. Cone termed “apotheosis”—“a special kind of recapitulation that reveals unexpected harmonic richness and textural excitement in a theme previously presented with deliberately restricted harmonization and relatively drab accompaniment” (Cone 1968, 84). This foreshortened return is followed by an extended, eight-measure tonic pedal built on a fragment of theme B(1), which Rink identifies as the coda. It leads to some more explicitly closing material, a coda to the coda as it were, that ends the piece. Rink’s graph of the Barcarolle’s background structure is shown in Example 2. While adopting the classification of themes in his formal plan, I will propose rethinking his identification of the structural cadence and the formal conclusions that follow therefrom.

[8] In his essay, Rink cites Gerald Abraham’s observation that Chopin’s introductions often seem to have no structural connections to the rest of the piece (Rink 1988, 218). Rink refutes this claim for the Barcarolle, based on linear motivic content and other factors. I will support and augment Rink’s argument using evidence of a different kind. First, though, let me introduce the personal, extramusical association that fueled my imagination. The barcarolle as a genre is, of course, a Venetian boat song. One of my most lasting impressions of a long-ago trip to Venice is of a sultry summer night that happened to be a festival day. Over the canals, fireworks repeatedly lit up the sky. Not until years later did this image occur to me as I played the opening measures of the Barcarolle: shot from below, the music lands high in the sky in a burst of pre-dominant energy, its sparkling parts slowly floating down through the air at slightly unequal speeds (Example 3).

[9] Since then the image has remained with me, hard to shake, despite the skepticism I’ve encountered from a minority of those with whom I’ve shared it. For the music of a composer widely regarded to have transcended the native associations of genres, a programmatic response to “barcarolle” may seem naïve. Nonetheless, the fireworks metaphor has helped me to articulate a question with implications for performance: does the Barcarolle begin in the water, or in the air? Does the introduction set the background character for the opening section, or does it set up a contrast against which the opening section defines its character? To me, the water imagery of the Barcarolle is conveyed in no small part by its lilting compound triple meter. Thus a “hydrocentric” introduction would initiate this lilt through accents and temporal stretching of the initial notes of each triplet group, defining a flexible, dotted-quarter pulse to be transmitted directly to the left-hand accompanimental figure supporting theme A. On the other hand, air is undifferentiated; it has no lilt, no pulse. Thus, an “aerocentric” introduction would proceed in even, unaccented values governed by an overall temporal shape, with the left hand accompaniment subsequently initiating the lilt as if from elsewhere. I wondered if there might be hydro-intro and aero-intro camps discernable in the recorded legacy of the piece. About a dozen different recordings, mostly historical, served as my initial basis for comparison—a small but representative sample that included classic renderings of the piece. Two of the recordings I knew best, the legendary Dinu Lipatti recording from 1948, and an Artur Rubinstein recording from about 1962, were exemplary: Lipatti is clearly hydro, and Rubinstein is definitely aero, whatever their own reasons for these approaches may have been. Audio Examples 1a and 1b demonstrate the different approaches: the reader may determine which is which.

[10] All of the recordings I listened to were unique and pianistically appealing, making it clear that my analytical metaphor, used as the basis for defining character and pulse in the passage, could hardly be prescriptive. Moreover, implementing the decision in performance would be contingent on many other factors involved in making a performance decision, including the properties of individual instruments and acoustic environments. Nonetheless, all things being equal, my own tendency is toward aero rather than hydro, consistent with my image of Venetian fireworks, and a preference for contrast over continuation in setting the stage for the amount of material to follow. But I can also attempt to support my orientation analytically. Looking at the introduction’s harmonic and contrapuntal structure, we find a descending series of sonorities outlining the ii chord over dominant pedal that Rink identifies. The composer’s single slur over the entire passage suggests uniformity: one broad, unarticulated gesture, in contrast to the half-measure and quarter-measure slurs giving impetus to the lilt of the left-hand accompaniment that follows. Looking more closely at the passage, we find a series of different intervals and contrapuntal states, contracting and expanding, lines moving in parallel and contrary motion. These do eventually coalesce into a half-measure pattern about halfway through, but in their linearity and constantly changing pitch-class content, with dissonance and consonance constantly superseding each other, they still fall well short of the consistent pitch-class repetition, consonance on the beat, and triadic figuration that follows in the left hand accompaniment. Applying the same lilt to the introduction as to the accompaniment masks these differences and limits the power of an association I will describe later on (see paragraph 22).

[11] The introduction terminates with a conspicuous, unresolved melodic A♯4, whose destiny we will trace as we proceed. Note that the left-hand accompaniment opens at measure 4 with a series of rising fifths, echoing the prominent rising fifth at the beginning of the introduction. At the entrance of theme A proper in measure 6, Chopin writes a counterintuitive dynamic marking: rather than rising and falling with the broad outlines of the meandering line, he marks an initial diminuendo, in contrast to the widespread doctrine of phrase arching, in which registral rise and fall are amplified by analogous dynamic shaping (see Example 4).⁽⁷⁾ This opening diminuendo can feel peculiar, and most of the recordings I listened to did not respect Chopin’s direction. However, thinking analytically in the context of the entire theme, one can imagine tailoring the composer’s dynamic markings to the hypermetric level, with dynamics rising to the hypermetric downbeat of measure 8, the theme’s fifth measure. Privileging the stress on the downbeat of measure 8 sets up a longer-range hypermetric effect: the downbeat of measure 10 receives a similar stress, but the downbeat of measure 12, at the beginning of an intensifying gesture, is relatively unstressed, enhancing the impact of the culminating stress on the downbeat of measure 14, about which more below. Thus heeding the initial diminuendo, as in my performance at the beginning of Audio Example 2, prods this performer to think more broadly, while exercising restraint in the first gesture.⁽⁸⁾

[12] The preponderance of diminuendos over crescendos throughout the piece also serves, in principle, to encourage the performer not to overplay. Chopin’s own playing, from all reports, tended strongly toward the softer end of the dynamic spectrum, eschewing any harshness. He also traced a constant dynamic ebb and flow, rarely sustaining a single dynamic level.⁽⁹⁾ Thus, his markings can be seen to reflect his own approach to performing the music. Nevertheless, a contemporary report on Chopin’s performance of the Barcarolle in his last public concert in 1848 marvels at the way he played the entire final section of the piece, measures 84–111, marked from forte to fortissimo in the score. In radical opposition to his own indications, Chopin maintained the music in a subtly nuanced pianissimo throughout that, according to the account, only the composer himself could bring off successfully in performance (see Eigeldinger 1998, 102–3). Chopin was well known for playing his own music differently at different times, on the spur of the moment (Eigeldinger 1998, 186–87n112). We might take this information as evidence justifying the performer’s carte blanche to disregard Chopin’s performance indications, much less to associate them with the results of analysis. Some of the historical recordings, particularly those from the earlier part of the twentieth century, display this attitude throughout.⁽¹⁰⁾ However, thoughtful consideration of the sheer abundance and care given to these indications, and their intimate relation to the music as it unfolds, leads to the conclusion that they represent more than mere suggestions to be ignored. The performer has something to learn from them, even when choosing not to follow them all.

[13] Looking further at the individual gestures comprising theme A, we find that the first three all terminate on the same A♯4 that ended the introduction (measure 7, beat 3; measure 9, beat 4; and measure 11, beat 2; refer again to Example 3). What may initially seem to be potentially tedious repetition appears in a different light, though, if one approaches this mid-nineteenth century music with a mid-nineteenth-century mindset informed by Moritz Hauptmann’s harmonic theory, which implies that a sense of harmonic motion may be transmitted through the change in identity, or chordal meaning, of common tones (Hauptmann 1853, 29–30 and 64–66). I would add intervallic relation to the bass as a factor contributing to tones’ identity in melodies where the associated chords are not in root position. The harmonic recontextualization of prominent melodic common tones as an embodiment of musical motion is a particularly common ploy of composers such as Schubert and Chopin. At the beginning of the Barcarolle, as illustrated in Figure 1, A♯ appears strikingly isolated as a dissonant sixth above the bass, lingering in the ear (see Example 3, last chord of measure 3). During the first two phrases of theme A, A♯ becomes the third of the tonic triad (first in measure 4, in a lower register, and in measure 6 as A♯4), and in the third phrase it moves again, becoming the fifth of D♯ minor (measure 10). It seems appropriate to attend to these changes in quality, coming as they do at successive phrase endings, and to use them as cues to differentiate the moments. A♯ continues to surface at key moments as the piece progresses, becoming a pervasive pitch-class motive.⁽¹¹⁾ While there are numerous localized instances of common-tone redefinition in the Barcarolle, only A♯ really becomes motivic throughout the piece.⁽¹²⁾

[14] At the end of theme A, the prominent A♯ gives way to a C♯, whose contextual meaning is continuously redefined in relation to the bass and its accompanying voice exchange with the alto. This leads to the piece’s first cadence, one of its most beautiful moments, at measures 14–15 (Example 4). The melody at measure 14 directly engages the introduction’s opening: the original leap C♯–G♯ is answered here by its inversion G♯–C♯, in the same rhythm, completing the octave, hypermetrically stressed, and followed as in the introduction by a kind of terraced floating. In this way the introduction and the cadence on C♯ frame all of the music to this point. Recognizing this relationship, and conceiving the cadential leap as an answer to the initial leap, has been helpful to me in delineating the overall shape of the music. The relation can be suggested by slightly underplaying the lower chord tones leading to G♯, as well as the E♯ below the C♯, rather than by voicing the sonorities fully. Listening to my group of historical recordings, I found only a couple of pianists who do this, notably Ashkenazy, but for me it works. Besides reviving the motive, this voicing of the climatic cadential , heard in Audio Example 2, also helps to project the melodic arch and to prepare the texture of parallel sixths that immediately follows.

[15] This is one passage that I had never been able to get to feel right despite repeated attempts at figuring it out. As part of this study, I decided to consult earlier editions of the piece to see if the variants in dynamic and articulation markings that are typical for Chopin editions might provide some clues for improving my interpretation (see note 8). In the Henle edition I had been using, there is a forte marking on the first beat of measure 14, and the right-hand chord on the second beat carries an accent above, indicating a stress greater than that on the first beat.⁽¹³⁾ The Paderewski edition from which I had originally learned the piece contains a more explicit crescendo to a forte marking on the second beat. However, the markings in the first editions and the nineteenth-century editions I consulted differ significantly from those in the modern editions. The English and German first editions are consistent in indicating forte at the beginning of measure 14 rather than on the second beat, while the French edition places the forte marking in the middle of the first beat, with an immediate diminuendo before the onset of the second beat.⁽¹⁴⁾ Neither the first editions nor the nineteenth-century editions cited here contain the second-beat accent; instead, they have a diminuendo hairpin beginning anywhere from right after the downbeat to the beginning of the second beat, indicating some degree of lighter stress on the second beat than on the first (see Example 4, which follows the English first edition). This suggests a performance effect that would be lost by following the modern markings. In the build-up to the cadential in measures 12–13, the passing chords with voice exchanges generate a harmonic rhythm of four dotted-quarter beats per measure. The modern-edition accent on the second beat of measure 14, which I had first learned, prolongs this regular stress pattern into the cadential measure, continuing the sense of intensification right up to the dominant seventh on the third beat, heard as an isolated instance of dotted-half pulse, as heard in my performance in Audio Example 3. But the original markings de-stress the second beat of measure 14, unmasking (for me) the actual dotted-half harmonic rhythm of the entire cadential measure. Emphasized in performance, this earlier change of pulse with unstressed second and fourth beats can unify the measure, creating a sense of sustained suspension through the cadential and melodic rise, followed by a corresponding sense of gradual release through the dominant seventh and terraced melodic descent. The end of Audio Example 2 demonstrates the results of this insight. The first time I tried this new accentuation pattern at the piano, I instantly achieved the satisfaction that had eluded me for years.⁽¹⁵⁾ While the basics of this change could have been achieved simply by following the instructions in the older editions, it was analysis and my new ability to consciously experience and project the harmonic rhythm that drove the transformation.⁽¹⁶⁾

[16] The music next moves sequentially through B major to land on an extended A♯ pedal (measures 20–22), which is heard initially as V/vi in F♯, with the melodic A♯ now locally, following Hauptmann (1853, 21–25) in root-meaning, that is, in its identity as the root of the chord. This A♯ major triad resolves, however, directly back to F♯ major at measure 24, in a chromatic third relation that identifies the A♯ triad (in my terminology) as the upper sharp mediant, located a major third above the tonic with harmonic content on the sharp side of the diatonic set (see Kopp 2002, 8–12), and with the melodic A♯ once again in third-meaning. I’ll have more to say about this presently. This progression is shown at the end of Figure 1 and the beginning of Example 5.

[17] Theme A (Rink’s A') returns at measure 24 in a somewhat heightened setting of double trills, melodic diminution, and textural amplification, this time concluding squarely in the tonic key, creating further new meanings for A♯, as shown in Figure 2. The repeated tonic chord following the cadence has its inner-voice A♯ emphasized by a grace note, like a chime. On its fourth appearance, in measure 35, A♯4 is transformed prominently into A♮, and the music takes a meaningful turn. A solitary line emerges, first relatively unmeasured, then meandering until it gradually acquires the characteristic barcarolle lilt, ultimately settling into the “foreign” key of A major (measure 40).⁽¹⁷⁾ The effect, for me, is as though the music had wafted through the air until landing on water in a new location (Example 6).

[18] The long section that follows establishes A major as the Barcarolle’s principal secondary key. From a Schenkerian perspective, A♮, like A♯, is capable of embodying the III Stufe as third-divider in a tonic arpeggiation within F♯ major (Schenker [1935] 1979, 29–30; fig. 14/1b). But within nineteenth-century common-tone tonality, A♮ and A♯ relate differently to the tonic. As a direct progression, the earlier move at measure 24, the descending major third from A♯ major to tonic F♯ major, is an instance of perhaps the most familiar of all chromatic mediant moves; the root of the first chord is preserved as a common tone in the second (becoming its third), and the fifth of the first chord functions as a leading tone to the new root. On the other hand, the ascending minor-third move from F♯ major to A major (mediated here by F♯ minor) defines a different relation. The root of the first chord vanishes, while the root of the second chord lies outside the previous diatonic set; the common tone is the root of neither chord (see Kopp 2002, 15–17).⁽¹⁸⁾ Especially given the prominence of the common tone A♯ in the previous section, the effect here is one of greater harmonic distance, despite the closer relation of A major to F♯ major on the circle of fifths.⁽¹⁹⁾ The independence of A major from F♯ major is accentuated here by the tenuous, single-line connection between sections. Nonetheless, as I have argued elsewhere, in common-tone tonality A major represents a stable chromatic mediant key within F♯ major.⁽²⁰⁾ This analysis has a potential bearing on performance, since theme B(1) is itself harmonically unstable, moving twice quasi-sequentially from A major through G♯ major to F♯ major, and from there to C♯ major, before returning to A major; see Figure 3.

[19] The accompaniment, with figuration that articulates the dotted-quarter rather than dotted-half note, embodies these choppier waters. If one thinks along conventional lines and regards A major as dissonant within F♯ major, then this section expresses the tendency of the foreign key to be drawn back to the tonic, or more likely toward the dominant given its place within the overall form. Thus, one would imagine the sequence as a progression from instability toward stability, hijacked back to the original state at the end. However, if one thinks of A major as a stable key from the start, then the passage takes on a different cast, as if the home key of F♯ major had repeatedly, and unsuccessfully, tried to pull A major back into its orbit. How might this translate into a performance idea? At measures 41–42, A major gives way to G♯ major through the agency of a French augmented-sixth chord. If A major were unstable, then it should yield easily, as one hears on most recordings, and in Audio Example 4a. But if it were stable, then it would need to sound as if pried away, achievable even while respecting Chopin’s decrescendo marks, as in Audio Example 4b. In the subsequent move to F♯ major, through the agency of a less dissonant dominant seventh chord, the arrival could likewise either be emphasized to show F♯’s dominance, or de-emphasized to reflect its contextual subservience to A major. In my own playing, I had always given the French sixth at the end of measure 41 the pried-away treatment, so analysis merely confirmed my existing intuition about the passage. However, I had never thought much about the following phrase except as a sequential repetition of the previous one, and had played it somewhat similarly, with slightly less intensity. Now, noting the less dissonant transitional chord to F♯ major, and with an understanding of the key’s local contextual significance, I make more of the difference, letting the dominant seventh at the end of measure 45 fall away rather than being pried, and arriving less decisively into F♯ major. The dynamic markings in the score can be understood to reflect this gentler approach. The result is a clearer long-range strategy for the entire theme than before, and potentially more interest for listeners, especially since the whole theme repeats.

[20] Following theme B(1)’s repetition, theme B(2) appears at measure 62, the music now entirely in A major (Example 7). This new theme, despite its quicker tempo, is perceptibly similar to theme A. The first phrases of both themes are framed by members of the tonic triad, having structural stepwise descents from local $\hat{5}$ at the beginning to local $\hat{3}$ at the end, along with prominent upward leaps of a minor sixth from local $\hat{3}$ to local $\hat{1}$ in the middle (here filled in by arpeggiation). Whereas the single stepwise descent of theme A’s first phrase happens at the end, theme (B)’s first phrase has two stepwise descents: one right at the beginning, and a second, less direct one at the end (Figure 4).⁽²¹⁾ There are further similarities: the accompaniment of theme B(2) returns to the half-measure units of theme A, and most of its melodic phrases also end on $\hat{3}$, which is now C♯. In the case of theme B(2), all of these arrivals are on the same chord by similar means, with similar performance indications for phrasing and dynamics. This repetitiveness becomes an issue to be addressed in the preparation of a performance, and a pianist who wishes to play them differently can make use of analytic evidence. The first arrival, in measure 63, is straightforward. The second, in measure 65, is less an arrival than a jumping-off point for harmonic motion that briskly transforms the contextual identity of the C♯ toward increasing dissonance: at first a stable triadic third, it becomes an element of a diminished seventh chord, and finally, at the beginning of the following measure, a suspension compelled to resolve. The third arrival, in measure 69, acquires a new metric context following some irregular phrasing that places it in the third measure, not the second, of a gesture containing not one but two moments of relative climax. At the theme’s end in measure 70, A major simply dissipates into the air, an exceptional upward gesture for this piece in that it never comes back down. This dissolution again underscores the independence of A major from tonic F♯; originally entering from nowhere on a thread, it now simply floats away rather than returning home. A dominant pedal on E ensues, in a new texture and character, further demonstrating the tenacity of A major within the formal scheme (Example 8).

[21] This dominant pedal, though, eventually leads away from the key, which is ultimately relinquished though a chromatic descent that drives another distinctive process of melodic common-tone redefinition. At measure 75, as depicted in Figure 5, F♮ is introduced as a downward-tending ninth (although it is the bass that ultimately descends). F♮ is then recontextualized as a neutral minor third and finally emerging as an upward-tending leading tone, E♯, implying tonic F♯. With only moderate concentration, one can hear and feel this change of tendency and respond to it in performance—in my case, with a lightening of the tone as the downward pull yields to upward. The harmonic motion toward F♯ is deflected only at the last moment to tonicize the more formally appropriate dominant, C♯. One might think that a harmonic twist as meaningful as this would be well worth translating into a performance strategy. But as Chopin has written the passage, other factors, particularly meter, make the arrival on C♯ sound more natural than it may appear in the analysis. It is difficult to show that kink at the end (i.e., the move to C♯ rather than F♯), and its effect might well not be clear to the listener. Leong and Korevaar consider the question of “performable analysis,” what can and cannot be achieved, noting that in much of the literature on performance and analysis, “elements of design (rhythm, motive, form) and surface-level structure translate directly into performance, while deeper-level structure informs performance decisions indirectly, if at all” (Leong and Korevaar 2005, [7]). By “translate directly” I do not understand the authors to mean that the analytic insight itself is projected in translation through performance, evoking Cook’s paradigm of representation, but rather that the insight leads directly to a performance decision that responds to the musical implications of the analysis. In any case, it may be that the irregular surface aspect of this progression is overpowered by the significant and overdue background arrival on the global dominant. The theater scholars Judith Milhous and Robert Hume, writing about the applicability of analytic insights to theatrical performance, have coined the expression “producible interpretation” to denote those insights that are communicable, in contrast to others whose abstraction precludes comprehensible transmission, or whose meaning cannot profitably be made apparent (Milhous and Hume 1985, 3–23). While the common-tone redefinition just described in measure 75 is, in my experience, producible, this quality of deflection to the C♯ in measure 76 may constitute an unproducible element. However, Chopin communicates its effect indirectly: just a few moments later, an enhanced chromatic approach to C♯ major serves to frame the harmonic arrival as something truly earned.

[22] The exquisite dolce sfogato section follows at measure 78. Sfogato indicates a light and delicate manner of playing. The section begins with a familiar element: the C♯–G♯ leap from the Barcarolle’s opening, now in a stable triadic setting. While the expressive context is different, the similarities are strong: the two attacks are metrically identical, and both leaps, with intervening notated rests, arrive on G♯5, followed by a wandering melodic line in short, equal durational values. Thus once again the introductory motive denotes a significant formal boundary, cueing an association with the F♯-major beginning of the piece after such a long digression to a chromatic key. The relationship between these two sections could conceivably be heightened in advance by choosing the aerocentric approach to the introduction, since it and the dolce sfogato equally suspend meter and project a sense of near-weightless meandering. As for the significance of the motivic relation, whether or not to emphasize it in order to further enhance the similarity in character for the listener is really not the issue. In the realm of producible interpretations, this one merits a maybe; moreover, listeners hear differently, certainly regarding a long-range association such as this one. For me the significance exists at the stages of making performance choices and crafting their manifestations in sound. Leading into this moment is the highly chromatic passage just described above, with quickly changing sonorities that encourage a focus on the linear-harmonic process and require separate pedaling of each chord to maintain clarity. Once C♯ major is attained, though, my attention immediately shifts from the linear-harmonic process to the resonance of the instrument; keeping the pedal down, I aim to create as ravishing and sonorous a background as I can manage upon which to float the melody. Prior to realizing the motivic association, I thought of the C♯4 on the downbeat purely as the ending of the previous process, and paid little attention to it. The G♯5 on the next beat was a beginning, pulled out of conceptual thin air. But the motivic bond has led me to reconceive the C♯ as an elided, essential component of the initial, evolving resonance of measure 78, to be consciously voiced. Imagining its connection to the G♯ has provided conceptual solid ground within the resonance, much richer than thin air, on which to place that note. What is more, a further group of associations accrues to the G♯ (Figure 4). Like themes A and B(2), the sfogato melody begins on local $\hat{5}$, tracing exactly the same chromatic lower-neighbor figure as theme A. This is followed by the ascending-sixth leap that is characteristic of both themes. But whereas those are minor sixth leaps from $\hat{3}$ up to $\hat{1}$, this one is a major sixth leap from $\hat{5}$ to $\hat{3}$, reaching higher than before, after which the melody retraces the stepwise descent of the previous themes and breaks through, reaching down an octave. This transformation, coupled with the overall upward succession of transpositional levels of these themes from F♯ major to A major to C♯ major (with concomitant rises in register), infuses the melody of measures 78–80 with what is for me an unmistakable aura of culmination and transcendence. The analytic realization that this moment ties to so much of what has gone before has influenced the way I play it, with additional gravity and continuity that strike me as an enhancement. The intended effect of this change on listeners would not be for them to share my insights, but rather to enjoy and appreciate the moment even more than they might have otherwise. And it is my love of this moment as a performer that drew me to think about the passage analytically in the first place.

[23] Toward the end of the dolce sfogato, the music proceeds through a chromatic bass ascent that prepares the return of themes A and B(2), heard in apotheosis at measures 84 and 93 respectively. Both are now further enhanced texturally, dynamically, and, in the case of theme B(2), harmonically. This expressive intensification is reinforced temporally, as Chopin restates the original material in a highly condensed format (refer back to Example 1). Theme A appears for the first time without its diminuendo at the beginning, as if liberated from constraint. Theme B(2) is now stated in tonic F♯ major rather than in A major (Example 9). A significant result of this transposition is that all of the repeated phrase endings on C♯ in the original theme B(2) now terminate on A♯, just as in theme A. This provides an abundance of new contexts for the common A♯, now increasingly more dissonant as shown in Figure 6: dominant seventh of B♯ (measure 96); suspended fourth (measure 97); and then a tritone above E in a pronounced dominant (measure 100). This process of intensification leads the music to its grand cadence at measure 103, at which point A♯ is at last brought down to F♯, completing the structural descent of Rink’s background sketch. For performance, my sense is that it is almost enough to pay attention to the phrase endings and shape them within the diminuendi rather than just letting them go by, although I do like to isolate the tritone moment at measure 100 through rubato. Note that Chopin writes a diminuendo in the first half of the measure, followed by ritenuto in the second half, precluding a too-ostentatious arrival into the cadence.⁽²²⁾

[24] What follows is an extended, highly chromatic tonic pedal point (Example 10). The prevailing performance tradition for this eight-measure passage is for the pianist to treat it like a coda, like closing material pulling well back dynamically and expressively from the previous section and gradually winding down toward the augmented-sixth filigree in measure 110. Rubinstein’s 1962 recording provides a classic example of this treatment. Yet Chopin writes sempre f for the passage, without supplying any of the plentiful, nuanced dynamic markings that characterize virtually every other passage in the piece.⁽²³⁾ Rink notes that this section, which he calls the first coda, contains the Barcarolle’s only true four-bar phrases (Rink 1988, 214). The hypermetric regularity, reinforced by the relative suspension of dynamic ebb and flow and the insistent measure-by-measure rhythmic repetition within four-measure cadential phrases, allows for a process of intensification unmatched anywhere else in the piece. As mentioned previously, the return up to measure 103 is highly attenuated, lasting only nineteen measures versus seventy-one in the original, which leaves plenty of space for further activity. The trills and crescendo of measure 106 clearly intensify toward the second four-bar phrase, during which harmony becomes denser, more chromatic, and dissonant than ever before in the piece. For all of these reasons I would suggest that this is not closing material; formal process is still active here. Whereas the previous cadence at measure 103 provided a dramatic harmonic arrival, measures 110–11 provide a dramatic contrapuntal arrival, set into motion at the downbeat of measure 110 by the extremely dissonant augmented sixth heard against its own octave of resolution below, and resolving into that octave in its own register at the following downbeat. Interestingly, the only difference between Jim Samson’s (1985) analysis of the Barcarolle’s form and Rink’s (1988) occurs at just this point. Samson includes the F♯ pedal within the main section of the return, associating it with the “missing” theme B(1) from before, and identifying measure 111 as the coda’s beginning (Samson 1985, 97).⁽²⁴⁾ Thinking of this F♯ pedal as part of the principal structure of the piece, an extension of apotheosis rather than coda material, helps me immeasurably in understanding and interpreting Chopin’s markings. In fact, in lieu of a single structural cadence, we could imagine a distributed or staged cadential process for the Barcarolle, first harmonic in measure 103, then contrapuntal in measure 111.⁽²⁵⁾

[25] The typical performance of the pedal passage, embodying decreased energy and drive, initiates the filigree of measure 110 at a moderate dynamic level, then either relaxes the tempo and dynamics during the ascent or else races up in a sparkling shimmer. Rounding the registral apex, it characteristically dissipates on the way down to a very soft landing. My performance in Audio Example 5a approximates this approach. Accordingly, at least one edition from the late nineteenth century actually moves Chopin’s calando indication back from the beginning of measure 111, after the cadential arrival, to the start of measure 110, while others introduce hairpins to the gesture and, in one case, a ritenuto marking.⁽²⁶⁾ In contrast, all first editions of the Barcarolle, despite their differences, continue the sempre f through measure 110 with nothing expressly indicated for the filigree, followed by a stress on the arrival at the downbeat of measure 111: perhaps a last, dissonant rocket zooming into the air and landing with a noticeable impact.⁽²⁷⁾ The modern Henle and Ekier editions reproduce the most forceful of the first-edition alternatives (Example 11), and my rendition of this approach is in Audio Example 5b. Of course, this is not to say that the filigree should ever be played unrelentingly forte and fast without nuance and grace; it must be supple and true to style. But having always shaped this passage in the conventional way as a retreat, I am rethinking my approach to maintain intensity and imbue the filigree with the energy of an arrival. Thus analysis, coupled with a bit of research, has led to a performance decision of consequence, against the common grain, that would not otherwise have been considered. Among my select group of recordings, a relatively recent one by Alicia de Larrocha from 1995 does come close to this effect, and there are doubtless others as well.

[26] After this event, the Barcarolle’s coda (or second coda, depending on one’s analytic stance) wraps things up by invoking three of the piece’s signature motives in turn; see Example 12. First, at measure 113, a closing tenor theme played by the inner fingers of the left hand traces an octave emanating upward from A♯3, twice arriving on A♯4. This final reminiscence in register of the pitch-class motive first introduced in measure 3 is significant, and while a conscious association may not be producible (in the sense of Milhous and Hume 1985), the pitch may be engraved enough in aural memory by this time that a bit of emphasis would sound fitting. Next, one last rise-and-fall figure realized in filigree leads to the conclusion, where a unison ascending tonic fourth (C♯–F♯) at measure 116 gives the definitive answer to the introduction’s initial dominant ascending fifth, framing the entire piece. So much for Abraham’s assertion (see paragraph 8), whether serious or tongue-in-cheek, that the introduction has no relation to the rest of the piece.

[27] Over the course of this essay I have attempted to demonstrate ways in which analysis can act in concert with other modes of thought and action in preparing the performance of a work of music. There are multiple types of analytic observation, involving rhythm and meter, harmony and counterpoint, common-tone and voice-leading relations, pitch-class and intervallic motives, and formal process. Most of the discussions of specific passages draw on a combination of these types. In the discussions, analysis interacts with other considerations in different ways and in different sequences, with varying paths through performance issues, analytic insights, study of scores and sources, extramusical images, and performance solutions, creating a mix of ideas and approaches specific to this particular piece. The decision to present the content in the chronological order of events was deliberate: organizing by analytic topic or mode of inquiry would inhibit the ability to communicate both this richness of interaction and a sense of the whole.

[28] To maintain a balance of perspectives, I have confined the analysis for the most part to those observations that address or are linked to issues of performance. I have left some analytic tools in the toolbox as they are not relevant to my focus, and some of my analyses are not as thorough as they would be if undertaken for other purposes. I have warded off unduly prescriptive conclusions by limiting performance directives to my own use, although the reader may find interest in some of those directives, and were I teaching the piece in the studio I would encourage students to test them out. Likewise, I have avoided negative evaluations of the cited recordings. I do not advocate for my reading of the piece as the preferred one, but simply document what I consider to be improvements in my own playing. Where my own experience bears on a discussion, I have described it. My hope is that the reader will come away from this article with a respect for the positive effects that analysis can bring to performance, and performance to analysis, as well as an enhanced appreciation of one of Chopin’s most admired works.

David Kopp
Boston University School of Music
855 Commonwealth Avenue
Boston, MA 02215-1303
dako@bu.edu

Works Cited

Berry, Wallace. 1989. Musical Form and Musical Performance. New York: W. W. Norton.

Blum, David. 1980. Casals and the Art of Interpretation. Berkeley: University of California Press.

Caplin, William. 1998. Classical Form: A Theory of Formal Functions for the Instrumental Music of Haydn, Mozart, and Beethoven. Oxford: Oxford University Press.

Cone, Edward T. 1968. Musical Form and Musical Performance. New York: W. W. Norton.

Cook, Nicholas. 1999. “Analyzing Performance, Performing Analysis.” In Rethinking Music, ed. N. Cook and M. Everist, 239–61. Oxford: Oxford University Press.

—————. 2012. “Introduction: Refocusing Theory.” Music Theory Online 18, no. 1.

—————. 2014. Beyond the Score: Music as Performance. Oxford: Oxford University Press.

Dunsby, Jonathan. 1989. Guest Editorial: “Performance and Analysis of Music.” Music Analysis 8, nos. 1–2: 5–20.

—————. 1995. Performing Music: Shared Concerns. Oxford: Oxford University Press.

Eigeldinger, Jean-Jacques. 1998. Chopin vu par ses élèves. Boudry-Neuchâtel: La Baconnière.

Hatten, Robert. 2014. “Performing Expressive Closure in Structurally Open Contexts: Chopin’s Prelude in A Minor and the Last Two Dances of Schumann’s Davidsbündlertänze.” Music Theory Online 20, no. 4.

Hauptmann, Moritz. 1853. Die Natur der Harmonik und der Metrik. Leipzig: Breitkopf und Härtel.

Hepokoski, James, and Warren Darcy. 2006. Elements of Sonata Theory. Oxford: Oxford University Press.

Higgins, Thomas. 1966. Chopin Interpretation: A Study of Performance Directions in Autographs and Other Sources. PhD diss., University of Iowa.

Kopp, David. 2002. Chromatic Transformations in Nineteenth-Century Music. Cambridge: Cambridge University Press.

Leong, Daphne, and David Korevaar. 2005. “The Performer’s Voice: Performance and Analysis in Ravel’s Concerto pour la main gauche.” Music Theory Online 11, no. 3.

Lester, Joel. 1995. “Performance and Analysis: Interaction and Interpretation.” In The Practice of Performance, ed. John Rink, 197–216. Cambridge: Cambridge University Press.

McCreless, Patrick. 2011. “The Pitch-Class Motive in Tonal Analysis: Some Historical and Critical Observations.” Res Musica 3: 52–67.

Milhous, Judith, and Robert D. Hume. 1985. Producible Interpretations: Eight English Plays 1675–1707. Carbondale: Southern Illinois University Press.

Narmour, Eugene. 1988. “On the Relationship of Analytical Theory to Performance and Interpretation.” In Explorations in Music, the Arts, and Ideas: Essays in Honor of Leonard B. Meyer, ed. Leonard B. Meyer, Eugene Narmour, and Ruth A. Solie, 317–40. Stuyvesant, NY: Pendragon Press.

Poli, Roberto. 2010. The Secret Life of Musical Notation: Defying Interpretive Traditions. Milwaukee: Amadeus Press.

Rink, John. 1988. “The Barcarolle: Auskomponierung and Apotheosis.” In Chopin Studies, ed. Jim Samson, 195–220. Cambridge: Cambridge University Press, 1988.

—————. 1990. Review of Musical Structure and Performance by Wallace Berry. Music Analysis 9, no. 3.

—————. 2002. “Analysis and (or?) Performance.” In Musical Performance: A Guide to Understanding, ed. John. Rink, 35–58. Cambridge: Cambridge University Press.

Rothstein, William. 1995. “Analysis and the Act of Performance.” In The Practice of Performance, ed. John Rink, 217–40. Cambridge: Cambridge University Press.

Samson, Jim. 1985. The Music of Chopin. London and Boston: Routledge and Kegan Paul.

—————. 1996. Chopin. New York: Schirmer Books.

Schmalfeldt, Janet. 1985. “On the Relation of Analysis to Performance: Beethoven’s Bagatelles Op. 126, Nos. 2 & 5.” Journal of Music Theory 29, no. 1: 1–31.

—————. 2005. “Response.” Music Theory Online 11, no. 1.

Schenker, Heinrich. (1935) 1979. Der freie Satz, Translated by E. Oster as Free Composition. New York: Longman.

Chopin, Barcarolle, op. 60, editions cited

Ekier, Karl, ed. 2002. Complete Works of Chopin, vol. 12. Warsaw: PWM edition.

First (English) edition. 1846. London: Wessel & Co.*

First (French) edition. 1846. Paris: Brandus & Cie.*

First (German) edition. 1846. Leipzig: Breitkopf & Härtel.*

Herttrich, Ernst, ed. 1978. Frédéric Chopin: Klavierstücke. Munich: G. Henle Verlag, 86–97.

Klindworth, Karl, and Xaver Scharwenka, eds. 1882. In Chopin, Complete Piano Works, vol. 10. London: Augener Editions , n.d., n.p.*

Kullak, Theodore, ed. 1882. In Friedrich Chopin’s Werke, Instruktiv Ausgabe, Band XIII. Berlin: Schlesinger'sche Buch-und Musikhandlung, 38–45.*

Mikuli, Carl, ed. 1895. In Frédéric Chopin, Complete Works for the Pianoforte, vol. 10. New York: G. Schirmer, 9–19.*

Paderewski, Ignacy, Ludwik Bronarski, and Józef Turczyński, eds. 1954. Frédéric Chopin: Complete Works (Fryderyk Chopin Dziela Wszystkie), vol. XI. Warsaw: Instytut Fryderyka Chopina, (printed Cracow, 1975), 31–40.

Scholtz, Hermann, ed. 1879. In Fr. Chopin Sämtliche Pianoforte-Werke, Band III. Leipzig: C.F. Peters, n.d., 9–19.*

*Accessed at Chopin’s First Editions Online, http://www.cfeo.org.uk.

**Accessed at Petrucci Music Library, IMSLP, http://imslp.org/wiki/Barcarolle,_Op.60_(Chopin,_Frédéric)

Manuscript facsimile images accessed at:

http://en.chopin.nifc.pl/institute/publications/facsimile/id/646
 Original in the Jagiellonian Library, Kraków.

http://pressandpolicy.bl.uk/Resource-Library/First-page-of-Chopin-s-famous-Barcarolle-1845-f5.aspx
 Original in the British Library, London.

Discography

Ashkenazy, Vladimir. 1968. Decca 591089.

de Larrocha, Alicia. 1995. RCA Red Seal 68959.

Lipatti, Dinu. 1948. Columbia 33CX 1386.

Rubinstein, Artur. 1999. EMI 1928, reissued in Rubinstein Collection, vol. 4, RCA Red Seal 62000.

Rubinstein, Artur. 1962. RCA 5617-2-RC.

Footnotes

1. This article is an expansion of a talk originally given in December 2008 at a conference held at Princeton University in memory of Edward T. Cone, and in a revised version in November 2010 as part of a special session on performance and analysis of nineteenth-century music at the Society for Music Theory annual meeting in Indianapolis.
Return to text

2. See Kopp 2002, 1–17.
Return to text

3. Schmalfeldt, summarizing and reflecting on these criticisms some two decades later, asserted that her attempt was to create “an utterly equal partnership” between her alter egos, noting that “I assumed that both of my characters would be easily recognized as aspects of myself” (2005, [3]).
Return to text

4. “Prescriptive” does not necessarily imply a negative value, insofar as any performance decision resulting from analysis can be thought of as prescriptive (see Rink 2002, 37).
Return to text

5. Thoroughly updating these analyses with respect to recent theories of sonata form would be an illuminating exercise that I will not accomplish here, although some clear correspondences will be identified.
Return to text

6. The structure of this tripartite section conforms to William Caplin’s (1998, 71) definition of the small ternary theme-type, characteristic of a range of forms.
Return to text

7. For one description of the practice, see Blum 1980, 15–22.
Return to text

8. The analysis of this passage, inspired by a performance question, in turn inspired some research. The issue of dynamic and articulation markings in Chopin’s music is complex, full of variants in manuscripts, first editions, and later editions; in many cases no single source can be taken as definitive. Even with modern critical editions, it can be instructive for the performer to study the sources and earlier editions to ponder the range of alternatives in problematic passages—a reminder that the score can provide only so much information about a piece, and that the idea of a piece changes over time. In measure 6 of the Barcarolle, well-known twentieth-century editions such as Paderewski and Henle/Herrtrich, from which I learned the piece, continue the diminuendo through to the downbeat of measure 7, the registral and metric high point of the gesture. This appears to run directly contrary to the natural phrase arching and creates an interpretive paradox. See for example Poli 2010, 6–7, although an agogic stress would still be possible at the end of the diminuendo. However, all three first editions of the Barcarolle, from 1846, have differing information. The French edition has the diminuendo extending only to the beginning of the fourth beat of measure 6, while the German edition has the diminuendo extending to the second eighth note of the fourth beat. The English edition lacks any diminuendo. Late nineteenth-century editions such as Peters/Scholtz 1880 and Schirmer/Mikuli 1895 also have diminuendos ending at or in the fourth beat, as does the recent National Edition, edited by Jan Ekier. This alternative leaves room at the end for a subdued but noticeable rise to the high point, with dynamic and agogic stress appropriate to this first in a series of related gestures at the beginning of the piece.
Return to text

9. See Higgins 1966, 77, based on a contemporary report.
Return to text

10. Artur Rubinstein’s fiery performance from 1928 is exemplary of this approach, in contrast with his later recordings.
Return to text

11. A pitch class can be understood as “motivic” if it returns in prominent, varied, and marked ways over the course of a movement or piece. McCreless (2011) traces the rise in usage of this term, first articulated by Steve Laitz in 1992.
Return to text

12. Rink, in the analysis shown in Figure 2 (1988), identifies C♯ as the Barcarolle’s Kopfton and stresses the importance of C♯ and F♯ as structural pitch classes. In a private conversation in 2009, however, he mentioned his shifting preference toward a $\hat{3}$-line Urlinie with A♯ as its Kopfton.
Return to text

13. The more recent National/Ekier edition has the same markings.
Return to text

14. There are two extant autograph manuscripts of the Barcarolle, both engraver’s copies. One autograph is consistent with the French first edition in measure 14. The other autograph is nearly consistent with the German first edition, although the diminuendo begins on the third eighth note of the first beat in the autograph, and on the second eighth in the print edition.
Return to text

15. The potential gain for the listener was also instantly demonstrated, as a family member familiar with my performance of the piece happened to listen in at that moment, and declared, “Wow, that sounded good!”
Return to text

16. The mid-beat climax suggested by the French first edition suggests an enhancement to this performance approach, setting up an unfulfilled expectation of stress on the second beat.
Return to text

17. Continuing the sonata analogy, this passage would represent the medial caesura (Hepokoski and Darcy 2006, 23–50). After this point, though, while the analogy can be extended, formal organization becomes more idiosyncratic.
Return to text

18. The Neo-Riemannian operations for the two relationships are similar: PL for A♯ major to F♯ major; PR for F♯ major to A major (or Pr and PR in my terminology), implying a similar transformational or voice-leading distance. These labels, however, are neutral regarding the harmonic values of the individual chord members—root, third, and fifth—which in turn contribute to the qualities of the relationships.
Return to text

19. Likewise, in the return from A major to F♯ major, which does not occur directly in the piece, the root of the first is displaced by a semitone into the second, while the approach to the tonic from below is by whole step.
Return to text

20. Kopp 2002, 5–8, 18–32.
Return to text

21. The analyst could readily enough produce more comprehensive diagrams than those in Figure 4 to account for the structural status of every note in themes A, B(2), and the sfogato theme. But this level of detail is not necessary for the performer to understand the thematic similarities.
Return to text

22. The diminuendo and ritenuto are not present in all editions.
Return to text

23. The French first edition has f marked at measure 103, a crescendo hairpin extending throughout measure 106, and sempre f at measure 107. The English and German editions have sempre f marked immediately at measure 103, the crescendo hairpin throughout measure 106, and nothing afterwards, suggesting a ff dynamic level. Both variants appear in later editions about equally. But in all cases an unbroken intensification is indicated.
Return to text

24. Samson subsequently abandoned this analysis, agreeing with the formal closure indicated in Rink’s Schenkerian graph (Samson 1996, 267).
Return to text

25. There is a potential resonance here with the notion of structural vs. expressive closure that Robert Hatten introduces in this issue (Hatten 2014). While I am arguing that in the Barcarolle both measures 103 and 111 participate in a process of structural closure, each has a different quality.
Return to text

26. Mikuli 1895, Kullak 1882, and Klindworth and Scharwenka 1882 respectively.
Return to text

27. The various editions show great variability here at one of the most dramatic moments of the piece. The French first edition has nothing at measure 110 b1, continuing the sempre f of measure 107, and sf followed by p at measure 111 b1. The English edition has sf at both measures 110 b1 and 111 b1, followed by p in the latter. The German edition has nothing at measure 110 b1, and fp at measure 111 b1. All of these variants found their way into later editions. The Peters/Scholtz shown in Example 10 follows the German, while the Henle/Herrtrich shown in Example 11 follows the English, with stress on both downbeats—my own preference, since this highlights the intervallic resolution.
Return to text

Copyright Statement

Copyright © 2014 by the Society for Music Theory. All rights reserved.

[1] Copyrights for individual items published in Music Theory Online (MTO) are held by their authors. Items appearing in MTO may be saved and stored in electronic or paper form, and may be shared among individuals for purposes of scholarly research or discussion, but may not be republished in any form, electronic or print, without prior, written permission from the author(s), and advance notification of the editors of MTO.

[2] Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:

This item appeared in Music Theory Online in [VOLUME #, ISSUE #] on [DAY/MONTH/YEAR]. It was authored by [FULL NAME, EMAIL ADDRESS], with whose written permission it is reprinted here.

[3] Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.

This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.