Reuniting the Muses: Cross-Disciplinary Analysis of Debussy’s Pelléas

and Prélude à “L’après midi d’un faune





Edward D. Latham



In the pursuit of interdisciplinary analytical schema, the presence of language itself in the act of analysis would seem to inevitably privilege poetics, literary criticism, semiotics, and linguistic philosophy.[1] This has led analysts to overlook the vital insights offered by the disciplines most integral to the creation (and recreation) of multi-dimensional  works of art, particularly in opera and ballet studies where theories of dance and drama are rarely brought to the interpretive table. This essay will attempt to “reunite the muses” by examining theories developed by two of the most influential performing artists of the past century - Konstantin Stanislavsky (1863-1938), the Russian actor, director, and co-founder of the Moscow Art Theatre, and Rudolf von Laban (1879-1958), the Hungarian-born choreographer and dance theorist - with regard to their definition of dramatic and gestural closure in opera and ballet, respectively. Analyses of excerpts from two works by Debussy, the opera Pelléas et Mélisande and the ballet Prélude à ‘L’après-midi d’un faune’ [Prelude to the Afternoon of a Faun], will exemplify the ways in which the theories of Stanislavsky and Laban can fundamentally alter the interpretation of opera and ballet, both for the analyst and for the performer.



Closure as an Inter-artistic Construct



In his study of parallelism in the arts, the aesthetician James Merriman notes that in order to compare features of music and drama those features must be possible in both mediums. He lists repetition, contrast, reversal, juxtaposition, and heterogeneity as potentially analyzable features.[2] If one were to distill Stanislavsky’s system of dramatic objectives down to a single feature, however, that feature would be closure. A character’s dramatic success is defined by the attainment of local objectives, main objectives, and a superobjective, and the attainment of each objective represents a kind of dramatic closure, a closing of a chapter in the character’s history. Obviously, though its definition varies depending on stylistic context, closure is also a prominent feature of music, and therefore it meets Merriman’s basic requirement for analysis.



Stanislavsky defined an objective as ‘the goal of a character’; the actions of the characters in a play are therefore motivated by their desire to achieve their objectives.[3] In order to arrive at a complete understanding of the role he or she is to play, the actor must undertake what Stanislavsky referred to as ‘the scoring of the role’, a process during which objectives are identified and refined.[4] To begin with, the role is divided hierarchically into units of varying length: the entire role, its various scenes and their subsections, and the individual lines themselves. Each unit is then assigned an objective: Stanislavsky used the term ‘superobjective’ to describe the character’s overarching goal for the entire play and identified the character’s objectives for each scene as ‘main objectives’.[5] The score of a role is typically constructed in the form of an outline, with the superobjective at the top, main objectives as the section headings, and line-by-line objectives as subheadings. Because the proceeding analyses address only local moments in Debussy’s works, complete outlines of character objectives will be eschewed, though they can be inferred from the ensuing commentary.



Closure in Pelléas et Mélisande


Though large sections of Debussy’s score are given over to non-functional triadic harmony, as well as to the octatonic, chromatic, and whole-tone pitch collections, he employs isolated instances of linear tonality to amplify or undercut points of dramatic closure or lack of closure. These moments generally involve one of three different types of dramatic situations, which might be called ‘positive’, ‘negative’ and ‘ironic’ situations, each differentiated by a unique relationship between the vocal line and the orchestra. In a ’positive’ situation, a character achieves a main objective, and both the orchestra and the vocal line close to the tonic. In a ‘negative’ situation, a character forfeits a main objective, and neither the orchestra nor the vocal line achieves closure, despite the expectations created by the establishment of tonality. ‘ironic’ situations, in which a character mistakenly believes an objective to be achieved, contain closure in the vocal line, indicating the character’s misconception, but an evasion of the tonic in the orchestra, either by means of a deceptive cadence or an evaded cadence in which the subsequent measures destabilize the tonic.


‘Positive’ situations are the most prevalent in Pelléas, and two examples are given in Figures 1 and 2. In the passage shown in Figure 1, from Act II, Scene 1, Mélisande’s objective is to convince Pelléas that the wedding ring given to her by Golaud, which has ‘accidentally’ fallen into a well, is not worth retrieving. Achieving this objective will help her to attain her main objective, which is to free herself from Golaud. She persuades Pelléas by telling him it is too far away, rippling the water so that the ring can no longer be seen, and then declaring that he was mistaken in saying he had seen it in the first place.


Debussy sets this moment as a 5-line in F major, with the fundamental descent taking place in Mélisande’s vocal line. The passage begins with a progression in mm. 92 to 93 that defines a first-inversion. F-major triad as tonic. The key is defined in a faintly mysterious way, however, due to the B natural in m. 92 that precludes the presence of the key-defining tritone, E-flat. Scale-degree 5 (C) is instituted as the primary tone in Mélisande’s vocal line on the word ‘loin’ [far] in ‘Elle est si loin de nous’ [it is so far from us], and is immediately decorated by an upper neighbor D. The orchestra moves to the dominant at the end of m. 94 to provide the harmonic context for the subsequent E half-diminished seventh chord, which becomes an extension of the dominant. Scale-degree 4 (B FLAT), is established in the melody at ‘Non, non, ce n’est pas elle’ [No, no, that its not it], and is prolonged through the unfolding of the tritone, B flat to E. 






Figure 1: Harmonic Voice-Leading Reduction of mm. 92-102 of Pelléas




Figure 2: Harmonic Voice-Leading Reduction of mm. 481-492 of Pelléas


At the arrival of scale-degree 3 in m. 96, the orchestra returns to the tonic, this time in unequivocal root position, and plays a transposition of the theme from mm. 92-93. In this way, the original harmonic ambiguity of the opening phrase is clarified. After a prolongation involving a consonant skip up to C in the melody, accompanied by a voice exchange in the orchestra at mm. 96 to 97, Mélisande returns to scale-degree 3, decorated by a double neighbor, at m. 99. The manner in which Debussy finishes the passage, with a strikingly accented chromatic descending-thirds progression followed by a brief, matter-of-fact descending-fifths progression, makes Mélisande’s question - ‘Qu’allons nous faire maintenant?’ [What will we do now?] - a rhetorical one. She does not intend to do anything about the lost ring; in fact, she meant to lose it.


The analysis of this brief excerpt is additionally enriched by the presence of numerous foreground motives, both melodic and harmonic, that act as indicators of the characters’ local line objectives. In Figure 1, for example, the dominant ninth, a symbol of desire throughout the opera, appears at m. 94, as Mélisande ripples the water, making the ring disappear. Moreover, Mélisande’s primary harmonic motive, F major, is the key of the excerpt, demonstrating her control of the situation, though Golaud’s motive lurks in the alternating major-seconds in the opening theme of the passage and its subsequent transposition.


Figure 2 shows a passage from Act III, Scene 4. In this scene, Golaud, whose superobjective is to keep Mélisande’s heart captive, is trying to persuade Yniold to tell him what he knows about the relationship between Pelléas and Mélisande. The attainment of this main objective will permit him to separate the two would-be lovers, thus bringing him closer to his superobjective. The excerpt begins with a prolongation of E, presented in Yniold’s melody at ‘Petit père! Vous m’avez fait mal!’ [Daddy! You have hurt me!], and it is E that may be identified retrospectively as the primary tone, scale-degree 3 in C major. At m. 485, E is coupled in the lower octave in Yniold’s melody at ‘Ici, ici, a mon petit bras’ [Here, here on my little arm]. Up to this point, the passage has projected an ambiguous sense of key: mm. 481 to 482 project A minor by sounding an A-minor triad on each downbeat and surrounding each of them with a  i-v-(iv6)-v6 progression, but the lack of a leading-tone G# leaves open the possibility for reinterpretation in C major. 



Measures 483 to 484, where the orchestra has a voice exchange that prolongs a B half-diminished seventh chord, begin to hint at C major, and the tension between the two keys is exploited by Debussy at m. 487, where he uses a rhythmic figure in the orchestra that recalls the harmonic progression from mm. 481 to 482. But now the progression sounds as though it is leading to a cadence in C major via the progression vi-II9-IV-V9. However, the April 12, 2006repetition at m. 488 of the same one-measure harmonic pattern places another A-minor triad on the downbeat, creating a deceptive cadence in C major. Debussy continues his cadential play in m. 489, where, after the same preparation, he puts a secondary dominant, [V9]® IV, on the downbeat, evading the cadence and pushing the line forward to a second deceptive cadence at m. 491 and finally to the perfect authentic cadence at m. 492.


Debussy’s avoidance of closure to the tonic emphasizes Golaud’s failure to get Yniold to cooperate. Golaud begins by apologizing for hurting Yniold’s arm, but this only leads to the deceptive cadence at m. 488. He then tries ordering Yniold to stop crying, but the evaded cadence at m. 489 shows Yniold has not been won over. Finally, he bribes Yniold with the promise of a bow and quiver - not exactly the perfect gift for a cherub between two would-be lovers! Yniold is clever, though: he demands to know what the bribe will be before offering any information and Golaud is required to make a concrete offer, hence the second deceptive cadence before closure to the tonic.


The linear continuity created by the dialogue between the vocal parts further enhances the cadential disruptions, particularly the evaded cadence. After Yniold’s initial prolongation and register transfer of E in mm. 481 to 485, Golaud moves the fundamental line to scale-degree 2 via a consonant skip from A to D at m. 488. Thus, when scale-degree 1 arrives over the [V9] ® IV in the next measure, it sounds more like an evasion because the goal tone has been prepared in the melody. Yniold sings scale-degree 2 at m. 490, revealing his interest in the bribe. He picks up the line and moves it forward; Golaud has made progress. When Golaud takes over the melodic line and provides the final closure to the tonic, he regains control of the situation.


A ‘negative’ situation is illustrated in Figure 3, an excerpt from Act IV, Scene 2. In this scene, Arkel, Golaud’s grandfather and the king of ‘Allemonde’, wants to embrace Mélisande. He is unable to achieve this main objective, however, because his tête-á-tête with Mélisande is interrupted by Golaud. The passage begins with the initiation of the primary tone B (5 in E major) in Arkel’s vocal line. After prolonging scale-degree 5 with a skip up to scale-degree 7 (D#) over a dominant ninth (the symbol of desire again), Arkel returns to scale-degree 5 over the tonic at m. 165. A pair of unfoldings ensue, prolonging the primary tone by expanding the upper neighbor C# from Arkel’s opening motive into a three-measure structure. The vocal line then moves to scale-degree 4, supported by the dominant, and there is an evaded cadence in the orchestra as Arkel asks ‘As tu peur de mes vieilles lèvres?’ [Are you afraid of my old lips?] One can imagine that, just as he is about to kiss her, Mélisande shrinks away, and the cadence is evaded.


Mélisande introduces scale-degree 3 over I6 at m. 175. She then embellishes it with a chromatic pitch, G natural, that is part of an inverted statement of pitch-class set 3-7 [025], the set-class form of her primary leitmotive. Arkel will not be deterred from E major, however, and he pushes the line down to F# (scale-degree 2) over the dominant in m. 179, placing Mélisande’s G natural into the context of E major as a chromatic passing tone. Scale-degree 2 is prolonged with an arpeggiation of the supertonic triad in the vocal line, but closure to the tonic is avoided in both the melody and the accompaniment. Arkel trails off awkwardly on the same incongruous F natural which set the earlier use of the word ‘mort’ [death] at m. 169, and the orchestra embellishes the tonic with a pair of suspensions deriving from Golaud’s motive. The harmony continually falls back to the dominant ninth, which is eventually reduced to just the two notes of Golaud’s motive as he enters and interrupts the encounter at mm. 185 to 186.




Figure 3: Harmonic Voice-Leading Reduction of mm. 203-205 of Pelléas





Figure 4: Harmonic Voice-Leading Reduction of mm. 388-401 of Pelléas



Figure 4 is an example of an ‘ironic’ situation, in which the protagonist believes an objective to be achieved, but the omniscient orchestra indicates otherwise. Figure 4 shows a passage from Act III, Scene 3, in which Golaud and Pelléas, having just left the catacombs beneath the castle, discuss the rendezvous between Pelléas and Mélisande at the tower the night before, which Golaud discovered. Golaud’s main objective in this scene is to shame Pelléas into staying away from Mélisande.


Beginning with motion to the dominant in E flat minor, the orchestra supports the initiation of the primary tone (B flat) in Golaud’s melody at the word ‘soir’ [night]. The harmonization of scale-degree 5 with the dominant instead of the tonic emphasizes the rough discontinuity of Golaud’s change of subject. Impatient with Pelléas’s inability to comprehend the meaning of his subtle threats and intimations in the previous scene, Golaud cuts right to the chase, without bothering to establish an E-flat minor tonic first. Golaud leaves no doubt about his goal, however, when his line arpeggiates the tonic triad at m. 391, and the orchestra affirms the tonic by entering on Eb in the next beat. 


Harmony, line, and motive are merged beautifully at mm. 393-394, when scale-degree 4, played by the orchestra and echoed by Golaud, is harmonized by a four-part setting of Mélisande’s symmetrical 3-7 motive in the orchestra, at its original transpositional level (the harmonic progression is II-V-VII-V-II). The orchestra then moves to the submediant (acting as a tonic substitute) at m. 396, to support scale-degree 3 in Golaud’s melody. In the following measure, Golaud’s impatience leads him to move to scale-degree 2 several measures too soon, rather than over the structural dominant. Then, in a violent outburst, he leaps away from scale-degree 2, reaching up to D flat, his highest note in the passage. Music theorist Walter Everett has described events like this one as moments where a character’s excesses of emotion cause him to override the linear progression of the melody, and that is what clearly happens here, as evidenced by the appearance of Melisande’s motive in Golaud’s melody in mm. 397 to 398.[6] When the orchestra reasserts scale-degree 2 at m. 400, Golaud, who considers the matter closed, does not pick up on the cue but instead jumps directly from scale-degree 5 to the tonic. The orchestra, knowing the truth of the matter, enters on b7, creating the effect of a [V42]  ® IV with the sparsest of means, but nonetheless evading harmonic closure.



From Analysis to Criticism: the Welsh National Opera Pelléas


Stanislavsky always intended his system to be put to practical use, not to remain solely at the theoretical level. By highlighting the ways in which Debussy’s score reinforces or undercuts the characters’ attainment of their objectives, the four analyses above might be of practical use to directors, conductors, and performers preparing a production of Pelléas. In order to remain faithful to the original sprit of Stanislavsky’s work, some of the possible applications of the knowledge gained from the analyses will now be discussed with regard to two excerpts from the Welsh National Opera video recording of Pelléas.[7]


The first excerpt, which corresponds to Figure 1, shows Alison Hagley as Mélisande, acting thoroughly distraught at the loss of the wedding ring given to her by Golaud. At the moment where the rapid descending arpeggio is heard in the orchestra (which I have interpreted as Mélisande’s attempt to hide the ring’s descent by rippling the water), Hagley makes a somewhat gratuitous cross to stage right. Even though the conductor, Pierre Boulez, accentuates the decisive, almost playful nature of the cadence in the orchestra by exaggerating Debussy’s staccato articulation and slightly increasing the tempo, director Peter Stein chooses to have Hagley remain genuinely agitated, leaving her nothing to play but vague emotion. Neill Archer, as Pelléas, however, registers his understanding of the meaning behind her question ‘What shall we do now?’ His head snaps up and his eyes widen as he replies, ‘You must not worry so about a ring.’


An interpretation of the same passage that incorporated an understanding of its musical and dramatic closure might make Melisande’s reaction more ambiguous. At the line, ‘Non, non, ce n’est pas elle’ [No, no, that is not it], Melisande could create a slight accelerando to show her determination, bending down and rippling the water with her hand. At ‘ce n’est plus elle’ [That is no longer it], she could place an accent on the word ‘elle’, straightening and drying her hands on her skirt, as if giving up the search. On her final line, ‘Qu’allons-nous faire maintenant?’ [What will we do now?], she ought to look directly at Pelleas, slightly clipping the final F natural to make the question a bit more pointed.


The second excerpt, which corresponds to Figure 3, presents the opposite problem. Whereas, in the scene at the well, Stein chose to ignore the tonal closure in both the orchestra and the vocal line, in Act IV, Scene 1 he does not take into account the lack of closure in the music. Instead of exploiting the tension between Arkel’s desire to embrace Mélisande and her desire to avoid the embrace, Stein permits Arkel not only to kiss Mélisande, both on the forehead and on the mouth, but also to caress her in a very intimate fashion. Although Hagley as Mélisande is clearly aware of the implications of Arkel’s request for her to come closer, since she demurely resists the pull of his arm, Stein’s interpretation of their encounter makes Golaud’s subsequent arrival over the unresolved dominant ninth seem less like an interruption than like a guilty coda - a husband walking in on the lovers the next morning.


Just as opera directors, conductors, and singers might benefit from the study of musico-dramatic closure and the lack thereof, choreographers and dancers might gain valuable insights into the works they create and perform through the use of ‘linear-gestural’ analysis. The notational system developed by Rudolf von Laban provides a stable and reliable medium in which to present a dance score for the kind of detailed study such an analysis would require, and it will therefore serve as a basis for the remainder of the present study.



Exploring Closure in Dance Through Labanotation



Rudolf von Laban was born in 1879 and spent his youth traveling throughout the Austro-Hungarian Empire with his father, a military governor.[8] After abandoning the military academy in which his father had pressed him to enroll, Laban traveled to Paris, familiarizing himself with the early dance notation system of the eighteenth-century French choreographer Raul-Auger Feuillet, and studying dance, art and architecture. He later moved to Munich, Zürich, and, after enduring persecution under the Third Reich, settled in London. He published ten books on dance, including his Schrifttanz [Written Dance] of 1928, his most important and influential work, and founded several dance institutes and organizations. He died in Surrey, England in 1958.


The most significant accomplishment of Laban, who is to dance theory what Schenker is to music theory and Stanislavsky is to dramatic theory, was the development of a comprehensive system for the notation of choreography, which he called ‘kinetography’, or the study of drawing movement. Later, American proponents of his notational system dubbed it ‘Labanotation’, and it rapidly became the gold standard of dance notation systems. As structured by Laban, a ‘kinetogram’, or dance score, is written on a three-line staff adapted from the standard five-line staff used in musical notation.[9] The second and fourth lines are omitted for the sake of visual clarity, but the insertion of the subsequent symbols is carried out with these hidden lines in mind. The staff is arranged vertically, in order to show forward progression or momentum, but can be rotated ninety degrees to run parallel to the music, if necessary.


In the introductory section of his ‘dictionary’ of Labanotation, Albrecht Knust, one of Laban’s foremost students and collaborators, provides an elegant outline of the basic aspects of Labanotation. Figures 5 and 6 reproduce excerpts from Knust’s first two pages of examples. As shown in Figure 5, the center line divides the dancer’s body into left and right halves, with movements performed by the right side of the body (that is, the dancer’s right) notated to the right and vice versa. Leg movements are notated immediately to the right and left of the center line if they support the weight of the body, as in walking, and in the next column if they do not, as in a gesture like the tendue [extension of the leg]. Movements of the upper torso and arms are notated in the outer columns. Figure 6 shows the basic symbols for direction and height, the former indicated by the shape of the symbol (as in the triangle pointing to the right), and the latter indicated by shading (striated for high, blank with a dot for medium, and filled for low). The length of a particular symbol on the staff represents its duration, as shown at the bottom of Figure 6. Figure 7 shows the basic symbols for direction, indicated by the shape of each symbol (e.g. the triangle pointing to the right). Height is indicated lby shading (striated for high,blank and a dot for medium and filled for low). The direction symbols in Figure 7 are shown in the high position.


Although space limitations prevent an explanation of them here, there are eight other categories of symbols, in addition to the  direction signs, including position signs, body signs, and relation signs, among others. Perhaps the most interesting of these are the preliminary indications, or ‘presigns’ given in the choreographical score before the dance begins, two categories of which are likened by Knust to the key signature and clef in musical notation.



Figure 5: The Center Line of the Body and the Limbs (Knust)





Figure 6: Indicating Duration of gestures (Knust)




Figure 7: The Direction Signs (Knust)




Nijinsky and Dance Notation


One of the primary purposes of Labanotation is the preservation and reconstruction of historical choreography, the finer details of which would eventually be lost due to the vagaries of the oral tradition of passing them from teacher to student. It was for this purpose that Ann Hutchinson Guest and Claudia Jeschke created their 1991 transcription of the dance score for Debussy’s Prélude à ‘L’après-midi d’un faune’, as choreographed by Vaslav Nijinsky.[10] Unlike memory-based Labanotation scores that represent the culmination of an oral tradition, however, Guest and Jeschke’s score of Faune is a translation of Nijinsky’s own notated score, which he created in 1915 using a system he developed himself.


Intriguingly, Nijinsky’s system, like Schenker’s, borrows elements of musical notation for many if its symbols (see Figure 8). Duration of gestures, for example, is indicated using traditional note values on a five-line staff. Based on the Stepanov system taught to him as a student at the Maryinsky Theater School in Russia, Nijinksy’s system nevertheless includes some significant improvements, including the use of a separate staff for each section of the body, the use of rests for exits and ties to indicate the retention of a particular gesture, and, perhaps most importantly, the use of ascending or descending ‘pitch’ to indicate the angle or direction of a gesture.


Because it is difficult to scan quickly and interpret accurately, however, it is doubtful that Nijinsky’s system will replace Labanotation as the primary system of dance notation. Nonetheless, Guest and Jeschke’s research into his score for Faune reveals a previously hidden aspect of Nijinsky’s professional persona: that of the theorist and analyst. It is this aspect of Nijinsky that warrants further investigation with respect to his score for the Prélude.




Closure in Debussy and Nijinsky’s Score


In creating his now-famous interpretation of Debussy’s Faune, from which musical elements did Nijinsky draw his inspiration, and how does his choreography interact with the music? An examination of ‘L’après-midi d’un faune’, the 1867 eclogue by Stephane Mallarmé to which Debussy’s title makes reference, reveals only general correspondence between it and Nijinsky’s ballet (i.e., the faun as central character, his encounter with the nymphs, and the use of grapes); in its details, the poem differs markedly from the story created by Nijinsky. In the eclogue, the faun meets only two nymphs, for example, not seven, and his subsequent ménage-à-trois with them has much more explicit and graphic sexual overtones than the comparatively innocent narrative portrayed in the ballet. It is ironic, then, in retrospect, that the ballet, and particularly its closing scene (where the Faun caresses the dress left behind by his Nymph), created such an uproar when it was premiered in Paris in 1912.





Figure 8: Nijinsky’s Notation System


The disparity between Mallarmé’s and Nijinsky’s versions of the story highlights an interesting ambiguity in Debussy’s title. Interpreted literally, ‘Prelude to the Afternoon of a Faun’ could signify music that describes an episode prefiguring the Faun’s afternoon adventures. Perhaps the Faun, inspired by his encounter with the Nymphs as portrayed by Nijinsky, returns later to seek them out, as portrayed by Mallarmé. Yet, it is equally possible that the title is meant to infer a more direct connection between the music and Mallarmé’s eclogue, as in ‘music for the afternoon of a faun’. It is to Debussy’s music and Nijinsky’s choreography that we must turn to give preference to one interpretation or the other.



As is often the case with Debussy scholarship, the analytical work that has been done on the Prelude divides into two camps. On the one hand, there are those, chief among them Richard Parks, who would claim it is an early work by Debussy the modernist: they tend to minimize its diatonic elements in favor of its motivic intervallic structures and chromaticism and generally prefer pitch-class set theory to tonal theories.[11] On the other hand, there are those, including Felix Salzer, Charles Burkhart, and Matthew Brown, who would argue that it is the work of Debussy the late Romantic: they tend to maximize its diatonic elements, absorbing as much of its recalcitrant chromaticism as possible with the porous sponge of Schenkerian theory.[12] In the case of the Prélude, though, a tonal reading is both appropriate and compelling. In his zeal to unearth complement relations, 4-17/18/19 complexes, and other atonal pitch relations, Parks seems to overlook a defining feature of the Prelude’s musical structure - namely, its restless search for, and drive toward, the root-position tonic E-major triad that arrives only at m. 106, four bars from the end of the piece.


The Schenkerian studies of the Prelude are largely in agreement on its broad formal and harmonic outlines: each of them chooses E major as the work’s tonic key, and none of them departs radically from the formal plan proposed most recently by Matthew Brown in 1993 (see Figure 9). They do differ, however, in both the scope of their analyses and the degree to which they are willing to subsume the B section of the Prelude (which outlines D flat Major) under an E-major background structure. Salzer, for example, analyzed only the first thirty measures of the piece, stopping conveniently short of the first major chromatic episode in mm. 31-36. Burkhart, too, tackles only a portion of the work (mm. 37-55), though his illustration of the chromatic ascent B-C-D flat as an enharmonic motivic enlargement of the B-B#-C# ascent of mm. 1-2 lays important groundwork for Brown’s subsequent analysis of the complete piece.


Matthew Brown’s analysis is by far the most daring. Not only does he analyze almost every bar of the Prelude (mm. 14-20 being the notable exception), he accounts for the thorny chromatic (whole-tone) episode of mm. 31-36 by relating it to the two occurrences of the structural dominant that surround it in mm. 30 and 37, and he incorporates Burkhart’s B-C-D flat as a transition section to the secondary key of D flat major. The key to Brown’s analytical success is his enharmonic reinterpretation of D flat as C#, an upper neighbor to the primary tone B (scale-degree 5) that becomes part of an expanded 5-6-5 melodic progression. C# as upper neighbor to B is also the primary focus of Salzer’s analysis of mm. 1-30, and the two analyses complement each other very well.


While Brown’s analysis is convincing in many respects, it contains four major flaws. First, the 5-line he proposes as a background structure, the entire descent of which takes place in the span of three measures, contains little harmonic support for scale-degree 3 (his analysis of mm. 79-106 is reproduced in Figure 10). Though he shows scale-degree 3 as supported by the cadential six-four in m. 105, a paradigm well documented by David Beach, the B shown in the tenor is actually an A and a C# in the music, severely weakening the sense of tonic harmony in the first part of the measure and privileging V9 instead. Second, Brown does not provide an analysis of mm. 14-20, which  mark  the  motion  G#-A-A#  as  a  transposed  instance  of Burkhart’s ascending chromatic motive, now in a more prominent register. Third, in his eagerness to show Burkhart’s motive in its most flattering light, he obscures several important structural harmonies - notably the E-major tonic.





Figure 9: Prélude à l’après midi d’un faune, Formal Plan (Brown)






Figure 10:  Prélude , mm 79 - 106 (Brown)





Example 1:  Prélude à l’après midi d’un faune , mm 14 - 20




Figure 11:  Prélude , mm 35 - 77 (Brown)



An alternative reading of the Prelude as a 3-line is shown in Figure 12. Like Brown’s analysis, the reading shown in the figure depends on an enharmonic transformation: the primary tone G#, prolonged via an upper-neighbor A# in m, 17 and an upper-neighbor A in mm. 23 and 42, becomes Ab at m. 45. Ab is then prolonged by two middleground descents to F (mm. 45-55 and 63-74), and an enharmonically re-spelled lower-neighbor Gb (m. 62). The return of the home key at m. 79 brings with it a prolongation of E, scale-degree 1, via a lower-neighbor Eb, and the eventual reinstatement of G# as primary tone at m. 94. The fundamental line closes to the tonic at m. 106.




Figure 12:  Prélude , Alternate Analysis as 3-Line


In addition to accounting for the aurally salient head note of the B-section theme, the reading shown in Figure 12 incorporates the missing measures 14-20, which, rather than being merely transitional, constitute an important prefiguration of both the “bathing” theme at m. 37 and the B-section theme at m. 55. As shown in Figure 13, these three themes are connected to the opening measure of the Prélude, and to each other, through their prominent use of pitch-class set 3-7 [025], the set-class form of Mélisande’s motive. The dates of composition for the opera and the ballet overlap: Debussy began working on Pelléas et Mélisande in September, 1893 and had finished a complete draft by August 1895; the Prélude was begun in 1891, and premiered in December 1894. Set-class 3-7, then, along with set-classes 4-27 and 5-34, both used extensively in Pelléas to signify Golaud’s passion for Mélisande, can be seen as intertextual symbols of desire, resonating in both works.



In Nijinsky’s setting of the Prélude, the erotic connotations of 3-7 and 4-27 are made explicit. Whereas for the majority of the opening A section the Faun mimes playing the flute theme onstage, when 4-27, the A# half-diminished seventh chord, is arpeggiated in mm. 4 and 7 he stops and turns to look offstage, as if checking to see whether his sinuous melody has borne fruit. His longing is expressed in the subsequent arpeggiation of 3-7 as part of the B flat dominant seventh chord in mm. 8-10, as he slowly returns to his original posture. Again, in mm. 14-20, as 3-7 returns in the melody, he puts down the flute, and picks up some grapes as if to devour them: Example 2 (on pg. 42) shows the movement of his arms in the outer columns as he picks up the grapes.




Figure 13: Set-class 3-7 [025] (Mélisande/Nymph Motive)


For the most part, the divisions in Nijinsky’s plot correlate with the large-scale formal divisions proposed by Brown (refer back to Figure 9): the Faun begins alone on stage, and is joined by the Nymphs in mm. 21-28. The whole-tone episode in mm. 31-36 creates an appropriately mysterious and exotic atmosphere for the unveiling of the head Nymph, who begins bathing at m. 37. The encounter, and subsequent pas de deux, between the Faun and the head Nymph takes place during the B section, mm. 55ff, and the Faun returns to his rock alone during the A’ section, at m. 94. 


What is not shown in Brown’s formal scheme, however, is Nijinsky’s response to the cadential evasions in the B section. Though it begins in Db Major, the B Section soon moves to V9 of E major at m. 60. Resolution to the tonic is evaded, however, by a motion to iv64, an A-minor chord in second inversion, at m. 61, which coincides with the head Nymph’s first rejection of the Faun’s advances. The Faun then leaps into the air in m. 62, beckoning to her again as the harmony shifts to V9 of G major (see Example 3). As this new dominant ninth, enharmonically reinterpreted as being built on an E double flat dominant ninth, again evades resolution by moving down by step to a Db-major triad, the Nymph rejects the faun again. The culmination of their battle of wills occurs in m. 73, where they finally embrace by linking elbows (see Example 4). It is important to emphasize that Nijinsky chooses not to have their union occur on the resolution to the Db-major tonic in m. 74. Instead, he acknowledges the unfinished, interrupted nature of their encounter - the other Nymphs enter, Donna Elvira-like, and spoil the Faun’s fun - by breaking the embrace at m. 74. Frustrated, the Faun gestures defiantly at the Nymphs and moves away from the head Nymph. The three stages of the encounter are shown in Illustration 1, a series of photographs of Nijinksy in the leading tole taken by Baron Adolf de Meyer soon after the premiere in 1912.[13]




Example 2: Faun Picks up Grapes (mm. 15-18)




Example 3: Faun Leaps into Air (mm. 61-63)




Example 4: Faun and Nymph Embrace (mm. 72 - 73)





Illustration 1: Nijinsky in the Title Role (from Guest and Jeschke, 80).






By constantly returning to the music’s dramatic underpinnings and to the actual experiences of the characters involved in the scenes discussed, the close readings presented here aim to provide relevant and compelling interpretations that will affect audience members and performers alike. Though equal to Schenkerian analysis in the richness and sophistication of their descriptions of dramatic and gestural processes, the theories of Stanislavsky and Laban discussed here remain relatively unencumbered by detailed technical vocabularies of their own, enabling the analyst simply to redefine and broaden concepts like closure. Care must be taken, however, at the conceptual level to avoid the unduly reverential application of models from other disciplines that leads not to integration but further separation of the constituent disciplines. That tendency is counterbalanced in the case of the Pelléas analyses through the inclusion of Type III (ironic) relationships that disrupt the 1:1 correlation of musical and dramatic closure, whereas in the Faune analyses Laban’s system was not directly involved in the interpretive process. Because Laban’s system is purely notational, it is value-neutral: its purpose is merely to create a highly detailed and accurate score for subsequent performances.


The interpretations offered here strive to respond to Lawrence Kramer’s charge that, despite musicology’s evolution in recent years toward a more interdisciplinary perspective, particularly through the incorporation of critical theory, little change has been effected in the concept of music itself. [14] Although he is right to claim that ‘the music itself’, what Jean-Jacques Nattiez has called the ‘neutral level’, continues to exist as an independent and powerful force in analysis and criticism, readings like those presented here challenge one of its most basic assumptions: the notion that closure, completeness, and organicism are the most desirable states for a musical entity. In an multidimensional artwork such as an opera or ballet, the moments of greatest significance to an audience are often those that reveal discontinuities, conflicts, or disruptions that generate tension and require subsequent resolution or cancellation. These moments are Kramer’s hermeneutic ‘windows’, opening onto vistas of unexplored interpretive possibilities.[15]



1 The application of concepts from semiotics and literary criticism to the study of music has produced some of the most well-known music theoretical works of the past thirty years, including works by Fred Lehrdahl and Ray Jackendoff, Jean-Jacques Nattiez, Robert Hatten, Joseph N. Straus and Kevin Korsyn.

2 James D. Merriman, ‘The Parallel of the Arts: Some Misgivings and a Faint Affirmation,’ The Journal of Aesthetics and Art Criticism 31 (1972): 155-61.

3 Konstantin Stanislavsky, Creating A Role, trans. Elizabeth Reynolds Hapgood (New York: Theatre Arts Books, 1961), 78.

4 Ibid., 56. See also Konstantin Stanislavsky, Stanislavsky’s Legacy: A Collection of Comments on a Variety of Aspects of an Actor’s Life and Art, ed. and trans. Elizabeth Reynolds Hapgood (New York: Theatre Arts Books, 1958), 181.

5 Stanislavsky, Creating A Role, 78.

6 Walter Everett, ‘Singing About the Fundamental Line’, paper read at the annual meeting of the Society for Music Theory, Phoenix, AZ, 1997.

7 Claude Debussy, Pelléas and Mélisande, video recording, Welsh National Opera, dir. Pierre Boulez (Deutsche Grammophon, 1994), #6303065910.

8 Vera Maletic, Body - Space - Expression: The Development of Rudolf Laban’s Movement and Dance Concepts (Mouton de Gruyter: New York, 1987).

9 Albrecht Knust, Dictionary of Kinetography Laban (Labanotation), 2 vols. (MacDonald and Evans: Plymouth, 1979).

10 Ann Hutchinson Guest and Claudia Jeschke, Nijinsky’s Faune Restored, No. 3, Language of Dance Series, ed. Ann Hutchinson Guest (Philadelphia: Gordon and Breach, 1991).

11 Richard S. Parks, The Music of Claude Debussy (New Haven: Yale University Press, 1989), Chapter 7.

12 Felix Salzer, Structural Hearing: Tonal Coherence in Music (New York: Dover Publications, 1962), 100; Charles Burkhart, ‘Schenker’s “Motivic Parallelisms”’, Journal of Music Theory 22/2 (Fall 1978): 145-75; Matthew Brown, ‘Tonality and Form in Debussy’s Prélude à “L’après-midi d’un faune”’, Music Theory Spectrum 15/2 (Fall 1993): 127-143.

13 Guest and Jeschke, 80.

14 Lawrence Kramer, “Wittgenstein’s Chopin: Interdisciplinary and “the Music Itself,” 2003 conference paper presented at Interdiscipline: New Languages for Criticism, Centre for Research in the Arts, Social Sciences and Humanities (CRASSH), Sidney Sussex College, Cambridge University, Cambridge, England.

15 The concept of “hermeneutic windows” is described in Lawrence Kramer, Musicas Cultural Practice, 1800-1900 (Berkeley: University of California Press, 1990), passim.