Society for Music Theory

Tonality and Form in Debussy's "Prelude a 'L'Apres-midi d'un faune" Author(s): Matthew Brown Source: Music Theory Spectrum, Vol. 15, No. 2 (Autumn, 1993), pp. 127-143 Published by: University of California Press on behalf of the Society for Music Theory Stable URL: http://www.jstor.org/stable/745811 Accessed: 23/06/2009 15:21
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucal. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

University of California Press and Society for Music Theory are collaborating with JSTOR to digitize, preserve and extend access to Music Theory Spectrum.

http://www.jstor.org

and Form in Debussy's Tonality d'un faune" "L'Apres-midi

Brown Matthew
Seventy-five years after his death, Debussy is often remembered as "the quiet revolutionary," who breathed new life into musical art.1 He is credited with challenging the authority of nineteenth-century tonal and formal practice, and with pushing music gently into the twentieth century. To quote Pierre Boulez: Just as modern poetry surely took root in certain of Baudelaire's poems, so one is justifiedin saying that modern music was awakened by L'Apres-midi d'un faune.2 For his part, Debussy did much to promote this radical image of his music; throughout his letters and journal articles, he launched a bitter campaign against the musical establishment and conventional compositional practice. Most of these criticisms are general in nature, and ridicule what Debussy saw

Prelude

as a "silly obsession with overprecise 'forms' and 'tonality.' "3 Sometimes, however, they address concrete musical issues. For example, Debussy denounced accepted notions of chord function; in a letter to Pierre Louys (22 January 1895), he announced that "tonic and dominant had become empty shadows of use only to stupid children."4 Similarly, Debussy rejected the traditional distinctions between consonance and dissonance. With typical panache he insisted: Nothing is more mysterious than a consonant chord! Despite all theories, both old and new, we are still not sure, first, why it is consonant, and second, why the other chordshave to bear the stigma of being dissonant.5

1See chapter 1 of Arthur Wenk's book Claude Debussy and TwentiethCenturyMusic (Boston: Twayne Publishers, 1983), 1-19; and Pierre Boulez, Relevdsd'apprenti,ed. Paul Thevenin (Paris: Seuil, 1966), 336. I would like to thank various people for helpful suggestions, especially Jennifer Williams Brown, David Grayson, MarieRolf, Arnold Whittall,and the two anonymous readers for this journal. 2PierreBoulez, Notes to CBS Record 32 11 0056, quoted in Glenn Watkins, Soundings (New York: Schirmer, 1988), 75.

3See Debussy, "Musicin the Open Air," La Revue blanche, 1 June 1901, in Francois Lesure, ed. Claude Debussy: Monsieur Croche et autres dcrits (Paris: Gallimard, 1971), 45; Francois Lesure and Richard LanghamSmith, eds. and trans., Debussy on Music (New York: Knopf, 1977), 41. 4FrancoisLesure, ed., Claude Debussy: Lettres1884-1918 (Paris: Hermann, 1980), 70-73; Francois Lesure and Roger Nichols, eds. and trans., Debussy Letters (Cambridge:Harvard University Press, 1987), 76-77. SDebussy,"A propos de 'Muguette'.-Au ConcertLamoureux,"Gil blas, 23 March 1903, in Lesure, ed., MonsieurCroche, 129-30; Lesure and Smith, Debussy on Music, 155. I have changed the translationsof the words parfait and imparfait.

128

MusicTheory Spectrum Is it not our duty ... to try andfindthe symphonic formulae best suitedto the audacious discoveries our moderntimes, so comof mittedas theyareto progress? century aeroplanes a right has The of to a musicof its own!9 Besides disdaining textbook formal stereotypes, so Debussy also praised composers, such as Mussorgsky, whose works "are impossible to relate to (the) accepted forms-the 'official' ones."10 Intentions are one thing, but actualities are often quite another matter. While Debussy's radical goals are hard to deny, we may still wonder whether he succeeded in freeing himself from common-practice tonality and nineteenthcentury formal conventions. Did Debussy really create symphonic forms fit for "the century of aeroplanes"? Obviously, any answerto this question depends on the way in which we decide to explain tonality and on the kinds of analytical priorities we make in assigning formal functions. Although there are many possible theories of tonality, this paper will use Schenkerian theory. This decision, however, requires some explanation. In particular, it contradicts the prevailing view that Debussy's music cannot be analyzed by strict Schenkerian paradigms. Surely, Schenker would have disapproved;he made no secret of his dislike for Debussy's music. In the preface to his edition of Beethoven's Piano Sonata in A Major, Op. 101, he complained that the sequences of sounds found in impressionistpieces are valuable as "an acousticphenomenon, but certainlynot as art."1'Else9Lesure, Monsieur Croche, 241; Lesure and Smith, Debussy on Music, 297. 'oLesure,MonsieurCroche,29; Lesure and Smith, Debussyon Music, 21. ""Von einer gewissen 'Linie' sprechen heute gern auch die Verfertiger Stiicke. Wo aber, wie in diesen, die Wirkung sogenannterimpressionistischer erst auf ein Tongerauschhinauslauft(das wie jedes Gerauschnur als akustische Erscheinung, aber noch nicht als Kunst gilt), dort sagt die 'Linie' des Tongerauschesgewiss nicht mehr die Linien als die steigend, fallend ja auch in anderenGerauschen(z.B. in Donner, Tischriicken,Wagenrollenusw.) sich

He likewise scoffed at those who prohibitedthe use of parallel chords and proposed that tonality should be fully chromatic and "enriched by other scales."6 Just as Debussy debunked many basic tenets of commonpractice tonality, so he also questioned conventional approaches to musical form, especially those found in symphonic repertories. In an essay for La Revue blanche (6 March 1901), he dismissed a recent orchestral work by Witkowski, claiming that it provided "only further proof of the uselessness of the symphony since Beethoven."7 He added: at that Mustwe conclude despiteso manyattempts transformation, of the past?Has not its wornout gilt merelybeen replacedby a orchesplatingof shiningcopper,the shoddyfinishof present-day tration?8 Some ten years later, the barbs were no less pointed. Writing for SIM (1 November 1913), Debussy censured his colleagues for imitating the stale symphonic forms of Franz Liszt and Richard Strauss:
the symphony-in all its elegance and formal order ... -is a thing

6In Debussy's conversationswith his teacher Guiraud, he questioned the rules prohibitingparallel sonorities, claiming that "there is no theory. You have merelyto listen. Pleasureis the law." In the same conversationsDebussy also declared: "Music is neither major nor minor. Minor thirds and major thirds should be combined, modulation thus becoming more flexible. The mode is that which one happens to choose at the moment. It is not constant." These conversationsare translatedin William Austin, ed., Debussy: Prelude to "The Afternoon of a Faun," Norton Critical Scores (New York: Norton, 1970), 128-31. For a facsimile of Maurice Emanuel's original transcription, see Leon Vallas, Claude Debussy: His Life and Works, trans. Maire and Grace O'Brien (London: Oxford University Press, 1933), 84. 7Debussy, La Revue blanche, 1 April 1901, in Lesure, MonsieurCroche, 25; Lesure and Smith, Debussy on Music, 15. 8Lesure,MonsieurCroche, 26; Lesure and Smith, Debussy on Music, 16.

d'un faune" Tonalityand Formin Debussy's Prelude a "L'Apres-midi

129

where, he also condemned Debussy for pandering to the "mediocrity of French taste."12 For the most part, Schenker's followers have accepted the limitations of their methods for explaining Debussy's music.13 Some, such as Felix Salzer, have seen these limitations as grounds for modifying Schenker's original model and, in so doing, expanding the notion of tonality in music. For example, Salzer claims that, instead of being derived from a Schenkerian Ursatz, Debussy's piano Prelude "Bruyeres" is generated from a background progression built from a chain of parallel fifths -I-3 in A ."14 Others, such as Adele Katz, acknowledge that only a few of Debussy's pieces follow strict Schenkerian paradigms, and that many of them require some new type of explanation.15 Most recently, Richard Parks has claimed that, with some exceptions, only Debussy's early scores (up to ca. 1889) conform in any consistent way.16 As he puts it:
bemerkbarmachen konnen." See Schenker'sErliuterungsausgabe Sonate der Op. 101, Ludwig van Beethoven(Vienna: Universal Edition, 1921), 23. Cited by David Paul Goldman, "Esotericismas a Determinant of Debussy's Harmonic Language," The Musical Quarterly75 (1991): 146, fn. 20. '2"Undgar ein Debussy, der ware demnachnicht einmal noch ein Talent, ein Musikeriiberhauptzu nennen, mag er auch dem Mittelmassin Frankreich geniigen und dort aus Grtinden kunstpolitischerNatur-Frankreich treibt Politik und Ranke auch in den Kiinsten-sogar als ein UberwinderWagners, also schon aus diesem Grunde allein als ein Erneuerer nationaler Musik gelten" (Schenker, Das Meisterwerkin der Musik, vol. 3 (1930), 108). 13AdeleKatz, Challengeto Musical Tradition(New York: Knopf, 1945), chap. 7, 248-93; Felix Salzer, StructuralHearing (New York: Boni, 1952). 4Salzer, StructuralHearing, fig. 478. '5According to Katz: "Therefore, as both harmonic analysis and the Schenkermethod were evolved out of the tonal techniques, it is probablethat a new system of analysis is needed to understandthe new concepts" (Challenge, 293). For a convenient summaryof Katz's position see James Baker, "SchenkerianAnalysis and Post-Tonal Music," in David Beach, ed., Aspects of SchenkerianTheory (New Haven: Yale University Press, 1983), 154-55. '6RichardS. Parks, The Music of Claude Debussy (New Haven and London: Yale University Press, 1989), 4, 18, 20.

After the early works especially, one rarely finds tonic-dominant closure in a structuralsense. In Schenkerianterms, this means that bass arpeggiationis not an integral feature of underlyingstructure. Linear progressions are far less common than in the works of eighteenth- and nineteenth-centurymasters, as is the multilayered concatenation of contrapuntal-harmonicrelationships which normally accompaniesthem. In Debussy, linear progressionstend to be of the simplest sort and only of local significance.'7 He adds that structural levels are "few and uncomplicated," and that the richness of Debussy's music lies mostly in the
foreground.
18

Nevertheless, there are problems with both of these responses. In Salzer's case, his revisions modify Schenkerian theory to such an extent that they erode the entire foundation of the model. For example, by classifying chords according to their degree of dissonance, Salzer contradicts Schenker's claim that in tonal music there is an absolute distinction between the behavior of consonances and that of dissonances. This distinction is crucial to the way in which Schenkerian theory connects the rules of harmony and voice leading; once the distinction is abolished, Schenker's notion of transformation becomes seriously weakened. To quote Edward Laufer: By and by one asks what is left, as all the specific techniques Schenker describedmust all go by the board or be diluted into an indistinct
blur ....

sis, then anything goes and likewise nothing.19

[I]f there is no technically consistent, non-speculative ba-

Similarly, while Katz and Parks are justified in restricting the scope of Schenkerian theory-to be explanatory, all theories must have boundaries-neither they, nor any other Schenkerians, have ever defined these limits in a systematic
17Ibid.,4. s8Ibid. "9Edward Laufer, "Review of Free Composition,"MusicTheorySpectrum 3 (1981): 161.

130

MusicTheorySpectrum On the other hand, by graphingthe Prelude, we can answer some of the intriguingformal questions posed by Debussy's orchestralmusic. In particular,we will see how he developed four techniques-incomplete progressions, parentheticalepisodes, motivic compression, and tonal modeling-that feature prominentlyin the Nocturnes(1897-99), La Mer (19035), and the Images for orchestra(1905-13), and that allowed him to move away from nineteenth-centuryformal models. It is hard to imagine a single work that captures the spirit of Debussy's style more obviously than the Prelude a
"L'Apres-midi d'un faune"; ever since its premiere at the

manner. In fact, both of them consistently underestimatenot only the extent to which Debussy's mature music contains orthodox tonal material, but also the degree to which Schenkerian theory can cope with extreme chromaticism,modal or exotic harmonies, free dissonances, parallel chords and many other anomalies in Debussy's style.20Of course, this does not mean that Schenkeriantheory can explain every aspect of all works by Debussy; it simply means that the model can help us determine, case by case, how and why partially, or even marginally, tonal pieces sometimes sound tonal and sometimes do not. To show how Schenkeriantheory helps us understandthe tonality of Debussy's music, this paper will examine the Prelude a "L'Apres-midi d'un faune" (1891-94). The work is an

Societe nationale on 22 December 1894, musicianshave tried to explain the elusive logic alluded to by Debussy.22These
spotting," as he made clear in an essay for Gil bias, 23 February 1903 (see Lesure, ed., MonsieurCroche, 102-4; Lesure and Smith, Debussy on Music, 126-27). Debussy occasionallyrebuked specific theorists: Riemann (Lesure, ed., Lettres,165; Lesure and Nichols, eds., Letters, 183); Louis (Lesure, ed., Lettres,118;Lesure and Nichols, eds., Letters,128-30); Lenormand(Lesure, ed., Lettres,227; Lesure and Nichols, ed., Letters,259-60); Dubois (Lesure, ed., MonsieurCroche, 212; Lesure and Smith, Debussy on Music, 268-69); and Emile Durand (Lesure, ed., MonsieurCroche,29-30; Lesure and Smith, Debussy on Music, 21). 22For analysesof the Preludesee: WilliamAustin, "Towardan Analytical Appreciation,"in Debussy: Preludeto "TheAfternoonof a Faun," ed. Austin, 71-96; Barraque,Debussy (Paris:Seuil, 1962), 85-91; LaurenceBerman, "Prelude to the Afternoon of a Faun and Jeux: Debussy's Summer Rites," Music 3 (1980): 225-38; Leonard Bernstein, The UnanNineteenth-Century swered Question (Cambridge: Harvard University Press, 1976), 238-59; Charles Burkhart, "Schenker's 'Motivic Parallelisms,'" Journal of Music Theory22 (1978): 155-58; John Crotty, "SymbolistInfluences in Debussy's "Prelude to the Afternoon of a Faun," In Theory Only 11 (1982): 17-30; Music (London: BBC, 1974), 9-18; Denijs David Cox, Debussy's Orchestral Dille, "Inleiding tot het vormbegripbij Debussy," in Melanges: Homage a Charles van den Borren (Anvers: De Nederlandsche Boekhandel, 1945), 145ff.; ChristianGoubault, Claude Debussy (Paris: Champion, 1986), 13640; Peter Gilke, "Musik aus dem Bannkreis einer literarischenAesthetik: Debussys 'Prelude a l'Apres-midid'un Faune,' "Jahrbuch Peters1978, Aufsatze zur Musik 1, ed. E. Klemm (Leipzig: Peters, 1979), 103-46; James Music 8 Hepokoski, "FormulaicOpenings in Debussy," Nineteenth-Century

appropriatetest case for several reasons. On the one hand, Debussy specifically drew attention to the novel tonal and formal properties of the Prelude. In a letter to Henri Gauthier-Villars(10 October 1896) he observed:
The Prelude a "L'Apres-midid'un faune," cher Monsieur, is it perhaps the dream left over at the bottom of the faun's flute? . . . It

which knowhow' for a alsodemonstrates disdain the 'constructional Thenagain,it has no respect is a burden ourfinestintellects. upon to for tonality! Rather,it's in a modewhichis intended containall of the nuances-I can give you a perfectlylogicaldemonstration
this.21
20Forfurther details, see Matthew Brown, "A Rational Reconstruction of SchenkerianTheory," (Ph.D. diss., Cornell University, 1989). 21Lesure ed., Lettres,79; Lesure and Nichols, 84-85. It is hard to imagine what this demonstrationwould have looked like because Debussy was extremelyskepticalabout the value of musicalanalysis.For example, in an essay for La Revue blanche, 1 April 1901, he wrote: "Grown ups tend to forget that as childrenthey were forbiddento open the insides of their dolls-a crime of high treason against the cause of mystery. . . . And yet they still insist on poking their aestheticnoses into thingsthat don't concernthem! Withouttheir dolls to break open, they still try to explain things, dismantlethem and quite heartlessly kill all their mystery" (Lesure, ed., Monsieur Croche, 23-24; Lesure and Smith, Debussy on Music, 13). He especially abhorred "theme

d'un faune" Tonalityand Formin Debussy's Prelude a "L'Apres-midi

131

analysesgenerallyagree about the location of the work'smain formal divisions. As shown in Example 1, the piece is clearly articulated at mm. 30, 37, 55, 79, 94, and 106. The famous flute theme is presented in mm. 1-30 and 94-106 on Cf in E major (mm. 1- 30 end in V of E). Measures 30-37 contain whole-tone diminutions of the flute theme, and mm. 79-93 present sequential statements on E (mm. 79-85) and Eb (mm. 86-93). The coda presents a simplified version of the flute theme, again in E. A contrastingtheme in the new key of Db is developed in mm. 55-78, while mm. 37-54 serve both as a transition to the Db theme and as a development of the flute theme. Debates arise, however, when we try to explain the motivic significanceand tonal properties of each section.23Four problems stand out: (1) Although mm. 1-30 center on E, the opening phrases are extremely abstruse tonally; the tonic E
(1984): 56-57; Roy Howat, Debussy in Proportion (Cambridge:Cambridge University Press, 1983), 149-53; Albert Jakobik, Debussy oder der lautlose Revolution in der Musik (Wirzburg: Konrad Triltsch, 1977), 30-44; Parks, Music of Debussy, especially 36-38, 224-25, 240-43, 263-67; Leonard B. Meyer, Style and Music (Philadelphia: University of Pennsylvania Press, 1989), 223-25; David Neumeyer and Susan Tepping, A Guide to Schenkerian Analysis (Englewood Cliffs, NJ: Prentice-Hall, 1992), 39-40; Walter Piston and Mark DeVoto, Harmony, 4th ed., rev. (New York: Norton, 1976), 51820; Salzer, Structural Hearing, vol. 1, pp. 209-10 and vol. 2, Ex. 455; Arthur Wenk, Claude Debussy and the Poets (Berkeley: University of California Press, 1976), 148-70; and Wenk, Debussy and Twentieth-Century Music, 57-61. 23Berman,Dille, and Goubault claim that the Prelude is best treated as a continuous ternary form: A (mm. 1-30), transition (mm. 30-54), B (mm. 55-78), A' (mm. 79-106), Coda (mm. 106-10). Barraqueand Howat regard the work as an arch form: A (mm. 1-30), A development (mm. 30-54), B (mm. 55-78), A' development (mm. 79-93), A' (mm. 94-106), Coda (mm. 106-10). Crotty offers a two-partreading:A (mm. 1-30, 30-37, 37-54); A' (mm. 55-78, 79-94, 94-106); Coda (mm. 106-10). Parks treats the Prelude as a variation form: Theme (mm. 1-10), Variation 1 (mm. 11-20), Var. 2 (mm. 21-30), Var. 3 (mm. 31-54), Var. 4 (mm. 55-78), Var. 5 (mm. 79-93), Var. 6 (mm. 94-99), Var. 7 (mm. 100-105), Var. 8 (mm. 106-10).

does not appear until m. 13, and is not confirmedby a closed progression until mm. 21-26. (2) It is not obvious how mm. 30-54 relate to their surroundings; the whole-tone chords (mm. 30-37) are particularlyhard to fathom from a tonal perspective. (3) While the B section (mm. 55-78) establishes a new theme and key area, the precise layout of this passage is far from obvious. (4) It is hard to decide whether mm. 79-106 are better regarded as two segments (mm. 79-93 and 94-106) or as a single span. To resolve these dilemmas, let us examine the four sections in depth. Few passages in the standardrepertory are more obscure than the opening of the Prelude. At first sight, the sinuous flute theme and its shimmering accompaniment seem quite removed from the familiar world of common-practicetonality. Eventually E emerges as tonic in m. 13, but it is unclear how the seventh chords on At, Bb, and D (mm. 4-12) prepare the tonic. Opinions differ widely about the tonal function of these harmonies. Some, such as Felix Salzer and James Hepokoski, propose that the tonic chord is implied from m. 1. Salzer's graph of mm. 1-30 adds a virtual tonic chord in parentheses and treats the opening Ct of the melody as a neighbor tone to the following B (m. 1). Although Salzer marks very few roman numerals (the V-I progression at m. 13 is a notable exception), he adds two small arrows in mm. 4 and 5. These suggest that the A# and Bb sevenths arise contrapuntally and link the virtual tonic of m. 1 to the D seventh in m. 11. Hepokoski likewise marks a tonic at the opening, and suggests that the opening flute arabesque expands the apparent "contradiction"between A#07 (VII of V in E) and B,7 (V7 of D# or V7 of V in Gt). The latter apparently acts as a German sixth to the D7 in m. 11, presumably with an elision to its normal resolution (V7 of D). Hepokoski then proposes that the bass tones D (m. 11) and
DO (m. 13) are neighbors to the root E (m. 13).

Other writers, however, have been reluctantto make functional ascriptionsof any sort. John Crotty regardsmm. 1-13

132

Music Theory Spectrum

Example 1. Debussy, Prelude a "L'Apres-midi d'un faune": main formal divisions _--__
A (mm. 1-30)

_----_-_ ___

--___

_____________

_________________-__

Whole-toneepisode ( (min. 31 -36)

_____ ______'-- ________J--__._ _ __ _ __ _

_ __ __ __

Transition

(mm. 37- 54)

(mm. 55-78)

l
(D

ri .. -...:,
--------- -

r
_ ------------------------

VP

A' ----~ -_-- --

-- -of-

v~~

Coda (mm.,o16-,o)

( ^A^ ^ i ^.^

, rt
r

d'un faune" Tonalityand Formin Debussy's Prelude a "L'Apres-midi Example2. Tonal analysisof the Prelude,mm. 1-13
a 6

133

as part of a double-tonic complex E-C# that Debussy worked out during the rest of the Prelude. William Austin simply shows the bass motion A#-Bb-D-B-E, and notes that although roman numerals could be added to some progressions, "the effort to label every chord seems out of proportion with any resulting insight."24Arthur Wenk takes a similar tack, claiming that the passage reflects Debussy's shift away from functional harmony to "static" or "circular" successions.25RichardParks, meanwhile, dispenses with tonal concepts altogether and asserts that the Prelude "opens with a complement relation: the flute melody's pcs form set 7-1, while 5-1 is embedded prominentlywithin the sixteenths that
descend to pc
7."26

f ""'f

A ii H.

1r
? S

o
Nb (

Lt.

Obviously, the differences between these interpretations are partly matters of emphasis or context, and partly results of the anomalies inherent to Debussy's score. It is perhaps testimony to the work's ambiguitythat so many readingshave been advanced. Example 2, however, presents an alternative Schenkerian analysis of mm. 1-13. Here, the passage is interpreted as a transformation of the progression VII7 of V-V9-I in E.27 This progression is given in Example 2a. Example 2b shows how the dominant is arpeggiated in the bass through its mixed third, D, and how the Ct in the upper voice in m. 4 is suspended over the D and B to form seventh and ninth chords in mm. 11-13. The resulting progression VII7-V9 is very similar to those described by Schenker in
24Austin, "Toward an Analytical Appreciation," 84-85.
25Wenk, Debussy and Twentieth-Century Music, 56-61. 26Parks, Music of Debussy, 156-57.
d

00
A.; --Nb

'I
VII7 of V V9 I

A1

??

Nb

VII7 ofV
6---

V9

I
5

27It might seem that this sonority might be more conveniently labeled lIV7. However, as Dave Headlam and I showed in our paper "Schenkerian Theory and the Limits of Tonality: The Problem of fIV" (Annual Meeting of the Society for MusicTheory, Oakland, 1990), Schenkeriantheory assumes that #IV Stufencannot be generated directlyfrom I in tonal contexts; instead, of they can only be derivedindirectlyeither from transformations other Stufen (for example, III of III, IV of II, VII of V, etc.), or from interpolations.

00

113
D5

0)()

'
VII 7 of V

-I T
V 9$-8 I

134

MusicTheory Spectrum

Der freie Satz.28Example 2c then shows how the initial VII7 of V in m. 4 is joined to the VII chord in m. 11 by a passing the Bb7. The latter chord arisescontrapuntally; alto and tenor F complete a chain of parallel thirds extending from E parts Debussy's draft and two-piano reduction than in the finished score. Finally, Example 2d adds the overall motion--C B-AO -of the opening tune, along with the expansion of the dominant Dt-B in m. 13. Example 2 clarifies several important issues. First, the graphssuggest that the tonal function of the opening four bars is indeterminant;althoughthe pitch Ct is clearly emphasized, its role is obscure. In fact, the G (mm. 1-2) implies that the harmonies probably change in these bars; such a change always occurs when the theme is harmonized, even when it raised to GO(mm. 100-101). To complicate matters further, the goal of the opening phrase seems to be A# rather than B. In m. 3, B seems to pass between C# and At, and only in later statements do CKand AO serve as neighbors to B. Second, whereas mm. 1-3 are ambiguous, mm. 4-13 can be derived by orthodox tonal transformations. Indeed, the rather strikingprogression bVII-V is actually quite common in Debussy's early compositions, and even controls the middleground of entire works such as the song "C'est l'extase" (Ariettes oubliees, 1887).29Third, the chromatic inner lines found in Example 2c recur later in the Prelude. As we will see, both mm. 20-30 and 79-106 are bound together by anal28Schenkerdiscussed VII-V progressions in various places in Der freie Satz, especially par. 246, fig. 111 (fig. 1llc shows how VII7 can occur with V9). See Heinrich Schenker, Derfreie Satz (Vienna: Universal, 1935); translated by Ernst Oster under the title Free Composition(New York, Longman, 1979). See also Matthew Brown, "The Diatonic and the Chromaticin Schenker's Theory of Harmonic Relations," Journal of Music Theory 30 (1986): 1-33. 29For differentanalysisof this song, see WallaceBerry, MusicalStructure a and Performance(New Haven: Yale University Press, 1989), 144-216.

(m. 4) to E (m. 13). This chain is perhaps more obvious in

ogous chains of parallel thirds. Fourth, Debussy veils our sense of tonality by manipulatingthe way in which melodic phrases intersect with their harmonic foundation. The progression VII7 of V-V9-I (mm. 1-13) actuallycuts across two distinctstatements of the flute theme in mm. 1-10 and 11-20. Having discussed the opening of the Prelude, let us now turn to mm. 30-54. Jean Barraque has rightly observed that this segment functions as some sort of development. He suggests that it can be divided into three discrete phases: mm. 30-37 offer a "double presentation of the principal theme harmonized by chords from the whole-tone scale"; mm.
37-50 introduce "the second theme . . . in the oboe" and

modulate towards the key of the middle section; and mm. 50-54 sustain "a pedal on the dominant of Dk."30When we look more closely at the wider tonal and motivic context of the passage, several other details seem significant. For example, although the oboe theme does not appear before m. 37, this gesture is derived from the pentatonic motive in m. 28 (see Exx. 3a and 3b).31 Indeed, the links between this section and its predecessor are reinforcedby the fact that the oboe theme appears with fragments of the main flute theme in mm. 46-47 (see Ex. 5b). More remarkably, as Charles Burkharthas demonstrated, mm. 37-55 form a single span, with a stepwise descent in the bass, from B (V of E) via Bb and A to Ab (V of Db,), and with a series of nested repetitions of the chromaticfigure B(C,)-C-Ct1(Db) in the upper parts (see Ex. 3c).32 Significantly,this chromatic figure is embedded within the main flute theme (see Ex. 3d). Burkhart's
30Barraque,Debussy, 88; translated in Austin, ed., Prelude, 163-64. 31Forelaborate accounts of the complex motivic connections see Austin, "Towardan Analytical Appreciation,"especiallypp. 76-78. Peter Gilke has stressed the significanceof pentatonicfiguresboth to the opening flute theme (Gilke, Exx. 1, 3, and 5), the oboe theme at m. 37 (Gtilke, Ex. 4), and the Db theme (Giilke, Ex. 2); see "Musikaus dem Bannkreis," 114ff. and 120ff. Ex. 9 very slightly;see Burkhart,"Schenker's 32I have adaptedBurkhart's 'Motivic Parallelisms,'" 156.

Tonality and Form in Debussy's Prelude a "L'Apres-midi d'un faune"

135

Example 3. Prelude a "L'Apres-midi d'un faune," mm. 37-55

a

1 3

1 3

1 A 3

'~~~

A 5

R ... "t ,.

Af i -

k I

--

S'_

I1h., , ,_

1, I

b, b'4 *.Ul

,, I ~! d ,

Ii
....... I

I]I

,,

000000
11--N ,:%. a__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

9-8 V

9-8

6-5 4-3

9-8

'1 -

d
V

O 0
I L_ N 1 IBBI

B- B#-Co

136

MusicTheorySpectrum

readingnot only connects mm. 37-54 to mm. 1-30, but it also shows how the transition functions as one, rather than two, separate spans. While mm. 37-54 are certainly unusual, they are a good deal more orthodox than mm. 30-37; Burkhartis surely correct to state that the latter contains "the work's most radical His analysis shows departure from traditional procedure."33 how the passage is built from the two whole-tone setsIt would be and Ct-Dt-F-G-A-B. C-D-E-Ft-Gt-Af however, to conclude that this passage cannot be dewrong, rived from tonal transformations.On the contrary, Example 4 shows that mm. 30-37 arise contrapuntally,from complex passing motions in the inner voices. The upper line is created by a motion from the inner voice: the Ft in the alto part (m. 30) ascends by step through G, A, B , and C to Cf (m. 37). The inner parts mostly ascend by step: the viola part moves up from B through Cl (m. 31-33) to D and E (mm. 34-36); the second violin part moves up from F through G (mm. 31-33) Gt and A# (mm. 34-36); the cello, bass, and bassoon parts move up from Ct through D and Dft (mm. 31-33) to E, F, and F# (mm. 34-36). The bass part simply shifts from G (mm. 32-33) via Bb (mm. 34-36) back to B (m. 37). The final sonority at the end of m. 36 (F#-A#-Et-C~) functions as an altered secondary dominant (V 7 of V), that resolves to V9 of E. The notion that whole-tone harmonies stem from altered dominantsis, of course, one that has been advancedby many writers, from Schoenbergto Tovey.34It is also consistent with

Example 4. Prelude a "L'Apres-midi d'un faune," mm. 30-39

5
AiM

6 I

?@ (3t

? ?4 9 b
'0V7

?
9-8
(I)

vL

(Vb5 of V)

the basic tenets of Schenkeriantheory. As becomes clear near the beginningof volume 1 of Kontrapunkt,Schenkerbelieved that when composers use exotic scales in tonal contexts, they do so "not to loosen [the] system in order to incorporate a foreign one, but, on the contrary, to use [the] major-minor system to express the foreign element."35Besides providing a tonal derivationfor mm. 30-37, Example 4 also implies that this passage does not advance the tonal flow of the Prelude; rather, it acts as an aside or parenthesisthat delays the modulation to Db. Schenker described such "delays" very beautifully in Der freie Satz:
In the art of music, as in life, motion toward the goal encounters obstacles, reverses, disappointments, and involves great distances, detours, expansions, interpolations,and, in short, retardationsof all kinds. Therein lies the source of all artisticdelaying, from which the creative mind can derive content that is ever new.36

33Ibid., 157. 34See, for example, Arnold Schoenberg, Harmonielehre,chap. 20 (Vienna: Universal, 1911; 3d ed., 1922); translatedby Roy E. Carterunder the title Theory of Harmony (Berkeley; University of California Press, 1978), 390-98. See also Donald FrancisTovey, "Harmony,"reprintedin The Forms of Music (Cleveland: World Publishing Co., 1965), 69.

35HeinrichSchenker, Kontrapunkt,vol. 1 (Stuttgart: Cotta, 1910), 44; translatedby John Rothgeb and Jurgen Thym under the title Counterpoint (New York: Schirmer, 1987), 28. 36Schenker,Der freie Satz, chap. 1, section 3, p. 18; Oster ed. and trans., 5.

d'un faune" Tonalityand Formin Debussy's Prelude a "L'Apres-midi

137

In the case of the Prelude, this brief delay helps to evoke the mysterious, dreamlike atmosphere that permeates Mallarme's eclogue. So far, we have seen that the Prelude is built from two main motives-the opening flute arabesque plus its derivatives, and the Db theme. Austin, however, identifies two other gestures, to which he refers as the syncopated and flowing motives.37Since these ideas play a vital role in the subsequent unfolding of the piece and help to clarify the formal function of the B section, it is worth examining them closely. Example 5 traces the life history of both ideas. Although the origins of the syncopated motive can be traced back to the horn parts in mm. 5 and 13, this gesture first becomes prominent in m. 39 (Ex. 5a). Almost immediately, it is combined with the opening flute motive (Ex. 5b). The flowing motive, meanwhile, is first heard in m. 28 (Ex. 5c) and becomes the progenitor of the oboe theme at m. 37. In the Db section, the flowing and syncopated motives are both absorbed into the B theme: the former appearsas a continuation of the B theme in mm. 61-62 (Ex. 5d); the latter appearswith the flowing motive beginning in m. 67 (Ex. 5e). The short codetta (mm. 74-78) then combines the syncopated motive in counterpointwith the B theme and flowingmotive (Ex. 5f). This example of presenting themes simultaneouslyor in close succession-what we might term "motivic compression"sets up the climax of the Prdlude in mm. 94-99, where the syncopated and flowing motives both return in counterpoint with the main flute theme (Ex. 5g). Example 5 highlights three essential features of thematic working in Debussy's music. First, it often includes very intricate motivic relationships. This fact apparentlycontradicts Debussy's frequent claims to simplifying symphonic composition. For example, in an essay for SIM (1 November 1913) he declared:
37Austin,"Toward an Analytical Appreciation," 75-78.

Let us purify music!Let us tryto relieveit of its congestion, find to a less cluttered kindof music.And let us be careful thatwe do not stifle all feeling underneath mass of superimposed a designsand motives:how can we hope to preserve finesse,our spirit,if we our insiston beingpreoccupied so manydetailsof composition? with We areattempting impossible the whenwe tryto organize braying a pack of tinythemes,all pushing jostlingeach otherfor a bite out of and old sentiment.38 poor Second, in Debussy's music, motives that initially seem innocuous may end up playing a vital role later. Since this process cuts across the main formal divisions of the Prelude, it helps to erode the boundaries and autonomy of each section. Third, Debussy's music is often extremely complex polyphonically, especially at critical moments in the form. In the Prelude, the densest textures occur at the climax (beginning in m. 94). This point is doubly ironic because Debussy is often portrayed as the archenemy of polyphony, and because Schenker traced the downfall of music in the twentieth century to a "decline of counterpoint."39 Before leaving the B section, we must briefly consider its harmonic structure. We have already seen how the thematic
38Lesure,ed., MonsieurCroche, 241; Lesure and Smith, Debussy on Music, 297. 39Forpublisheddiscussionsof Debussy's contrapuntalpractices,see Peter DeLone, "ClaudeDebussy: Contrapuntiste MalgreLui," CollegeMusic Symposium 17 (1977): 48-63; and WilliamAustin, Musicin the Twentieth Century, (New York: Norton, 1966), especially p. 20. Debussy himself discussed the significanceof counterpointin his music; see Lesure and Smith, Debussy on Music, 84, 94, and 278. Schenker, of course, made frequent attacks on music, such as the following in Derfreie Satz: "The present twentieth-century decline of counterpointhas brought about the decline of diatony. This is the fault of the musicians, who still have not grasped the fact that as long as the fifth determines the natural sonority-and that will always be so-a voiceleading technique based on the fifth, as nature requires, cannot lead to any diatony other than the diatony which our art has exhibited up to the present day. All attempts to deprive nature of her rights will shatter against the wall of her resistance" (Oster, ed. and trans., par. 4, p. 11).

138

MusicTheory Spectrum

Example 5. Motivic compression in the Prelude

^~(g)D-~ __a* -I e'r

^--

Syncopatedmotive
riI

Flute theme

II

I,

-j n0"g n -J gem -'ig

-?
3

?3

| I I ?[ f\
1

Syncopated motive B theme

Flowing motive Flowing motive

0
d

>

blbF
?IQ

F rFr vi fr,
B theme __ )
-

I
I
(6 7)I A

lt
Syncopated motive

"

Flowing motive I, I

t,

r Ii'
IL

l
B theme

'

II

r
F.
I

A -\ ro 'i
I

, rm _:^Trr
r II

rr
#
Flowing motive

I rr r,
Flowing motive

Syncopated motive Flute theme

II Us

Flowing motive
3mr^

'^

-S=Fmff_________?3

"c )11 'c*1_-

,P.

,n I

7---- -I

r .r r7--

-I' -:_

Syncopated motive

d'un faune" Tonalityand Formin Debussy's Prelude a "L'Apres-midi

139

Example 6. Prelude a "L'Apres-midi d'un faune," mm. 55-78

6 6
J A . j

1
b-bbb

1
I.

I
'/

hbJ,l.J
i ~

1
'b;

:b iJ

JJ

IV I1

t) bbbb P 6*^k
I

b-

b 6 , 4~
IVb3 bVI
of VI

- < "bL
aug. 6 I

IV VII 1Iof

of Vl

content changes between the first and the second statements of the Db theme. Example 6 shows that this change coincides with a remarkableshift in the theme's harmonic support. To understand this transformation it is best to begin with the

second statement (see Ex. 6a). These measures present a diatonic descent throughthe D octave in the upper part that is supported by the simple progression I-IV-II-V-I. The firststatement, however, is more complex (see Ex. 6b). Here,

140

MusicTheory Spectrum

the octave descent is transformed chromatically: Db-Cb,Bb -Ab-Gb-Fb -Elb-C-Db. Notice how the augmentedsixth chord resolves traditionally:D moves down by step in the bass, while the tritone Ft-C in the upper voice resolves inwards to F and Db (horns 2 and 4, violin 2). As shown in Example 6c, the supportingprogression, with its mixed subdominant, flat submediant, and augmented-sixthsonorities, foreshadows the chromatic excursions found in Debussy's later works, such as the piece "Reflets dans l'eau" (Images, Book I, 1904-5). Here the upper line moves from I via
,I

less do they initiate the final section of a piece. How, then, can we explain their presence in m. 94? Example 7a suggeststhat these dissonancesarise from contrapuntal lines that start in m. 79. The inner parts F, in m. 94 are part of a long chain of chromatic parallel thirds that
passes from G#(m. 79) via EC(m. 83), G (m. 86), and D (m. F 90) to (m. 94). This line then descends to E (m. 96) before

and t 3i to v and }.

We now come to the final sections of the Prelude. As mentioned earlier, experts disagree about the formalfunction of mm. 79-106. Some, such as Barraque and Howat, argue that these measures are built from two distinct spans: mm. 79-93 develop the flute theme sequentiallyon E and Eb; mm. 94-106 restore the main theme to Ct for the start of the final section proper.40 Conversely, Denis Dille, LaurenceBerman, and others suggest that these bars constitute a single span analogous to that of the opening A section.41 Whichreadingshould we prefer?Certainly,there are good reasons for dividing mm. 79-106 into two spans. Although the flute theme is transposed to E in m. 79, it is not restored to the original startingpitch COuntil m. 94. Furthermore,the scherzandocharacterof the flute theme (mm. 79-93) recalls the whole-tone variations (mm. 30-37). Tonally, however, this reading is less convincing. No one would deny that the tonic 6 returns in m. 79, but when the bass tone E returns at m. 94, it no longer supports a simple tonic triad. Instead, we find an E13identical to the one in m. 26. Dissonances of this sort do not normally appear at the start of a span; still
40Barraqu6, Debussy, 88; translated in Austin, ed., Prelude, 164; Howat, Debussy in Proportion, 149. 41Dille's view is summarized by Austin in "Toward an Analytical Appreciation," 74; Berman, "Debussy's Summer Rites," 231.

shifting direction to join the upper line in m. 104. The soprano C# in m. 94, meanwhile, arises as an expansion of the chromaticmotive B-B -C that is projected from m. 79. The CO,which is even locally tonicized beginning in m. 96, eventually slides down through Ct and Bb to meet the inner thirds at m. 104. Besides demonstrating the overall integrity of mm. 79106, Example 7b also indicates that the final section is built from the same tonal transformationsas mm. 20-30. In mm. 20-30, the upper line also expands the motive B-Bt-C# while the inner parts contain a chain of chromatic thirds extending from G (m. 20) via B (m. 22), A (m. 23), Gd (m. 25), G (m. 25) to F (m. 26). As with mm. 79-106, these

thirds finally shift up to meet the upper line for the cadence. As Leonard B. Meyer has noted, the cadences in mm. 29-30 and 103-6 are connected motivically; both contain the same motives, marked x and y in Example 7c.42The significance of this reading is twofold. It suggests that the final A' section is not a simple repetition of the opening section, but is an ingenious composing-out of a small segment, mm. 20-30. Furthermore, since this passage conflates the main flute theme with the syncopated and flowing motives and inflects COas a local tonic, it provides a subtle reminiscence of the B section. This allusion to the B section unifies the Prelude and demonstrates the complex, long-range connections that invariablybind Debussy's music together. Tonally, then, mm.

42Meyer, Style and Music, 223-25.

d'un faune" Tonalityand Formin Debussy's Prelude a "L'Apres-midi

141

Example 7. Tonal models in the Prelude

4 A. 3 2 1

I I

II Nb

I II

79-106 constitute a single unit, bound together by an elaborate contrapuntal framework; they do not subdivide conveniently into two separate tonal spans. To sum up: When analyzed from a Schenkerianperspective, Debussy's Prelude a "L'Apres-midi d'un faune" is best

.-.--\ L

6I~ --'

V6 !2 VV-6

----IE

,V,1

_ __3
4, - ~ ;ttIa- d-

3I 3

regardedas a continuous ternaryform-A (mm. 1-30), transition (mm. 37-54), B (mm. 55-78), A' (mm. 79-106), and coda (mm. 106-10). However, this scheme is obscuredin four ways: by starting with an incomplete progression VII7 of V-V9-I; by adding a parentheticalwhole-tone episode (mm. 30-36); by developing the syncopated and triplet figures across the formal boundaries; and by modeling the final A' section (mm. 79-106) on part of the opening section (mm. 20-30). Although these features are unusual in themselves, especially when judged against the standardsof nineteenthcentury tonal theory, they do not represent a complete repudiation of tonal paradigms. What is so remarkable about the Prelude is that Debussy managed to combine these techniques so boldly over such a large span. Perhaps we should view the rather elaborate explanations presented above as signs of Debussy's cultural elitism and his love of the esoteric-concerns that he described in a famous letter to Ernest Chausson (3 September 1893):
Music really ought to have been a hermetical science, enshrined in texts so hard and laborious to decipher as to discouragethe herd of people who treat it as casually as they do a handkerchief! I'd go further and, instead of spreadingmusic among the populace, I propose the foundation of a "Society of Musical Esotericism."43

,c

xt

YI,... 1

1? I<h H.2

I2

'

33

43"Vraiment musiqueauraitdfi tre une science herm6tique,gard6epar la des textes d'une interpretationtellement longue et difficile qu'elle auraitcertainement d6courag6la tropeau de gens qui s'en servent avec la d6sinvolture que l'on met a se servir d'un mouchoir de poche! Or, et en outre, au lieu de chercher a repandre Fart dans le public, je propose la fondation d'une "Societe d'Esotericisme Musical" et vous verrez que M. Helman n'en sera

142

MusicTheory Spectrum

As it happens, this letter dates from the time when Debussy was finishing work on the Prelude. Although it is beyond the scope of this essay to trace Debussy's use of the four techniques enumerated above in all his subsequentorchestralworks, a few examples are in order. To begin with, of Debussy's remaining eleven orchestral movements, all but one-"Par les rues et les chemins" (Iberia, first movement)-begin with some sort of incomplete progression. For the most part, these works open with introductions that move to the tonic for the start of the main movement proper. However, "Les parfums de la nuit" and "Le matin d'un jour de fete" (Iberia, second and third movements) are more complex. The former is a classic example of an auxiliary-cadence piece: the bulk of the movement prolongs the dominant of Ft with the tonic finally occurring near the end (m. 92).44The latter, meanwhile, not only begins away from the tonic, but it actually ends somewhere else: the movement starts in Eb, moves to C, and closes in G with material from the middle section of "Par les rues et les chemins." We can also find parentheticalepisodes and interpolations in Debussy's later orchestral scores. Perhaps the most obvious examples occur in the transition between the final movements of Iberia. Here Debussy added a subordinate theme from "Le matin d'un jour de fete" near the end of "Les parfums de la nuit" (mm. 127-28), and he inserted a flashback to one of the main themes in the second movement near the beginning of the third (mm. 5-6). Significantly,the latter has "interruption" connections with Debussy's piano Prelude "La s6ernade interrompue":in mm. 80-84 and 87-89 of the
pas, ni M. de Monnieres non plus!" See Lesure, ed., Lettres,51; Lesure and Nichols, eds., Letters, 52. This letter is normally connected with Debussy's interest in the occult. 44Formore details of this and the following analytical observations, see my forthcomingbook Debussy: 'Iberia'(Oxford: Oxford University Press).

Prelude Debussy actually inserted the main theme from "Le matin d'un jour de fete"! On a largerscale, the entire fanfare section from the middle of "Fetes" (Nocturnes,second movement) is a giant interpolation: mm. 116-73 are actually inserted within an ascending line that spans from m. 27 to m. 208. As in Iberia, the insert recalls material from another work by Debussy: at the climax of the fanfare section, we hear Rodrigue's leitmotiv from Debussy's aborted "Wagnerian" opera Rodrigue et Chimene (1890-92).45

Several interesting instances of motivic compression also occur in La Merand the orchestralImages. A particularly fine example appears at the climax of "Jeux de vagues" (La Mer, second movement).46In mm. 163-218, the main theme returns and is combined contrapuntallywith other subordinate themes as well as with its own diminutions and augmentations. The same phenomenon can be seen in "Rondes de printemps"(Images). In two long sections (mm. 87-118 and 138-205), Debussy develops the folk tune "Nous n'ironsplus au bois" contrapuntallywith itself, and with other themes, including another folk tune "Do, do l'enfant do." Lastly, Debussy relied on tonal modeling in several later symphonicscores. In "Nuages" (Nocturnes, firstmovement), he modeled the final section (mm. 80-93) on the last segment of the opening A section (mm. 43-56). In recomposing the earlier material, he inserted two repetitions of a horn motive from mm. 23-24 and 27-28. More remarkably, the second half of "De l'aube a midi sur la mer" (La Mer, first movement), is an elaborate composing out of the contrapuntal frameworkof mm. 31-84: both move from Db via BI to Ab and back, via Gb to Ab. Similar experiments in tonal mod45Comparethe chromatic motive and Tristan chords (mm. 170-73) of "Fetes" with Rodrigue's theme, transcribedby Orledge (Exx. 12 and 13) in Debussy and the Theater(Cambridge:CambridgeUniversityPress, 1982), 31. 46See Wolfgang Domling, Claude Debussy: La Mer (Munich: Wilhelm Finck Verlag, 1976).

d'un faune" Tonalityand Form in Debussy's Prelude a "L'Apres-midi

143

eling can also be found in "Par les rues et les chemins"; here, the final A section and the coda are built from roughly the same set of tonal transformations as the middleground of the entire movement. To what extent, then, did Debussy manage to move away from nineteenth-century tonal and formal conventions? From the preceding discussion, it seems that Debussy's path away from traditional composition was gradual. Indeed, most of the techniques used in the Prelude have clear precedents in nineteenth-century tonal practice, at least as explained by Schenkerian theory. This suggests that Debussy was not the iconoclast that Boulez and others believe. His genius was perhaps not so much in creating entirely new principles of harmony and form, but rather in exploiting established techniques in bold and unusual ways. In many ways, this essay confirms the conclusions of Arnold Whittall: Debussy, at his best, was always a dramatic composer. What is dramatized, what brings tension and dynamism to the music, is the skilfully balanced relationship between chromaticism and diatonibut cism, both of which may show modal characteristics which never lose sight of the triadicconstructionsand progressionsof earliertonal music. Debussy's harmony functions precisely in the sense that it

gives meaning, and movement, to this relationship. As a language it can best be described as "expandedtonality", a language in which tonality still acts as a basic term, giving perspective to all other harmonic activity.47 Even if Debussy was unable to ignore common-practice tonality altogether, he did create a remarkable body of symphonic music that was indeed fit for "the century of aeroplanes." ABSTRACT Debussy is often credited with revolutionizingmusical form by rejecting nineteenth-centurytonal and formalpractice. This essay considers this claim by examining the tonal and thematic frameworkof the Prelude a "L'Apres-midid'un faune." Although this analysis shows that the Prelude can be explained by traditional tonal procedures, it identifies four techniques-incomplete progressions,parenthetical episodes, motivic compression, and tonal models-that allowed Debussy to move away from conventional symphonicmodels. The paper ends by showing how he developed these techniques in later orchestralworks such as the Nocturnes, La Mer, and Iberia.
47ArnoldWhittall, "Tonality and the Whole-Tone Scale in the Music of Debussy," The Music Review 36 (1975): 271.